Showing 151–200 of 1,075 results for author: Han, W

Evolution of cooperation with the diversity of cooperation tendencies

Authors: Linya Huang, Wenchen Han

Abstract: The complete cooperation and the complete defection are two typical strategies considered in evolutionary games in many previous works. However, in real life, strategies of individuals are full of variety rather than only two complete ones. In this work, the diversity of strategies is introduced into the weak prisoners' dilemma game, which is measured by the diversity of the cooperation tendency.… ▽ More The complete cooperation and the complete defection are two typical strategies considered in evolutionary games in many previous works. However, in real life, strategies of individuals are full of variety rather than only two complete ones. In this work, the diversity of strategies is introduced into the weak prisoners' dilemma game, which is measured by the diversity of the cooperation tendency. A higher diversity means more cooperation tendencies are provided. The complete cooperation strategy is the full cooperation tendency and the complete defection strategy is without any cooperation tendency. Agents with other cooperation tendencies behave as partial cooperators and as partial defectors simultaneously. The numerical simulation shows that increasing the diversity of the cooperation tendency promotes the cooperation level, not only the number of cooperators but also the average tendency over the whole population, until the diversity reaches its saturated value. Furthermore, our work points out maintaining cooperation is based on the cooperation efficiency approximating to the reward of cooperators and that the cooperation efficiency oscillates and quickly decreases to zero when cooperator clusters cannot resist the invasion of defectors. When the effect of the noise for the Femi update mechanism is considered, a higher diversity of strategies not only improves the cooperation level of the whole population but also supports the survival of more rational agents. △ Less

Submitted 18 June, 2024; originally announced June 2024.

Comments: 7 pages, 5 figures, 1 table

Retrieval Meets Reasoning: Dynamic In-Context Editing for Long-Text Understanding

Authors: Weizhi Fei, Xueyan Niu, Guoqing Xie, Yanhua Zhang, Bo Bai, Lei Deng, Wei Han

Abstract: Current Large Language Models (LLMs) face inherent limitations due to their pre-defined context lengths, which impede their capacity for multi-hop reasoning within extensive textual contexts. While existing techniques like Retrieval-Augmented Generation (RAG) have attempted to bridge this gap by sourcing external information, they fall short when direct answers are not readily available. We introd… ▽ More Current Large Language Models (LLMs) face inherent limitations due to their pre-defined context lengths, which impede their capacity for multi-hop reasoning within extensive textual contexts. While existing techniques like Retrieval-Augmented Generation (RAG) have attempted to bridge this gap by sourcing external information, they fall short when direct answers are not readily available. We introduce a novel approach that re-imagines information retrieval through dynamic in-context editing, inspired by recent breakthroughs in knowledge editing. By treating lengthy contexts as malleable external knowledge, our method interactively gathers and integrates relevant information, thereby enabling LLMs to perform sophisticated reasoning steps. Experimental results demonstrate that our method effectively empowers context-limited LLMs, such as Llama2, to engage in multi-hop reasoning with improved performance, which outperforms state-of-the-art context window extrapolation methods and even compares favorably to more advanced commercial long-context models. Our interactive method not only enhances reasoning capabilities but also mitigates the associated training and computational costs, making it a pragmatic solution for enhancing LLMs' reasoning within expansive contexts. △ Less

Submitted 18 June, 2024; originally announced June 2024.

Prefixing Attention Sinks can Mitigate Activation Outliers for Large Language Model Quantization

Authors: Seungwoo Son, Wonpyo Park, Woohyun Han, Kyuyeun Kim, Jaeho Lee

Abstract: Despite recent advances in LLM quantization, activation quantization remains to be challenging due to the activation outliers. Conventional remedies, e.g., mixing precisions for different channels, introduce extra overhead and reduce the speedup. In this work, we develop a simple yet effective strategy to facilitate per-tensor activation quantization by preventing the generation of problematic tok… ▽ More Despite recent advances in LLM quantization, activation quantization remains to be challenging due to the activation outliers. Conventional remedies, e.g., mixing precisions for different channels, introduce extra overhead and reduce the speedup. In this work, we develop a simple yet effective strategy to facilitate per-tensor activation quantization by preventing the generation of problematic tokens. Precisely, we propose a method to find a set of key-value cache, coined CushionCache, which mitigates outliers in subsequent tokens when inserted as a prefix. CushionCache works in two steps: First, we greedily search for a prompt token sequence that minimizes the maximum activation values in subsequent tokens. Then, we further tune the token cache to regularize the activations of subsequent tokens to be more quantization-friendly. The proposed method successfully addresses activation outliers of LLMs, providing a substantial performance boost for per-tensor activation quantization methods. We thoroughly evaluate our method over a wide range of models and benchmarks and find that it significantly surpasses the established baseline of per-tensor W8A8 quantization and can be seamlessly integrated with the recent activation quantization method. △ Less

Submitted 4 October, 2024; v1 submitted 17 June, 2024; originally announced June 2024.

Comments: EMNLP 2024 Main (Long)

CustAny: Customizing Anything from A Single Example

Authors: Lingjie Kong, Kai Wu, Xiaobin Hu, Wenhui Han, Jinlong Peng, Chengming Xu, Donghao Luo, Mengtian Li, Jiangning Zhang, Chengjie Wang, Yanwei Fu

Abstract: Recent advances in diffusion-based text-to-image models have simplified creating high-fidelity images, but preserving the identity (ID) of specific elements, like a personal dog, is still challenging. Object customization, using reference images and textual descriptions, is key to addressing this issue. Current object customization methods are either object-specific, requiring extensive fine-tunin… ▽ More Recent advances in diffusion-based text-to-image models have simplified creating high-fidelity images, but preserving the identity (ID) of specific elements, like a personal dog, is still challenging. Object customization, using reference images and textual descriptions, is key to addressing this issue. Current object customization methods are either object-specific, requiring extensive fine-tuning, or object-agnostic, offering zero-shot customization but limited to specialized domains. The primary issue of promoting zero-shot object customization from specific domains to the general domain is to establish a large-scale general ID dataset for model pre-training, which is time-consuming and labor-intensive. In this paper, we propose a novel pipeline to construct a large dataset of general objects and build the Multi-Category ID-Consistent (MC-IDC) dataset, featuring 315k text-image samples across 10k categories. With the help of MC-IDC, we introduce Customizing Anything (CustAny), a zero-shot framework that maintains ID fidelity and supports flexible text editing for general objects. CustAny features three key components: a general ID extraction module, a dual-level ID injection module, and an ID-aware decoupling module, allowing it to customize any object from a single reference image and text prompt. Experiments demonstrate that CustAny outperforms existing methods in both general object customization and specialized domains like human customization and virtual try-on. Our contributions include a large-scale dataset, the CustAny framework and novel ID processing to advance this field. Code and dataset will be released soon in https://github.com/LingjieKong-fdu/CustAny. △ Less

Submitted 22 November, 2024; v1 submitted 17 June, 2024; originally announced June 2024.

Ents: An Efficient Three-party Training Framework for Decision Trees by Communication Optimization

Authors: Guopeng Lin, Weili Han, Wenqiang Ruan, Ruisheng Zhou, Lushan Song, Bingshuai Li, Yunfeng Shao

Abstract: Multi-party training frameworks for decision trees based on secure multi-party computation enable multiple parties to train high-performance models on distributed private data with privacy preservation. The training process essentially involves frequent dataset splitting according to the splitting criterion (e.g. Gini impurity). However, existing multi-party training frameworks for decision trees… ▽ More Multi-party training frameworks for decision trees based on secure multi-party computation enable multiple parties to train high-performance models on distributed private data with privacy preservation. The training process essentially involves frequent dataset splitting according to the splitting criterion (e.g. Gini impurity). However, existing multi-party training frameworks for decision trees demonstrate communication inefficiency due to the following issues: (1) They suffer from huge communication overhead in securely splitting a dataset with continuous attributes. (2) They suffer from huge communication overhead due to performing almost all the computations on a large ring to accommodate the secure computations for the splitting criterion. In this paper, we are motivated to present an efficient three-party training framework, namely Ents, for decision trees by communication optimization. For the first issue, we present a series of training protocols based on the secure radix sort protocols to efficiently and securely split a dataset with continuous attributes. For the second issue, we propose an efficient share conversion protocol to convert shares between a small ring and a large ring to reduce the communication overhead incurred by performing almost all the computations on a large ring. Experimental results from eight widely used datasets show that Ents outperforms state-of-the-art frameworks by $5.5\times \sim 9.3\times$ in communication sizes and $3.9\times \sim 5.3\times$ in communication rounds. In terms of training time, Ents yields an improvement of $3.5\times \sim 6.7\times$. To demonstrate its practicality, Ents requires less than three hours to securely train a decision tree on a widely used real-world dataset (Skin Segmentation) with more than 245,000 samples in the WAN setting. △ Less

Submitted 3 July, 2024; v1 submitted 12 June, 2024; originally announced June 2024.

Comments: This paper is the full version of a paper to appear in ACM CCS 2024

Relational Proxy Loss for Audio-Text based Keyword Spotting

Authors: Youngmoon Jung, Seungjin Lee, Joon-Young Yang, Jaeyoung Roh, Chang Woo Han, Hoon-Young Cho

Abstract: In recent years, there has been an increasing focus on user convenience, leading to increased interest in text-based keyword enrollment systems for keyword spotting (KWS). Since the system utilizes text input during the enrollment phase and audio input during actual usage, we call this task audio-text based KWS. To enable this task, both acoustic and text encoders are typically trained using deep… ▽ More In recent years, there has been an increasing focus on user convenience, leading to increased interest in text-based keyword enrollment systems for keyword spotting (KWS). Since the system utilizes text input during the enrollment phase and audio input during actual usage, we call this task audio-text based KWS. To enable this task, both acoustic and text encoders are typically trained using deep metric learning loss functions, such as triplet- and proxy-based losses. This study aims to improve existing methods by leveraging the structural relations within acoustic embeddings and within text embeddings. Unlike previous studies that only compare acoustic and text embeddings on a point-to-point basis, our approach focuses on the relational structures within the embedding space by introducing the concept of Relational Proxy Loss (RPL). By incorporating RPL, we demonstrated improved performance on the Wall Street Journal (WSJ) corpus. △ Less

Submitted 7 June, 2024; originally announced June 2024.

Comments: 5 pages, 2 figures, Accepted by Interspeech 2024

Bootstrapping Referring Multi-Object Tracking

Authors: Yani Zhang, Dongming Wu, Wencheng Han, Xingping Dong

Abstract: Referring multi-object tracking (RMOT) aims at detecting and tracking multiple objects following human instruction represented by a natural language expression. Existing RMOT benchmarks are usually formulated through manual annotations, integrated with static regulations. This approach results in a dearth of notable diversity and a constrained scope of implementation. In this work, our key idea is… ▽ More Referring multi-object tracking (RMOT) aims at detecting and tracking multiple objects following human instruction represented by a natural language expression. Existing RMOT benchmarks are usually formulated through manual annotations, integrated with static regulations. This approach results in a dearth of notable diversity and a constrained scope of implementation. In this work, our key idea is to bootstrap the task of referring multi-object tracking by introducing discriminative language words as much as possible. In specific, we first develop Refer-KITTI into a large-scale dataset, named Refer-KITTI-V2. It starts with 2,719 manual annotations, addressing the issue of class imbalance and introducing more keywords to make it closer to real-world scenarios compared to Refer-KITTI. They are further expanded to a total of 9,758 annotations by prompting large language models, which create 617 different words, surpassing previous RMOT benchmarks. In addition, the end-to-end framework in RMOT is also bootstrapped by a simple yet elegant temporal advancement strategy, which achieves better performance than previous approaches. The source code and dataset is available at https://github.com/zyn213/TempRMOT. △ Less

Submitted 7 June, 2024; originally announced June 2024.

Touch100k: A Large-Scale Touch-Language-Vision Dataset for Touch-Centric Multimodal Representation

Authors: Ning Cheng, Changhao Guan, Jing Gao, Weihao Wang, You Li, Fandong Meng, Jie Zhou, Bin Fang, Jinan Xu, Wenjuan Han

Abstract: Touch holds a pivotal position in enhancing the perceptual and interactive capabilities of both humans and robots. Despite its significance, current tactile research mainly focuses on visual and tactile modalities, overlooking the language domain. Inspired by this, we construct Touch100k, a paired touch-language-vision dataset at the scale of 100k, featuring tactile sensation descriptions in multi… ▽ More Touch holds a pivotal position in enhancing the perceptual and interactive capabilities of both humans and robots. Despite its significance, current tactile research mainly focuses on visual and tactile modalities, overlooking the language domain. Inspired by this, we construct Touch100k, a paired touch-language-vision dataset at the scale of 100k, featuring tactile sensation descriptions in multiple granularities (i.e., sentence-level natural expressions with rich semantics, including contextual and dynamic relationships, and phrase-level descriptions capturing the key features of tactile sensations). Based on the dataset, we propose a pre-training method, Touch-Language-Vision Representation Learning through Curriculum Linking (TLV-Link, for short), inspired by the concept of curriculum learning. TLV-Link aims to learn a tactile representation for the GelSight sensor and capture the relationship between tactile, language, and visual modalities. We evaluate our representation's performance across two task categories (namely, material property identification and robot grasping prediction), focusing on tactile representation and zero-shot touch understanding. The experimental evaluation showcases the effectiveness of our representation. By enabling TLV-Link to achieve substantial improvements and establish a new state-of-the-art in touch-centric multimodal representation learning, Touch100k demonstrates its value as a valuable resource for research. Project page: https://cocacola-lab.github.io/Touch100k/. △ Less

Submitted 6 June, 2024; originally announced June 2024.

Evaluating Durability: Benchmark Insights into Multimodal Watermarking

Authors: Jielin Qiu, William Han, Xuandong Zhao, Shangbang Long, Christos Faloutsos, Lei Li

Abstract: With the development of large models, watermarks are increasingly employed to assert copyright, verify authenticity, or monitor content distribution. As applications become more multimodal, the utility of watermarking techniques becomes even more critical. The effectiveness and reliability of these watermarks largely depend on their robustness to various disturbances. However, the robustness of th… ▽ More With the development of large models, watermarks are increasingly employed to assert copyright, verify authenticity, or monitor content distribution. As applications become more multimodal, the utility of watermarking techniques becomes even more critical. The effectiveness and reliability of these watermarks largely depend on their robustness to various disturbances. However, the robustness of these watermarks in real-world scenarios, particularly under perturbations and corruption, is not well understood. To highlight the significance of robustness in watermarking techniques, our study evaluated the robustness of watermarked content generated by image and text generation models against common real-world image corruptions and text perturbations. Our results could pave the way for the development of more robust watermarking techniques in the future. Our project website can be found at \url{https://mmwatermark-robustness.github.io/}. △ Less

Submitted 5 June, 2024; originally announced June 2024.

Revisiting and Maximizing Temporal Knowledge in Semi-supervised Semantic Segmentation

Authors: Wooseok Shin, Hyun Joon Park, Jin Sob Kim, Sung Won Han

Abstract: In semi-supervised semantic segmentation, the Mean Teacher- and co-training-based approaches are employed to mitigate confirmation bias and coupling problems. However, despite their high performance, these approaches frequently involve complex training pipelines and a substantial computational burden, limiting the scalability and compatibility of these methods. In this paper, we propose a PrevMatc… ▽ More In semi-supervised semantic segmentation, the Mean Teacher- and co-training-based approaches are employed to mitigate confirmation bias and coupling problems. However, despite their high performance, these approaches frequently involve complex training pipelines and a substantial computational burden, limiting the scalability and compatibility of these methods. In this paper, we propose a PrevMatch framework that effectively mitigates the aforementioned limitations by maximizing the utilization of the temporal knowledge obtained during the training process. The PrevMatch framework relies on two core strategies: (1) we reconsider the use of temporal knowledge and thus directly utilize previous models obtained during training to generate additional pseudo-label guidance, referred to as previous guidance. (2) we design a highly randomized ensemble strategy to maximize the effectiveness of the previous guidance. Experimental results on four benchmark semantic segmentation datasets confirm that the proposed method consistently outperforms existing methods across various evaluation protocols. In particular, with DeepLabV3+ and ResNet-101 network settings, PrevMatch outperforms the existing state-of-the-art method, Diverse Co-training, by +1.6 mIoU on Pascal VOC with only 92 annotated images, while achieving 2.4 times faster training. Furthermore, the results indicate that PrevMatch induces stable optimization, particularly in benefiting classes that exhibit poor performance. Code is available at https://github.com/wooseok-shin/PrevMatch △ Less

Submitted 30 May, 2024; originally announced May 2024.

Comments: 14 pages, 5 figures, submitted to IEEE TPAMI. This work has been submitted to the IEEE for possible publication

Numerical analysis of a 1/2-equation model of turbulence

Authors: Wei-Wei Han, Rui Fang, William Layton

Abstract: The recent 1/2-equation model of turbulence is a simplification of the standard Kolmogorov-Prandtl 1-equation URANS model. Surprisingly, initial numerical tests indicated that the 1/2-equation model produces comparable velocity statistics at reduced cost. It is also a test problem and first step for developing numerical analysis to address a full 1-equation model. This report begins the numerical… ▽ More The recent 1/2-equation model of turbulence is a simplification of the standard Kolmogorov-Prandtl 1-equation URANS model. Surprisingly, initial numerical tests indicated that the 1/2-equation model produces comparable velocity statistics at reduced cost. It is also a test problem and first step for developing numerical analysis to address a full 1-equation model. This report begins the numerical analysis of the 1/2 equation model. Stability, convergence and error estimates are proven for a semi-discrete and fully discrete approximation. Finally, numerical tests are conducted to validate our convergence theory. △ Less

Submitted 29 May, 2024; originally announced May 2024.

Crowdsourcing with Difficulty: A Bayesian Rating Model for Heterogeneous Items

Authors: Seong Woo Han, Ozan Adıgüzel, Bob Carpenter

Abstract: In applied statistics and machine learning, the "gold standards" used for training are often biased and almost always noisy. Dawid and Skene's justifiably popular crowdsourcing model adjusts for rater (coder, annotator) sensitivity and specificity, but fails to capture distributional properties of rating data gathered for training, which in turn biases training. In this study, we introduce a gener… ▽ More In applied statistics and machine learning, the "gold standards" used for training are often biased and almost always noisy. Dawid and Skene's justifiably popular crowdsourcing model adjusts for rater (coder, annotator) sensitivity and specificity, but fails to capture distributional properties of rating data gathered for training, which in turn biases training. In this study, we introduce a general purpose measurement-error model with which we can infer consensus categories by adding item-level effects for difficulty, discriminativeness, and guessability. We further show how to constrain the bimodal posterior of these models to avoid (or if necessary, allow) adversarial raters. We validate our model's goodness of fit with posterior predictive checks, the Bayesian analogue of $χ^2$ tests. Dawid and Skene's model is rejected by goodness of fit tests, whereas our new model, which adjusts for item heterogeneity, is not rejected. We illustrate our new model with two well-studied data sets, binary rating data for caries in dental X-rays and implication in natural language. △ Less

Submitted 21 October, 2024; v1 submitted 29 May, 2024; originally announced May 2024.

Precision measurement of the branching fraction of \boldmath $J/ψ\rightarrow K^+K^-$ via $ψ(2S)\rightarrow π^+π^-J/ψ$

Authors: BESIII Collaboration, M. Ablikim, M. N. Achasov, P. Adlarson, X. C. Ai, R. Aliberti, A. Amoroso, M. R. An, Q. An, Y. Bai, O. Bakina, I. Balossino, Y. Ban, H. -R. Bao, V. Batozskaya, K. Begzsuren, N. Berger, M. Berlowski, M. Bertani, D. Bettoni, F. Bianchi, E. Bianco, A. Bortone, I. Boyko, R. A. Briere , et al. (604 additional authors not shown)

Abstract: Using a sample of $448.1 \times 10^6$ $ψ(2S)$ events collected with the BESIII detector, we perform a study of the decay $J/ψ\rightarrow K^+K^-$ via $ψ(2S)\rightarrow π^+π^-J/ψ$. The branching fraction of $J/ψ\rightarrow K^+K^-$ is determined to be $\mathcal{B}_{K^+K^-}=(3.072\pm 0.023({\rm stat.})\pm 0.050({\rm syst.}))\times 10^{-4}$, which is consistent with previous measurements but with sig… ▽ More Using a sample of $448.1 \times 10^6$ $ψ(2S)$ events collected with the BESIII detector, we perform a study of the decay $J/ψ\rightarrow K^+K^-$ via $ψ(2S)\rightarrow π^+π^-J/ψ$. The branching fraction of $J/ψ\rightarrow K^+K^-$ is determined to be $\mathcal{B}_{K^+K^-}=(3.072\pm 0.023({\rm stat.})\pm 0.050({\rm syst.}))\times 10^{-4}$, which is consistent with previous measurements but with significantly improved precision. △ Less

Submitted 21 May, 2024; originally announced May 2024.

Comments: to be submitted to PRD

Transformer in Touch: A Survey

Authors: Jing Gao, Ning Cheng, Bin Fang, Wenjuan Han

Abstract: The Transformer model, initially achieving significant success in the field of natural language processing, has recently shown great potential in the application of tactile perception. This review aims to comprehensively outline the application and development of Transformers in tactile technology. We first introduce the two fundamental concepts behind the success of the Transformer: the self-atte… ▽ More The Transformer model, initially achieving significant success in the field of natural language processing, has recently shown great potential in the application of tactile perception. This review aims to comprehensively outline the application and development of Transformers in tactile technology. We first introduce the two fundamental concepts behind the success of the Transformer: the self-attention mechanism and large-scale pre-training. Then, we delve into the application of Transformers in various tactile tasks, including but not limited to object recognition, cross-modal generation, and object manipulation, offering a concise summary of the core methodologies, performance benchmarks, and design highlights. Finally, we suggest potential areas for further research and future work, aiming to generate more interest within the community, tackle existing challenges, and encourage the use of Transformer models in the tactile field. △ Less

Submitted 21 May, 2024; originally announced May 2024.

Comments: 27 pages, 2 tables, 5 figures, accepted by ICIC 2024

The Effect of Higher Harmonics On Gravitational Wave Dark Sirens

Authors: Jian-Dong Liu, Wen-Biao Han, Qianyun Yun, Shu-Cheng Yang

Abstract: The gravitational wave (GW) signal from the merger of two black holes can serve as a standard sirens for cosmological inference. However, a degeneracy exists between the luminosity distance and the inclination angle between the binary system's orbital angular momentum and the observer's line of sight, limiting the precise measurement of the luminosity distance. In this study, we investigate how hi… ▽ More The gravitational wave (GW) signal from the merger of two black holes can serve as a standard sirens for cosmological inference. However, a degeneracy exists between the luminosity distance and the inclination angle between the binary system's orbital angular momentum and the observer's line of sight, limiting the precise measurement of the luminosity distance. In this study, we investigate how higher harmonics affect luminosity distance estimation for third-generation (3G) GW detectors in binary black hole mergers. Our findings demonstrate that considering higher harmonics significantly enhances distance inference results compared with using only the (2,2) mode. This improved accuracy in distance estimates also strengthens constraints on host galaxies, enabling more precise measurements of the Hubble constant. These results highlight the significant influence of higher harmonics on the range estimation accuracy of 3G ground-based GW detectors. △ Less

Submitted 18 May, 2024; originally announced May 2024.

Comments: 14 pages, 6 figures

An Embarrassingly Simple Approach to Enhance Transformer Performance in Genomic Selection for Crop Breeding

Authors: Renqi Chen, Wenwei Han, Haohao Zhang, Haoyang Su, Zhefan Wang, Xiaolei Liu, Hao Jiang, Wanli Ouyang, Nanqing Dong

Abstract: Genomic selection (GS), as a critical crop breeding strategy, plays a key role in enhancing food production and addressing the global hunger crisis. The predominant approaches in GS currently revolve around employing statistical methods for prediction. However, statistical methods often come with two main limitations: strong statistical priors and linear assumptions. A recent trend is to capture t… ▽ More Genomic selection (GS), as a critical crop breeding strategy, plays a key role in enhancing food production and addressing the global hunger crisis. The predominant approaches in GS currently revolve around employing statistical methods for prediction. However, statistical methods often come with two main limitations: strong statistical priors and linear assumptions. A recent trend is to capture the non-linear relationships between markers by deep learning. However, as crop datasets are commonly long sequences with limited samples, the robustness of deep learning models, especially Transformers, remains a challenge. In this work, to unleash the unexplored potential of attention mechanism for the task of interest, we propose a simple yet effective Transformer-based framework that enables end-to-end training of the whole sequence. Via experiments on rice3k and wheat3k datasets, we show that, with simple tricks such as k-mer tokenization and random masking, Transformer can achieve overall superior performance against seminal methods on GS tasks of interest. △ Less

Submitted 24 June, 2024; v1 submitted 15 May, 2024; originally announced May 2024.

Comments: Accepted by IJCAI2024. Code is available at https://github.com/RenqiChen/Genomic-Selection

ODFormer: Semantic Fundus Image Segmentation Using Transformer for Optic Nerve Head Detection

Authors: Jiayi Wang, Yi-An Mao, Xiaoyu Ma, Sicen Guo, Yuting Shao, Xiao Lv, Wenting Han, Mark Christopher, Linda M. Zangwill, Yanlong Bi, Rui Fan

Abstract: Optic nerve head (ONH) detection has been a crucial area of study in ophthalmology for years. However, the significant discrepancy between fundus image datasets, each generated using a single type of fundus camera, poses challenges to the generalizability of ONH detection approaches developed based on semantic segmentation networks. Despite the numerous recent advancements in general-purpose seman… ▽ More Optic nerve head (ONH) detection has been a crucial area of study in ophthalmology for years. However, the significant discrepancy between fundus image datasets, each generated using a single type of fundus camera, poses challenges to the generalizability of ONH detection approaches developed based on semantic segmentation networks. Despite the numerous recent advancements in general-purpose semantic segmentation methods using convolutional neural networks (CNNs) and Transformers, there is currently a lack of benchmarks for these state-of-the-art (SoTA) networks specifically trained for ONH detection. Therefore, in this article, we make contributions from three key aspects: network design, the publication of a dataset, and the establishment of a comprehensive benchmark. Our newly developed ONH detection network, referred to as ODFormer, is based upon the Swin Transformer architecture and incorporates two novel components: a multi-scale context aggregator and a lightweight bidirectional feature recalibrator. Our published large-scale dataset, known as TongjiU-DROD, provides multi-resolution fundus images for each participant, captured using two distinct types of cameras. Our established benchmark involves three datasets: DRIONS-DB, DRISHTI-GS1, and TongjiU-DROD, created by researchers from different countries and containing fundus images captured from participants of diverse races and ages. Extensive experimental results demonstrate that our proposed ODFormer outperforms other state-of-the-art (SoTA) networks in terms of performance and generalizability. Our dataset and source code are publicly available at mias.group/ODFormer. △ Less

Submitted 2 June, 2024; v1 submitted 15 April, 2024; originally announced May 2024.

Search for the leptonic decays $D^{*+}\to e^+ν_e$ and $D^{*+}\to μ^+ν_μ$

Authors: BESIII Collaboration, M. Ablikim, M. N. Achasov, P. Adlarson, M. Albrecht, R. Aliberti, A. Amoroso, M. R. An, Q. An, Y. Bai, O. Bakina, R. Baldini Ferroli, I. Balossino, Y. Ban, V. Batozskaya, D. Becker, K. Begzsuren, N. Berger, M. Bertani, D. Bettoni, F. Bianchi, E. Bianco, J. Bloms, A. Bortone, I. Boyko , et al. (559 additional authors not shown)

Abstract: We present the first search for the leptonic decays $D^{*+}\to e^+ν_e$ and $D^{*+}\to μ^+ν_μ$ by analyzing a data sample of electron-positron collisions recorded with the BESIII detector at center-of-mass energies between 4.178 and 4.226 GeV, corresponding to an integrated luminosity of 6.32~fb$^{-1}$. No significant signal is observed. The upper limits on the branching fractions for… ▽ More We present the first search for the leptonic decays $D^{*+}\to e^+ν_e$ and $D^{*+}\to μ^+ν_μ$ by analyzing a data sample of electron-positron collisions recorded with the BESIII detector at center-of-mass energies between 4.178 and 4.226 GeV, corresponding to an integrated luminosity of 6.32~fb$^{-1}$. No significant signal is observed. The upper limits on the branching fractions for $D^{*+}\to e^+ν_e$ and $D^{*+}\to μ^+ν_μ$ are set to be $1.1 \times 10^{-5}$ and $4.3 \times 10^{-6}$ at 90\% confidence level, respectively. △ Less

Submitted 14 May, 2024; originally announced May 2024.

Comments: 14 pages, 7 figures

Journal ref: Phys. Rev. D 110, 012003 (2024)

Beyond Scaling Laws: Understanding Transformer Performance with Associative Memory

Authors: Xueyan Niu, Bo Bai, Lei Deng, Wei Han

Abstract: Increasing the size of a Transformer does not always lead to enhanced performance. This phenomenon cannot be explained by the empirical scaling laws. Furthermore, the model's enhanced performance is closely associated with its memorization of the training samples. We present a theoretical framework that sheds light on the memorization during pre-training of transformer-based language models. We mo… ▽ More Increasing the size of a Transformer does not always lead to enhanced performance. This phenomenon cannot be explained by the empirical scaling laws. Furthermore, the model's enhanced performance is closely associated with its memorization of the training samples. We present a theoretical framework that sheds light on the memorization during pre-training of transformer-based language models. We model the behavior of Transformers with associative memories using Hopfield networks, such that each transformer block effectively conducts an approximate nearest-neighbor search. In particular, the energy function in modern continuous Hopfield networks serves as an explanation for the attention mechanism, which we approximate with a distance-based energy function. By observing that the softmax function corresponds to the gradient of the LogSumExp function in the energy, and employing the majorization-minimization technique, we construct a global energy function designed to capture the layered architecture. We demonstrate a dependency between the model size and the dataset size for the model to achieve optimal performance, and we show that the achievable cross-entropy loss is bounded from below. △ Less

Submitted 27 November, 2024; v1 submitted 14 May, 2024; originally announced May 2024.

Potential and Limitations of LLMs in Capturing Structured Semantics: A Case Study on SRL

Authors: Ning Cheng, Zhaohui Yan, Ziming Wang, Zhijie Li, Jiaming Yu, Zilong Zheng, Kewei Tu, Jinan Xu, Wenjuan Han

Abstract: Large Language Models (LLMs) play a crucial role in capturing structured semantics to enhance language understanding, improve interpretability, and reduce bias. Nevertheless, an ongoing controversy exists over the extent to which LLMs can grasp structured semantics. To assess this, we propose using Semantic Role Labeling (SRL) as a fundamental task to explore LLMs' ability to extract structured se… ▽ More Large Language Models (LLMs) play a crucial role in capturing structured semantics to enhance language understanding, improve interpretability, and reduce bias. Nevertheless, an ongoing controversy exists over the extent to which LLMs can grasp structured semantics. To assess this, we propose using Semantic Role Labeling (SRL) as a fundamental task to explore LLMs' ability to extract structured semantics. In our assessment, we employ the prompting approach, which leads to the creation of our few-shot SRL parser, called PromptSRL. PromptSRL enables LLMs to map natural languages to explicit semantic structures, which provides an interpretable window into the properties of LLMs. We find interesting potential: LLMs can indeed capture semantic structures, and scaling-up doesn't always mirror potential. Additionally, limitations of LLMs are observed in C-arguments, etc. Lastly, we are surprised to discover that significant overlap in the errors is made by both LLMs and untrained humans, accounting for almost 30% of all errors. △ Less

Submitted 10 May, 2024; originally announced May 2024.

Comments: Accepted by ICIC 2024

Tests of gravitational wave propagation with LIGO-Virgo catalog

Authors: Xian-Liang Wang, Shu-Cheng Yang, Wen-Biao Han

Abstract: In the framework of general relativity (GR), gravitational waves (GWs) travel at the speed of light across all frequencies. However, massive gravity and weak equivalence principle (WEP) violation may lead to frequency-dependent variations in the propagation speed of GWs, which can be examined by comparing the theoretical and observed discrepancies in the arrival times of GW signals at various freq… ▽ More In the framework of general relativity (GR), gravitational waves (GWs) travel at the speed of light across all frequencies. However, massive gravity and weak equivalence principle (WEP) violation may lead to frequency-dependent variations in the propagation speed of GWs, which can be examined by comparing the theoretical and observed discrepancies in the arrival times of GW signals at various frequencies. This provides us with an opportunity to test these theories. For massive gravity, we consider that gravitons may have a nonzero rest mass. For WEP violations, we hypothesize that different massless particles exposed to the same gravitational source should exhibit varying gravitational time delays. The gravitational time delay induced by massive gravitational sources is proportional to $γ+1$, where the parameter $γ=1$ in GR. Therefore, we can quantify these two deviations using phenomenological parameters $m_g$ and $|Δγ|$, respectively. In this study, we use selected GW data from binary black hole coalescences in the LIGO-Virgo catalogs GWTC-2.1 and GWTC-3 to place constraints on the parameters $m_g$ and $|Δγ|$. We analyze the relationship between $m_g$ and luminosity distance,as well as between $|Δγ|$ and both luminosity distance sky location of GW events to determine the presence of graviton mass and WEP violation. Nevertheless, we find no evidence of such relationships. We also compute Bayes factors for models that assume the existence of graviton mass and WEP violation compared to the standard GW model, respectively. The absolute value of the natural logarithm of the Bayes factor is generally less than 2. Our analysis reveals no significant preference for either model. Additionally, the Bayes factors between these two models do not provide obvious evidence in favor of either one. △ Less

Submitted 5 January, 2025; v1 submitted 22 April, 2024; originally announced April 2024.

Comments: 11 pages, 10 figures

Fuxi-DA: A Generalized Deep Learning Data Assimilation Framework for Assimilating Satellite Observations

Authors: Xiaoze Xu, Xiuyu Sun, Wei Han, Xiaohui Zhong, Lei Chen, Hao Li

Abstract: Data assimilation (DA), as an indispensable component within contemporary Numerical Weather Prediction (NWP) systems, plays a crucial role in generating the analysis that significantly impacts forecast performance. Nevertheless, the development of an efficient DA system poses significant challenges, particularly in establishing intricate relationships between the background data and the vast amoun… ▽ More Data assimilation (DA), as an indispensable component within contemporary Numerical Weather Prediction (NWP) systems, plays a crucial role in generating the analysis that significantly impacts forecast performance. Nevertheless, the development of an efficient DA system poses significant challenges, particularly in establishing intricate relationships between the background data and the vast amount of multi-source observation data within limited time windows in operational settings. To address these challenges, researchers design complex pre-processing methods for each observation type, leveraging approximate modeling and the power of super-computing clusters to expedite solutions. The emergence of deep learning (DL) models has been a game-changer, offering unified multi-modal modeling, enhanced nonlinear representation capabilities, and superior parallelization. These advantages have spurred efforts to integrate DL models into various domains of weather modeling. Remarkably, DL models have shown promise in matching, even surpassing, the forecast accuracy of leading operational NWP models worldwide. This success motivates the exploration of DL-based DA frameworks tailored for weather forecasting models. In this study, we introduces FuxiDA, a generalized DL-based DA framework for assimilating satellite observations. By assimilating data from Advanced Geosynchronous Radiation Imager (AGRI) aboard Fengyun-4B, FuXi-DA consistently mitigates analysis errors and significantly improves forecast performance. Furthermore, through a series of single-observation experiments, Fuxi-DA has been validated against established atmospheric physics, demonstrating its consistency and reliability. △ Less

Submitted 12 April, 2024; originally announced April 2024.

Measurement of $e^{+}e^{-}\to ωη^{\prime}$ cross sections at $\sqrt{s}=$ 2.000 to 3.080 GeV

Authors: BESIII Collaboration, M. Ablikim, M. N. Achasov, P. Adlarson, X. C. Ai, R. Aliberti, A. Amoroso, M. R. An, Q. An, Y. Bai, O. Bakina, I. Balossino, Y. Ban, V. Batozskaya, K. Begzsuren, N. Berger, M. Berlowski, M. Bertani, D. Bettoni, F. Bianchi, E. Bianco, A. Bortone, I. Boyko, R. A. Briere, A. Brueggemann , et al. (599 additional authors not shown)

Abstract: The Born cross sections for the process $e^{+}e^{-}\to ωη^{\prime}$ are measured at 22 center-of-mass energies from 2.000 to 3.080 GeV using data collected with the BESIII detector at the BEPCII collider. A resonant structure is observed with a statistical significance of 9.6$σ$. A Breit-Wigner fit determines its mass to be $M_R=(2153\pm30\pm31)~{\rm{MeV}}/c^{2}$ and its width to be… ▽ More The Born cross sections for the process $e^{+}e^{-}\to ωη^{\prime}$ are measured at 22 center-of-mass energies from 2.000 to 3.080 GeV using data collected with the BESIII detector at the BEPCII collider. A resonant structure is observed with a statistical significance of 9.6$σ$. A Breit-Wigner fit determines its mass to be $M_R=(2153\pm30\pm31)~{\rm{MeV}}/c^{2}$ and its width to be $Γ_{R}=(167\pm77\pm7)~\rm{MeV}$, where the first uncertainties are statistical and the second are systematic. △ Less

Submitted 10 April, 2024; originally announced April 2024.

JDEC: JPEG Decoding via Enhanced Continuous Cosine Coefficients

Authors: Woo Kyoung Han, Sunghoon Im, Jaedeok Kim, Kyong Hwan Jin

Abstract: We propose a practical approach to JPEG image decoding, utilizing a local implicit neural representation with continuous cosine formulation. The JPEG algorithm significantly quantizes discrete cosine transform (DCT) spectra to achieve a high compression rate, inevitably resulting in quality degradation while encoding an image. We have designed a continuous cosine spectrum estimator to address the… ▽ More We propose a practical approach to JPEG image decoding, utilizing a local implicit neural representation with continuous cosine formulation. The JPEG algorithm significantly quantizes discrete cosine transform (DCT) spectra to achieve a high compression rate, inevitably resulting in quality degradation while encoding an image. We have designed a continuous cosine spectrum estimator to address the quality degradation issue that restores the distorted spectrum. By leveraging local DCT formulations, our network has the privilege to exploit dequantization and upsampling simultaneously. Our proposed model enables decoding compressed images directly across different quality factors using a single pre-trained model without relying on a conventional JPEG decoder. As a result, our proposed network achieves state-of-the-art performance in flexible color image JPEG artifact removal tasks. Our source code is available at https://github.com/WooKyoungHan/JDEC. △ Less

Submitted 2 April, 2024; originally announced April 2024.

Search for di-photon decays of an axion-like particle in radiative $J/ψ$ decays

Authors: BESIII Collaboration, M. Ablikim, M. N. Achasov, P. Adlarson, O. Afedulidis, X. C. Ai, R. Aliberti, A. Amoroso, M. R. An, Q. An, Y. Bai, O. Bakina, I. Balossino, Y. Ban, H. -R. Bao, V. Batozskaya, K. Begzsuren, N. Berger, M. Berlowski, M. Bertani, D. Bettoni, F. Bianchi, E. Bianco, A. Bortone, I. Boyko , et al. (604 additional authors not shown)

Abstract: We search for the di-photon decay of a light pseudoscalar axion-like particle, $a$, in radiative $J/ψ$ decays, using 10 billion $J/ψ$ events collected with the BESIII detector. We find no evidence of a signal and set upper limits at the $95\%$ confidence level on the product branching fraction $\mathcal{B}(J/ψ\to γa) \times \mathcal{B}(a \to γγ)$ and the axion-like particle photon coupling constan… ▽ More We search for the di-photon decay of a light pseudoscalar axion-like particle, $a$, in radiative $J/ψ$ decays, using 10 billion $J/ψ$ events collected with the BESIII detector. We find no evidence of a signal and set upper limits at the $95\%$ confidence level on the product branching fraction $\mathcal{B}(J/ψ\to γa) \times \mathcal{B}(a \to γγ)$ and the axion-like particle photon coupling constant $g_{a γγ}$ in the ranges of $(3.7-48.5) \times 10^{-8}$ and $(2.2 -101.8)\times 10^{-4}$ GeV$^{-1}$, respectively, for $0.18 \le m_a \le 2.85$ GeV/$c^2$. These are the most stringent limits to date in this mass region. △ Less

Submitted 6 August, 2024; v1 submitted 6 April, 2024; originally announced April 2024.

Comments: 9 pages, 5 figures, Published in Phys. Rev. D 110, L031101 (2024)

Report number: BESIII Analysis Memo - 671

Observation of Gravitational Waves from the Coalescence of a $2.5\text{-}4.5~M_\odot$ Compact Object and a Neutron Star

Authors: The LIGO Scientific Collaboration, the Virgo Collaboration, the KAGRA Collaboration, A. G. Abac, R. Abbott, I. Abouelfettouh, F. Acernese, K. Ackley, S. Adhicary, N. Adhikari, R. X. Adhikari, V. K. Adkins, D. Agarwal, M. Agathos, M. Aghaei Abchouyeh, O. D. Aguiar, I. Aguilar, L. Aiello, A. Ain, P. Ajith, S. Akçay, T. Akutsu, S. Albanesi, R. A. Alfaidi, A. Al-Jodah , et al. (1771 additional authors not shown)

Abstract: We report the observation of a coalescing compact binary with component masses $2.5\text{-}4.5~M_\odot$ and $1.2\text{-}2.0~M_\odot$ (all measurements quoted at the 90% credible level). The gravitational-wave signal GW230529_181500 was observed during the fourth observing run of the LIGO-Virgo-KAGRA detector network on 2023 May 29 by the LIGO Livingston Observatory. The primary component of the so… ▽ More We report the observation of a coalescing compact binary with component masses $2.5\text{-}4.5~M_\odot$ and $1.2\text{-}2.0~M_\odot$ (all measurements quoted at the 90% credible level). The gravitational-wave signal GW230529_181500 was observed during the fourth observing run of the LIGO-Virgo-KAGRA detector network on 2023 May 29 by the LIGO Livingston Observatory. The primary component of the source has a mass less than $5~M_\odot$ at 99% credibility. We cannot definitively determine from gravitational-wave data alone whether either component of the source is a neutron star or a black hole. However, given existing estimates of the maximum neutron star mass, we find the most probable interpretation of the source to be the coalescence of a neutron star with a black hole that has a mass between the most massive neutron stars and the least massive black holes observed in the Galaxy. We provisionally estimate a merger rate density of $55^{+127}_{-47}~\text{Gpc}^{-3}\,\text{yr}^{-1}$ for compact binary coalescences with properties similar to the source of GW230529_181500; assuming that the source is a neutron star-black hole merger, GW230529_181500-like sources constitute about 60% of the total merger rate inferred for neutron star-black hole coalescences. The discovery of this system implies an increase in the expected rate of neutron star-black hole mergers with electromagnetic counterparts and provides further evidence for compact objects existing within the purported lower mass gap. △ Less

Submitted 26 July, 2024; v1 submitted 5 April, 2024; originally announced April 2024.

Comments: 45 pages (10 pages author list, 13 pages main text, 1 page acknowledgements, 13 pages appendices, 8 pages bibliography), 17 figures, 16 tables. Update to match version published in The Astrophysical Journal Letters. Data products available from https://zenodo.org/records/10845779

Report number: LIGO-P2300352

Journal ref: ApJL 970, L34 (2024)

MineLand: Simulating Large-Scale Multi-Agent Interactions with Limited Multimodal Senses and Physical Needs

Authors: Xianhao Yu, Jiaqi Fu, Renjia Deng, Wenjuan Han

Abstract: While Vision-Language Models (VLMs) hold promise for tasks requiring extensive collaboration, traditional multi-agent simulators have facilitated rich explorations of an interactive artificial society that reflects collective behavior. However, these existing simulators face significant limitations. Firstly, they struggle with handling large numbers of agents due to high resource demands. Secondly… ▽ More While Vision-Language Models (VLMs) hold promise for tasks requiring extensive collaboration, traditional multi-agent simulators have facilitated rich explorations of an interactive artificial society that reflects collective behavior. However, these existing simulators face significant limitations. Firstly, they struggle with handling large numbers of agents due to high resource demands. Secondly, they often assume agents possess perfect information and limitless capabilities, hindering the ecological validity of simulated social interactions. To bridge this gap, we propose a multi-agent Minecraft simulator, MineLand, that bridges this gap by introducing three key features: large-scale scalability, limited multimodal senses, and physical needs. Our simulator supports 64 or more agents. Agents have limited visual, auditory, and environmental awareness, forcing them to actively communicate and collaborate to fulfill physical needs like food and resources. Additionally, we further introduce an AI agent framework, Alex, inspired by multitasking theory, enabling agents to handle intricate coordination and scheduling. Our experiments demonstrate that the simulator, the corresponding benchmark, and the AI agent framework contribute to more ecological and nuanced collective behavior.The source code of MineLand and Alex is openly available at https://github.com/cocacola-lab/MineLand. △ Less

Submitted 23 May, 2024; v1 submitted 28 March, 2024; originally announced March 2024.

Comments: Project website: https://github.com/cocacola-lab/MineLand

Observation of the semileptonic decays $D^0\rightarrow K_S^0π^-π^0 e^+ ν_e$ and $D^+\rightarrow K_S^0π^+π^- e^+ ν_e$

Authors: BESIII Collaboration, M. Ablikim, M. N. Achasov, P. Adlarson, X. C. Ai, R. Aliberti, A. Amoroso, M. R. An, Q. An, Y. Bai, O. Bakina, I. Balossino, Y. Ban, V. Batozskaya, K. Begzsuren, N. Berger, M. Berlowski, M. Bertani, D. Bettoni, F. Bianchi, E. Bianco, A. Bortone, I. Boyko, R. A. Briere, A. Brueggemann , et al. (600 additional authors not shown)

Abstract: By analyzing $e^+e^-$ annihilation data corresponding to an integrated luminosity of 2.93 $\rm fb^{-1}$ collected at a center-of-mass energy of 3.773 GeV with the \text{BESIII} detector, the first observation of the semileptonic decays $D^0\rightarrow K_S^0π^-π^0 e^+ ν_e$ and $D^+\rightarrow K_S^0π^+π^- e^+ ν_e$ is reported. With a dominant hadronic contribution from $K_1(1270)$, the branching fra… ▽ More By analyzing $e^+e^-$ annihilation data corresponding to an integrated luminosity of 2.93 $\rm fb^{-1}$ collected at a center-of-mass energy of 3.773 GeV with the \text{BESIII} detector, the first observation of the semileptonic decays $D^0\rightarrow K_S^0π^-π^0 e^+ ν_e$ and $D^+\rightarrow K_S^0π^+π^- e^+ ν_e$ is reported. With a dominant hadronic contribution from $K_1(1270)$, the branching fractions are measured to be $\mathcal{B}(D^0\rightarrow {K}_1(1270)^-(\to K^0_Sπ^-π^0)e^+ν_e)=(1.69^{+0.53}_{-0.46}\pm0.15)\times10^{-4}$ and $\mathcal{B}(D^+\to \bar{K}_1(1270)^0(\to K^0_Sπ^+π^-)e^+ν_e)=(1.47^{+0.45}_{-0.40}\pm0.20)\times10^{-4}$ with statistical significance of 5.4$σ$ and 5.6$σ$, respectively. When combined with measurements of the $K_1(1270)\to K^+π^-π$ decays, the absolute branching fractions are determined to be $\mathcal{B}(D^0\to K_1(1270)^-e^+ν_e)=(1.05^{+0.33}_{-0.28}\pm0.12\pm0.12)\times10^{-3}$ and $\mathcal{B}(D^+\to \bar{K}_1(1270)^0e^+ν_e)=(1.29^{+0.40}_{-0.35}\pm0.18\pm0.15)\times10^{-3}$. The first and second uncertainties are statistical and systematic, respectively, and the third uncertainties originate from the assumed branching fractions of the $K_1(1270)\to Kππ$ decays. △ Less

Submitted 27 March, 2024; originally announced March 2024.

Comments: 19pages

Receiver Resonant Frequency Adaptive Tracking in Wireless Power Transfer Systems Using Primary Variable Capacitor

Authors: Chang Liu, Wei Han, Guangyu Yan, Bowang Zhang, Chunlin Li

Abstract: Parameter variations within the resonant network of wireless power transfer (WPT) systems can cause drift in the resonant frequency, leading to a detuned system that requires higher power capacity and experiences reduced transfer efficiency. To address this issue, this paper presents an adaptive online receiver resonant frequency tracking scheme based solely on primary-side detection. The proposed… ▽ More Parameter variations within the resonant network of wireless power transfer (WPT) systems can cause drift in the resonant frequency, leading to a detuned system that requires higher power capacity and experiences reduced transfer efficiency. To address this issue, this paper presents an adaptive online receiver resonant frequency tracking scheme based solely on primary-side detection. The proposed method effectively compensates for parameter fluctuations in both primary and secondary resonators. The core of this approach is a switch-controlled capacitor (SCC) with a control angle calibrated during a system self-check process prior to high-power charging. Additionally, a two-step perturb-and-observe algorithm has been developed to perform online tracking while minimizing disturbances to the output power. Post-tracking, zero-voltage switching (ZVS) conditions can be achieved within a specified detuning range. To validate the efficacy of the proposed system, a 200W experimental platform was constructed. The measured results demonstrate that resonance is consistently maintained within the 79-90 kHz frequency range, as specified by the SAE J2954 standard. The maximum frequency tracking error and efficiency increase are 0.7 kHz and 9%, respectively. Notably, the tracking process is completed in less than 1 ms. △ Less

Submitted 12 September, 2024; v1 submitted 26 March, 2024; originally announced March 2024.

Comments: 11 pages,16 figures

Entity6K: A Large Open-Domain Evaluation Dataset for Real-World Entity Recognition

Authors: Jielin Qiu, William Han, Winfred Wang, Zhengyuan Yang, Linjie Li, Jianfeng Wang, Christos Faloutsos, Lei Li, Lijuan Wang

Abstract: Open-domain real-world entity recognition is essential yet challenging, involving identifying various entities in diverse environments. The lack of a suitable evaluation dataset has been a major obstacle in this field due to the vast number of entities and the extensive human effort required for data curation. We introduce Entity6K, a comprehensive dataset for real-world entity recognition, featur… ▽ More Open-domain real-world entity recognition is essential yet challenging, involving identifying various entities in diverse environments. The lack of a suitable evaluation dataset has been a major obstacle in this field due to the vast number of entities and the extensive human effort required for data curation. We introduce Entity6K, a comprehensive dataset for real-world entity recognition, featuring 5,700 entities across 26 categories, each supported by 5 human-verified images with annotations. Entity6K offers a diverse range of entity names and categorizations, addressing a gap in existing datasets. We conducted benchmarks with existing models on tasks like image captioning, object detection, zero-shot classification, and dense captioning to demonstrate Entity6K's effectiveness in evaluating models' entity recognition capabilities. We believe Entity6K will be a valuable resource for advancing accurate entity recognition in open-domain settings. △ Less

Submitted 18 March, 2024; originally announced March 2024.

Anomalous Raman Response in 2D Magnetic FeTe under Uniaxial Strain: Tetragonal and Hexagonal Polymorphs

Authors: Wuxiao Han, Tiansong Zhang, Pengcheng Zhao, Longfei Yang, Mo Cheng, Jianping Shi, Yabin Chen

Abstract: Two-dimensional (2D) Fe-chalcogenides have emerged with rich structures, magnetisms and superconductivities, which sparked the growing research interests in the torturous transition mechanism and tunable properties for their potential applications in nanoelectronics. Uniaxial strain can produce a lattice distortion to study symmetry breaking induced exotic properties in 2D magnets. Herein, the ano… ▽ More Two-dimensional (2D) Fe-chalcogenides have emerged with rich structures, magnetisms and superconductivities, which sparked the growing research interests in the torturous transition mechanism and tunable properties for their potential applications in nanoelectronics. Uniaxial strain can produce a lattice distortion to study symmetry breaking induced exotic properties in 2D magnets. Herein, the anomalous Raman spectrum of 2D tetragonal (t-) and hexagonal (h-) FeTe were systematically investigated via uniaxial strain engineering strategy. We found that both t- and h-FeTe keep the structural stability under different uniaxial tensile or compressive strain up to +/- 0.4%. Intriguingly, the lattice vibrations along both in-plane and out-of-plane directions exceptionally hardened (softened) under tensile (compressive) strain, distinguished from the behaviors of many conventional 2D systems. Furthermore, the difference in thickness-dependent strain effect can be well explained by their structural discrepancy between two polymorphs of FeTe. Our results could provide a unique platform to elaborate the vibrational properties of many novel 2D materials. △ Less

Submitted 14 March, 2024; originally announced March 2024.

Comments: 21 pages, 5 figures

Towards Comprehensive Multimodal Perception: Introducing the Touch-Language-Vision Dataset

Authors: Ning Cheng, You Li, Jing Gao, Bin Fang, Jinan Xu, Wenjuan Han

Abstract: Tactility provides crucial support and enhancement for the perception and interaction capabilities of both humans and robots. Nevertheless, the multimodal research related to touch primarily focuses on visual and tactile modalities, with limited exploration in the domain of language. Beyond vocabulary, sentence-level descriptions contain richer semantics. Based on this, we construct a touch-langua… ▽ More Tactility provides crucial support and enhancement for the perception and interaction capabilities of both humans and robots. Nevertheless, the multimodal research related to touch primarily focuses on visual and tactile modalities, with limited exploration in the domain of language. Beyond vocabulary, sentence-level descriptions contain richer semantics. Based on this, we construct a touch-language-vision dataset named TLV (Touch-Language-Vision) by human-machine cascade collaboration, featuring sentence-level descriptions for multimode alignment. The new dataset is used to fine-tune our proposed lightweight training framework, STLV-Align (Synergistic Touch-Language-Vision Alignment), achieving effective semantic alignment with minimal parameter adjustments (1%). Project Page: https://xiaoen0.github.io/touch.page/. △ Less

Submitted 17 June, 2024; v1 submitted 14 March, 2024; originally announced March 2024.

Comments: Accepted by ICIC 2024

MCformer: Multivariate Time Series Forecasting with Mixed-Channels Transformer

Authors: Wenyong Han, Tao Zhu Member, Liming Chen, Huansheng Ning, Yang Luo, Yaping Wan

Abstract: The massive generation of time-series data by largescale Internet of Things (IoT) devices necessitates the exploration of more effective models for multivariate time-series forecasting. In previous models, there was a predominant use of the Channel Dependence (CD) strategy (where each channel represents a univariate sequence). Current state-of-the-art (SOTA) models primarily rely on the Channel In… ▽ More The massive generation of time-series data by largescale Internet of Things (IoT) devices necessitates the exploration of more effective models for multivariate time-series forecasting. In previous models, there was a predominant use of the Channel Dependence (CD) strategy (where each channel represents a univariate sequence). Current state-of-the-art (SOTA) models primarily rely on the Channel Independence (CI) strategy. The CI strategy treats all channels as a single channel, expanding the dataset to improve generalization performance and avoiding inter-channel correlation that disrupts long-term features. However, the CI strategy faces the challenge of interchannel correlation forgetting. To address this issue, we propose an innovative Mixed Channels strategy, combining the data expansion advantages of the CI strategy with the ability to counteract inter-channel correlation forgetting. Based on this strategy, we introduce MCformer, a multivariate time-series forecasting model with mixed channel features. The model blends a specific number of channels, leveraging an attention mechanism to effectively capture inter-channel correlation information when modeling long-term features. Experimental results demonstrate that the Mixed Channels strategy outperforms pure CI strategy in multivariate time-series forecasting tasks. △ Less

Submitted 14 March, 2024; originally announced March 2024.

CoBra: Complementary Branch Fusing Class and Semantic Knowledge for Robust Weakly Supervised Semantic Segmentation

Authors: Woojung Han, Seil Kang, Kyobin Choo, Seong Jae Hwang

Abstract: Leveraging semantically precise pseudo masks derived from image-level class knowledge for segmentation, namely image-level Weakly Supervised Semantic Segmentation (WSSS), still remains challenging. While Class Activation Maps (CAMs) using CNNs have steadily been contributing to the success of WSSS, the resulting activation maps often narrowly focus on class-specific parts (e.g., only face of human… ▽ More Leveraging semantically precise pseudo masks derived from image-level class knowledge for segmentation, namely image-level Weakly Supervised Semantic Segmentation (WSSS), still remains challenging. While Class Activation Maps (CAMs) using CNNs have steadily been contributing to the success of WSSS, the resulting activation maps often narrowly focus on class-specific parts (e.g., only face of human). On the other hand, recent works based on vision transformers (ViT) have shown promising results based on their self-attention mechanism to capture the semantic parts but fail in capturing complete class-specific details (e.g., entire body parts of human but also with a dog nearby). In this work, we propose Complementary Branch (CoBra), a novel dual branch framework consisting of two distinct architectures which provide valuable complementary knowledge of class (from CNN) and semantic (from ViT) to each branch. In particular, we learn Class-Aware Projection (CAP) for the CNN branch and Semantic-Aware Projection (SAP) for the ViT branch to explicitly fuse their complementary knowledge and facilitate a new type of extra patch-level supervision. Our model, through CoBra, fuses CNN and ViT's complementary outputs to create robust pseudo masks that integrate both class and semantic information effectively. Extensive experiments qualitatively and quantitatively investigate how CNN and ViT complement each other on the PASCAL VOC 2012 dataset, showing a state-of-the-art WSSS result. This includes not only the masks generated by our model, but also the segmentation results derived from utilizing these masks as pseudo labels. △ Less

Submitted 27 May, 2024; v1 submitted 5 February, 2024; originally announced March 2024.

Online Multi-Contact Feedback Model Predictive Control for Interactive Robotic Tasks

Authors: Seo Wook Han, Maged Iskandar, Jinoh Lee, Min Jun Kim

Abstract: In this paper, we propose a model predictive control (MPC) that accomplishes interactive robotic tasks, in which multiple contacts may occur at unknown locations. To address such scenarios, we made an explicit contact feedback loop in the MPC framework. An algorithm called Multi-Contact Particle Filter with Exploration Particle (MCP-EP) is employed to establish real-time feedback of multi-contact… ▽ More In this paper, we propose a model predictive control (MPC) that accomplishes interactive robotic tasks, in which multiple contacts may occur at unknown locations. To address such scenarios, we made an explicit contact feedback loop in the MPC framework. An algorithm called Multi-Contact Particle Filter with Exploration Particle (MCP-EP) is employed to establish real-time feedback of multi-contact information. Then the interaction locations and forces are accommodated in the MPC framework via a spring contact model. Moreover, we achieved real-time control for a 7 degrees of freedom robot without any simplifying assumptions by employing a Differential-Dynamic-Programming algorithm. We achieved 6.8kHz, 1.9kHz, and 1.8kHz update rates of the MPC for 0, 1, and 2 contacts, respectively. This allows the robot to handle unexpected contacts in real time. Real-world experiments show the effectiveness of the proposed method in various scenarios. △ Less

Submitted 13 March, 2024; originally announced March 2024.

Comments: This paper has been accepted for publication at the IEEE International Conference on Robotics and Automation (ICRA), Yokohama, 2024

Advancing Text-Driven Chest X-Ray Generation with Policy-Based Reinforcement Learning

Authors: Woojung Han, Chanyoung Kim, Dayun Ju, Yumin Shim, Seong Jae Hwang

Abstract: Recent advances in text-conditioned image generation diffusion models have begun paving the way for new opportunities in modern medical domain, in particular, generating Chest X-rays (CXRs) from diagnostic reports. Nonetheless, to further drive the diffusion models to generate CXRs that faithfully reflect the complexity and diversity of real data, it has become evident that a nontrivial learning a… ▽ More Recent advances in text-conditioned image generation diffusion models have begun paving the way for new opportunities in modern medical domain, in particular, generating Chest X-rays (CXRs) from diagnostic reports. Nonetheless, to further drive the diffusion models to generate CXRs that faithfully reflect the complexity and diversity of real data, it has become evident that a nontrivial learning approach is needed. In light of this, we propose CXRL, a framework motivated by the potential of reinforcement learning (RL). Specifically, we integrate a policy gradient RL approach with well-designed multiple distinctive CXR-domain specific reward models. This approach guides the diffusion denoising trajectory, achieving precise CXR posture and pathological details. Here, considering the complex medical image environment, we present "RL with Comparative Feedback" (RLCF) for the reward mechanism, a human-like comparative evaluation that is known to be more effective and reliable in complex scenarios compared to direct evaluation. Our CXRL framework includes jointly optimizing learnable adaptive condition embeddings (ACE) and the image generator, enabling the model to produce more accurate and higher perceptual CXR quality. Our extensive evaluation of the MIMIC-CXR-JPG dataset demonstrates the effectiveness of our RL-based tuning approach. Consequently, our CXRL generates pathologically realistic CXRs, establishing a new standard for generating CXRs with high fidelity to real-world clinical scenarios. △ Less

Submitted 11 March, 2024; originally announced March 2024.

HealMe: Harnessing Cognitive Reframing in Large Language Models for Psychotherapy

Authors: Mengxi Xiao, Qianqian Xie, Ziyan Kuang, Zhicheng Liu, Kailai Yang, Min Peng, Weiguang Han, Jimin Huang

Abstract: Large Language Models (LLMs) can play a vital role in psychotherapy by adeptly handling the crucial task of cognitive reframing and overcoming challenges such as shame, distrust, therapist skill variability, and resource scarcity. Previous LLMs in cognitive reframing mainly converted negative emotions to positive ones, but these approaches have limited efficacy, often not promoting clients' self-d… ▽ More Large Language Models (LLMs) can play a vital role in psychotherapy by adeptly handling the crucial task of cognitive reframing and overcoming challenges such as shame, distrust, therapist skill variability, and resource scarcity. Previous LLMs in cognitive reframing mainly converted negative emotions to positive ones, but these approaches have limited efficacy, often not promoting clients' self-discovery of alternative perspectives. In this paper, we unveil the Helping and Empowering through Adaptive Language in Mental Enhancement (HealMe) model. This novel cognitive reframing therapy method effectively addresses deep-rooted negative thoughts and fosters rational, balanced perspectives. Diverging from traditional LLM methods, HealMe employs empathetic dialogue based on psychotherapeutic frameworks. It systematically guides clients through distinguishing circumstances from feelings, brainstorming alternative viewpoints, and developing empathetic, actionable suggestions. Moreover, we adopt the first comprehensive and expertly crafted psychological evaluation metrics, specifically designed to rigorously assess the performance of cognitive reframing, in both AI-simulated dialogues and real-world therapeutic conversations. Experimental results show that our model outperforms others in terms of empathy, guidance, and logical coherence, demonstrating its effectiveness and potential positive impact on psychotherapy. △ Less

Submitted 29 July, 2024; v1 submitted 26 February, 2024; originally announced March 2024.

Comments: 19 pages, 4 figures

ACM Class: J.4

Journal ref: 2024 ACL main

Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context

Authors: Gemini Team, Petko Georgiev, Ving Ian Lei, Ryan Burnell, Libin Bai, Anmol Gulati, Garrett Tanzer, Damien Vincent, Zhufeng Pan, Shibo Wang, Soroosh Mariooryad, Yifan Ding, Xinyang Geng, Fred Alcober, Roy Frostig, Mark Omernick, Lexi Walker, Cosmin Paduraru, Christina Sorokin, Andrea Tacchetti, Colin Gaffney, Samira Daruki, Olcan Sercinoglu, Zach Gleicher, Juliette Love , et al. (1112 additional authors not shown)

Abstract: In this report, we introduce the Gemini 1.5 family of models, representing the next generation of highly compute-efficient multimodal models capable of recalling and reasoning over fine-grained information from millions of tokens of context, including multiple long documents and hours of video and audio. The family includes two new models: (1) an updated Gemini 1.5 Pro, which exceeds the February… ▽ More In this report, we introduce the Gemini 1.5 family of models, representing the next generation of highly compute-efficient multimodal models capable of recalling and reasoning over fine-grained information from millions of tokens of context, including multiple long documents and hours of video and audio. The family includes two new models: (1) an updated Gemini 1.5 Pro, which exceeds the February version on the great majority of capabilities and benchmarks; (2) Gemini 1.5 Flash, a more lightweight variant designed for efficiency with minimal regression in quality. Gemini 1.5 models achieve near-perfect recall on long-context retrieval tasks across modalities, improve the state-of-the-art in long-document QA, long-video QA and long-context ASR, and match or surpass Gemini 1.0 Ultra's state-of-the-art performance across a broad set of benchmarks. Studying the limits of Gemini 1.5's long-context ability, we find continued improvement in next-token prediction and near-perfect retrieval (>99%) up to at least 10M tokens, a generational leap over existing models such as Claude 3.0 (200k) and GPT-4 Turbo (128k). Finally, we highlight real-world use cases, such as Gemini 1.5 collaborating with professionals on completing their tasks achieving 26 to 75% time savings across 10 different job categories, as well as surprising new capabilities of large language models at the frontier; when given a grammar manual for Kalamang, a language with fewer than 200 speakers worldwide, the model learns to translate English to Kalamang at a similar level to a person who learned from the same content. △ Less

Submitted 16 December, 2024; v1 submitted 8 March, 2024; originally announced March 2024.

Ultralight vector dark matter search using data from the KAGRA O3GK run

Authors: The LIGO Scientific Collaboration, the Virgo Collaboration, the KAGRA Collaboration, A. G. Abac, R. Abbott, H. Abe, I. Abouelfettouh, F. Acernese, K. Ackley, C. Adamcewicz, S. Adhicary, N. Adhikari, R. X. Adhikari, V. K. Adkins, V. B. Adya, C. Affeldt, D. Agarwal, M. Agathos, O. D. Aguiar, I. Aguilar, L. Aiello, A. Ain, P. Ajith, T. Akutsu, S. Albanesi , et al. (1778 additional authors not shown)

Abstract: Among the various candidates for dark matter (DM), ultralight vector DM can be probed by laser interferometric gravitational wave detectors through the measurement of oscillating length changes in the arm cavities. In this context, KAGRA has a unique feature due to differing compositions of its mirrors, enhancing the signal of vector DM in the length change in the auxiliary channels. Here we prese… ▽ More Among the various candidates for dark matter (DM), ultralight vector DM can be probed by laser interferometric gravitational wave detectors through the measurement of oscillating length changes in the arm cavities. In this context, KAGRA has a unique feature due to differing compositions of its mirrors, enhancing the signal of vector DM in the length change in the auxiliary channels. Here we present the result of a search for $U(1)_{B-L}$ gauge boson DM using the KAGRA data from auxiliary length channels during the first joint observation run together with GEO600. By applying our search pipeline, which takes into account the stochastic nature of ultralight DM, upper bounds on the coupling strength between the $U(1)_{B-L}$ gauge boson and ordinary matter are obtained for a range of DM masses. While our constraints are less stringent than those derived from previous experiments, this study demonstrates the applicability of our method to the lower-mass vector DM search, which is made difficult in this measurement by the short observation time compared to the auto-correlation time scale of DM. △ Less

Submitted 5 March, 2024; originally announced March 2024.

Comments: 20 pages, 5 figures

Report number: LIGO-P2300250

Precise Extraction of Deep Learning Models via Side-Channel Attacks on Edge/Endpoint Devices

Authors: Younghan Lee, Sohee Jun, Yungi Cho, Woorim Han, Hyungon Moon, Yunheung Paek

Abstract: With growing popularity, deep learning (DL) models are becoming larger-scale, and only the companies with vast training datasets and immense computing power can manage their business serving such large models. Most of those DL models are proprietary to the companies who thus strive to keep their private models safe from the model extraction attack (MEA), whose aim is to steal the model by training… ▽ More With growing popularity, deep learning (DL) models are becoming larger-scale, and only the companies with vast training datasets and immense computing power can manage their business serving such large models. Most of those DL models are proprietary to the companies who thus strive to keep their private models safe from the model extraction attack (MEA), whose aim is to steal the model by training surrogate models. Nowadays, companies are inclined to offload the models from central servers to edge/endpoint devices. As revealed in the latest studies, adversaries exploit this opportunity as new attack vectors to launch side-channel attack (SCA) on the device running victim model and obtain various pieces of the model information, such as the model architecture (MA) and image dimension (ID). Our work provides a comprehensive understanding of such a relationship for the first time and would benefit future MEA studies in both offensive and defensive sides in that they may learn which pieces of information exposed by SCA are more important than the others. Our analysis additionally reveals that by grasping the victim model information from SCA, MEA can get highly effective and successful even without any prior knowledge of the model. Finally, to evince the practicality of our analysis results, we empirically apply SCA, and subsequently, carry out MEA under realistic threat assumptions. The results show up to 5.8 times better performance than when the adversary has no model information about the victim model. △ Less

Submitted 5 March, 2024; originally announced March 2024.

Comments: Accepted by 27th European Symposium on Research in Computer Security (ESORICS 2022)

FLGuard: Byzantine-Robust Federated Learning via Ensemble of Contrastive Models

Authors: Younghan Lee, Yungi Cho, Woorim Han, Ho Bae, Yunheung Paek

Abstract: Federated Learning (FL) thrives in training a global model with numerous clients by only sharing the parameters of their local models trained with their private training datasets. Therefore, without revealing the private dataset, the clients can obtain a deep learning (DL) model with high performance. However, recent research proposed poisoning attacks that cause a catastrophic loss in the accurac… ▽ More Federated Learning (FL) thrives in training a global model with numerous clients by only sharing the parameters of their local models trained with their private training datasets. Therefore, without revealing the private dataset, the clients can obtain a deep learning (DL) model with high performance. However, recent research proposed poisoning attacks that cause a catastrophic loss in the accuracy of the global model when adversaries, posed as benign clients, are present in a group of clients. Therefore, recent studies suggested byzantine-robust FL methods that allow the server to train an accurate global model even with the adversaries present in the system. However, many existing methods require the knowledge of the number of malicious clients or the auxiliary (clean) dataset or the effectiveness reportedly decreased hugely when the private dataset was non-independently and identically distributed (non-IID). In this work, we propose FLGuard, a novel byzantine-robust FL method that detects malicious clients and discards malicious local updates by utilizing the contrastive learning technique, which showed a tremendous improvement as a self-supervised learning method. With contrastive models, we design FLGuard as an ensemble scheme to maximize the defensive capability. We evaluate FLGuard extensively under various poisoning attacks and compare the accuracy of the global model with existing byzantine-robust FL methods. FLGuard outperforms the state-of-the-art defense methods in most cases and shows drastic improvement, especially in non-IID settings. https://github.com/201younghanlee/FLGuard △ Less

Submitted 5 March, 2024; originally announced March 2024.

Comments: Accepted by 28th European Symposium on Research in Computer Security (ESORICS 2023)

Recent Advances, Applications, and Open Challenges in Machine Learning for Health: Reflections from Research Roundtables at ML4H 2023 Symposium

Authors: Hyewon Jeong, Sarah Jabbour, Yuzhe Yang, Rahul Thapta, Hussein Mozannar, William Jongwon Han, Nikita Mehandru, Michael Wornow, Vladislav Lialin, Xin Liu, Alejandro Lozano, Jiacheng Zhu, Rafal Dariusz Kocielnik, Keith Harrigian, Haoran Zhang, Edward Lee, Milos Vukadinovic, Aparna Balagopalan, Vincent Jeanselme, Katherine Matton, Ilker Demirel, Jason Fries, Parisa Rashidi, Brett Beaulieu-Jones, Xuhai Orson Xu , et al. (18 additional authors not shown)

Abstract: The third ML4H symposium was held in person on December 10, 2023, in New Orleans, Louisiana, USA. The symposium included research roundtable sessions to foster discussions between participants and senior researchers on timely and relevant topics for the \ac{ML4H} community. Encouraged by the successful virtual roundtables in the previous year, we organized eleven in-person roundtables and four vir… ▽ More The third ML4H symposium was held in person on December 10, 2023, in New Orleans, Louisiana, USA. The symposium included research roundtable sessions to foster discussions between participants and senior researchers on timely and relevant topics for the \ac{ML4H} community. Encouraged by the successful virtual roundtables in the previous year, we organized eleven in-person roundtables and four virtual roundtables at ML4H 2022. The organization of the research roundtables at the conference involved 17 Senior Chairs and 19 Junior Chairs across 11 tables. Each roundtable session included invited senior chairs (with substantial experience in the field), junior chairs (responsible for facilitating the discussion), and attendees from diverse backgrounds with interest in the session's topic. Herein we detail the organization process and compile takeaways from these roundtable discussions, including recent advances, applications, and open challenges for each topic. We conclude with a summary and lessons learned across all roundtables. This document serves as a comprehensive review paper, summarizing the recent advancements in machine learning for healthcare as contributed by foremost researchers in the field. △ Less

Submitted 5 April, 2024; v1 submitted 3 March, 2024; originally announced March 2024.

Comments: ML4H 2023, Research Roundtables

EAGLE: Eigen Aggregation Learning for Object-Centric Unsupervised Semantic Segmentation

Authors: Chanyoung Kim, Woojung Han, Dayun Ju, Seong Jae Hwang

Abstract: Semantic segmentation has innately relied on extensive pixel-level annotated data, leading to the emergence of unsupervised methodologies. Among them, leveraging self-supervised Vision Transformers for unsupervised semantic segmentation (USS) has been making steady progress with expressive deep features. Yet, for semantically segmenting images with complex objects, a predominant challenge remains:… ▽ More Semantic segmentation has innately relied on extensive pixel-level annotated data, leading to the emergence of unsupervised methodologies. Among them, leveraging self-supervised Vision Transformers for unsupervised semantic segmentation (USS) has been making steady progress with expressive deep features. Yet, for semantically segmenting images with complex objects, a predominant challenge remains: the lack of explicit object-level semantic encoding in patch-level features. This technical limitation often leads to inadequate segmentation of complex objects with diverse structures. To address this gap, we present a novel approach, EAGLE, which emphasizes object-centric representation learning for unsupervised semantic segmentation. Specifically, we introduce EiCue, a spectral technique providing semantic and structural cues through an eigenbasis derived from the semantic similarity matrix of deep image features and color affinity from an image. Further, by incorporating our object-centric contrastive loss with EiCue, we guide our model to learn object-level representations with intra- and inter-image object-feature consistency, thereby enhancing semantic accuracy. Extensive experiments on COCO-Stuff, Cityscapes, and Potsdam-3 datasets demonstrate the state-of-the-art USS results of EAGLE with accurate and consistent semantic segmentation across complex scenes. △ Less

Submitted 5 April, 2024; v1 submitted 3 March, 2024; originally announced March 2024.

Exploring Advanced Methodologies in Security Evaluation for LLMs

Authors: Jun Huang, Jiawei Zhang, Qi Wang, Weihong Han, Yanchun Zhang

Abstract: Large Language Models (LLMs) represent an advanced evolution of earlier, simpler language models. They boast enhanced abilities to handle complex language patterns and generate coherent text, images, audios, and videos. Furthermore, they can be fine-tuned for specific tasks. This versatility has led to the proliferation and extensive use of numerous commercialized large models. However, the rapid… ▽ More Large Language Models (LLMs) represent an advanced evolution of earlier, simpler language models. They boast enhanced abilities to handle complex language patterns and generate coherent text, images, audios, and videos. Furthermore, they can be fine-tuned for specific tasks. This versatility has led to the proliferation and extensive use of numerous commercialized large models. However, the rapid expansion of LLMs has raised security and ethical concerns within the academic community. This emphasizes the need for ongoing research into security evaluation during their development and deployment. Over the past few years, a substantial body of research has been dedicated to the security evaluation of large-scale models. This article an in-depth review of the most recent advancements in this field, providing a comprehensive analysis of commonly used evaluation metrics, advanced evaluation frameworks, and the routine evaluation processes for LLMs. Furthermore, we also discuss the future directions for advancing the security evaluation of LLMs. △ Less

Submitted 28 February, 2024; v1 submitted 27 February, 2024; originally announced February 2024.

Towards Robust Instruction Tuning on Multimodal Large Language Models

Authors: Wei Han, Hui Chen, Soujanya Poria

Abstract: Fine-tuning large language models (LLMs) on multi-task instruction-following data has been proven to be a powerful learning paradigm for improving their zero-shot capabilities on new tasks. Recent works about high-quality instruction-following data generation and selection require amounts of human labor to conceive model-understandable instructions for the given tasks and carefully filter the LLM-… ▽ More Fine-tuning large language models (LLMs) on multi-task instruction-following data has been proven to be a powerful learning paradigm for improving their zero-shot capabilities on new tasks. Recent works about high-quality instruction-following data generation and selection require amounts of human labor to conceive model-understandable instructions for the given tasks and carefully filter the LLM-generated data. In this work, we introduce an automatic instruction augmentation method named INSTRAUG in multimodal tasks. It starts from a handful of basic and straightforward meta instructions but can expand an instruction-following dataset by 30 times. Results on two popular multimodal instructionfollowing benchmarks MULTIINSTRUCT and InstructBLIP show that INSTRAUG can significantly improve the alignment of multimodal large language models (MLLMs) across 12 multimodal tasks, which is even equivalent to the benefits of scaling up training data multiple times. △ Less

Submitted 14 June, 2024; v1 submitted 22 February, 2024; originally announced February 2024.

Comments: 24 pages, 7 figures

FinBen: A Holistic Financial Benchmark for Large Language Models

Authors: Qianqian Xie, Weiguang Han, Zhengyu Chen, Ruoyu Xiang, Xiao Zhang, Yueru He, Mengxi Xiao, Dong Li, Yongfu Dai, Duanyu Feng, Yijing Xu, Haoqiang Kang, Ziyan Kuang, Chenhan Yuan, Kailai Yang, Zheheng Luo, Tianlin Zhang, Zhiwei Liu, Guojun Xiong, Zhiyang Deng, Yuechen Jiang, Zhiyuan Yao, Haohang Li, Yangyang Yu, Gang Hu , et al. (9 additional authors not shown)

Abstract: LLMs have transformed NLP and shown promise in various fields, yet their potential in finance is underexplored due to a lack of comprehensive evaluation benchmarks, the rapid development of LLMs, and the complexity of financial tasks. In this paper, we introduce FinBen, the first extensive open-source evaluation benchmark, including 36 datasets spanning 24 financial tasks, covering seven critical… ▽ More LLMs have transformed NLP and shown promise in various fields, yet their potential in finance is underexplored due to a lack of comprehensive evaluation benchmarks, the rapid development of LLMs, and the complexity of financial tasks. In this paper, we introduce FinBen, the first extensive open-source evaluation benchmark, including 36 datasets spanning 24 financial tasks, covering seven critical aspects: information extraction (IE), textual analysis, question answering (QA), text generation, risk management, forecasting, and decision-making. FinBen offers several key innovations: a broader range of tasks and datasets, the first evaluation of stock trading, novel agent and Retrieval-Augmented Generation (RAG) evaluation, and three novel open-source evaluation datasets for text summarization, question answering, and stock trading. Our evaluation of 15 representative LLMs, including GPT-4, ChatGPT, and the latest Gemini, reveals several key findings: While LLMs excel in IE and textual analysis, they struggle with advanced reasoning and complex tasks like text generation and forecasting. GPT-4 excels in IE and stock trading, while Gemini is better at text generation and forecasting. Instruction-tuned LLMs improve textual analysis but offer limited benefits for complex tasks such as QA. FinBen has been used to host the first financial LLMs shared task at the FinNLP-AgentScen workshop during IJCAI-2024, attracting 12 teams. Their novel solutions outperformed GPT-4, showcasing FinBen's potential to drive innovation in financial LLMs. All datasets, results, and codes are released for the research community: https://github.com/The-FinAI/PIXIU. △ Less

Submitted 18 June, 2024; v1 submitted 19 February, 2024; originally announced February 2024.

Comments: 26 pages, 11 figures

Search for the production of deuterons and antideuterons in e^+e^- annihilation at center-of-mass energies between 4.13 and 4.70 GeV

Authors: BESIII Collaboration, M. Ablikim, M. N. Achasov, P. Adlarson, O. Afedulidis, R. Aliberti, A. Amoroso, M. R. An, Q. An, Y. Bai, O. Bakina, I. Balossino, Y. Ban, V. Batozskaya, K. Begzsuren, N. Berger, M. Berlowski, M. Bertani, D. Bettoni, F. Bianchi, E. Bianco, A. Bortone, I. Boyko, R. A. Briere, A. Brueggemann , et al. (593 additional authors not shown)

Abstract: Using a data sample of $e^+e^-$ collision data corresponding to an integrated luminosity of 19 fb$^{-1}$ collected with the BESIII detector at the BEPCII collider, we search for the production of deuterons and antideuterons via $e^+e^-\to ppπ^-\bar{d}+c.c.$ for the first time at center-of-mass energies between 4.13 and 4.70 GeV. No significant signal is observed and the upper limit of the… ▽ More Using a data sample of $e^+e^-$ collision data corresponding to an integrated luminosity of 19 fb$^{-1}$ collected with the BESIII detector at the BEPCII collider, we search for the production of deuterons and antideuterons via $e^+e^-\to ppπ^-\bar{d}+c.c.$ for the first time at center-of-mass energies between 4.13 and 4.70 GeV. No significant signal is observed and the upper limit of the $e^+e^-\to ppπ^-\bar{d}+c.c.$ cross section is determined to be from 9.0 to 145 fb depending on the center-of-mass energy at the $90\%$ confidence level. △ Less

Submitted 17 February, 2024; originally announced February 2024.

Dólares or Dollars? Unraveling the Bilingual Prowess of Financial LLMs Between Spanish and English

Authors: Xiao Zhang, Ruoyu Xiang, Chenhan Yuan, Duanyu Feng, Weiguang Han, Alejandro Lopez-Lira, Xiao-Yang Liu, Sophia Ananiadou, Min Peng, Jimin Huang, Qianqian Xie

Abstract: Despite Spanish's pivotal role in the global finance industry, a pronounced gap exists in Spanish financial natural language processing (NLP) and application studies compared to English, especially in the era of large language models (LLMs). To bridge this gap, we unveil Toisón de Oro, the first bilingual framework that establishes instruction datasets, finetuned LLMs, and evaluation benchmark for… ▽ More Despite Spanish's pivotal role in the global finance industry, a pronounced gap exists in Spanish financial natural language processing (NLP) and application studies compared to English, especially in the era of large language models (LLMs). To bridge this gap, we unveil Toisón de Oro, the first bilingual framework that establishes instruction datasets, finetuned LLMs, and evaluation benchmark for financial LLMs in Spanish joint with English. We construct a rigorously curated bilingual instruction dataset including over 144K Spanish and English samples from 15 datasets covering 7 tasks. Harnessing this, we introduce FinMA-ES, an LLM designed for bilingual financial applications. We evaluate our model and existing LLMs using FLARE-ES, the first comprehensive bilingual evaluation benchmark with 21 datasets covering 9 tasks. The FLARE-ES benchmark results reveal a significant multilingual performance gap and bias in existing LLMs. FinMA-ES models surpass SOTA LLMs such as GPT-4 in Spanish financial tasks, due to strategic instruction tuning and leveraging data from diverse linguistic resources, highlighting the positive impact of cross-linguistic transfer. All our datasets, models, and benchmarks have been released. △ Less

Submitted 11 February, 2024; originally announced February 2024.

Comments: 10 pages, 2 figures

VisLingInstruct: Elevating Zero-Shot Learning in Multi-Modal Language Models with Autonomous Instruction Optimization

Authors: Dongsheng Zhu, Xunzhu Tang, Weidong Han, Jinghui Lu, Yukun Zhao, Guoliang Xing, Junfeng Wang, Dawei Yin

Abstract: This paper presents VisLingInstruct, a novel approach to advancing Multi-Modal Language Models (MMLMs) in zero-shot learning. Current MMLMs show impressive zero-shot abilities in multi-modal tasks, but their performance depends heavily on the quality of instructions. VisLingInstruct tackles this by autonomously evaluating and optimizing instructional texts through In-Context Learning, improving th… ▽ More This paper presents VisLingInstruct, a novel approach to advancing Multi-Modal Language Models (MMLMs) in zero-shot learning. Current MMLMs show impressive zero-shot abilities in multi-modal tasks, but their performance depends heavily on the quality of instructions. VisLingInstruct tackles this by autonomously evaluating and optimizing instructional texts through In-Context Learning, improving the synergy between visual perception and linguistic expression in MMLMs. Alongside this instructional advancement, we have also optimized the visual feature extraction modules in MMLMs, further augmenting their responsiveness to textual content. Our comprehensive experiments on MMLMs, based on FlanT5 and Vicuna, show that VisLingInstruct significantly improves zero-shot performance in visual multi-modal tasks. Notably, it achieves a 13.1% and 9% increase in accuracy over the prior state-of-the-art on the TextVQA and HatefulMemes datasets. Our main code is available at https://github.com/Zhudongsheng75/VisLingInstruct. △ Less

Submitted 20 June, 2024; v1 submitted 11 February, 2024; originally announced February 2024.

Comments: Accepted to NAACL2024 main conference

TransGPT: Multi-modal Generative Pre-trained Transformer for Transportation

Authors: Peng Wang, Xiang Wei, Fangxu Hu, Wenjuan Han

Abstract: Natural language processing (NLP) is a key component of intelligent transportation systems (ITS), but it faces many challenges in the transportation domain, such as domain-specific knowledge and data, and multi-modal inputs and outputs. This paper presents TransGPT, a novel (multi-modal) large language model for the transportation domain, which consists of two independent variants: TransGPT-SM for… ▽ More Natural language processing (NLP) is a key component of intelligent transportation systems (ITS), but it faces many challenges in the transportation domain, such as domain-specific knowledge and data, and multi-modal inputs and outputs. This paper presents TransGPT, a novel (multi-modal) large language model for the transportation domain, which consists of two independent variants: TransGPT-SM for single-modal data and TransGPT-MM for multi-modal data. TransGPT-SM is finetuned on a single-modal Transportation dataset (STD) that contains textual data from various sources in the transportation domain. TransGPT-MM is finetuned on a multi-modal Transportation dataset (MTD) that we manually collected from three areas of the transportation domain: driving tests, traffic signs, and landmarks. We evaluate TransGPT on several benchmark datasets for different tasks in the transportation domain, and show that it outperforms baseline models on most tasks. We also showcase the potential applications of TransGPT for traffic analysis and modeling, such as generating synthetic traffic scenarios, explaining traffic phenomena, answering traffic-related questions, providing traffic recommendations, and generating traffic reports. This work advances the state-of-the-art of NLP in the transportation domain and provides a useful tool for ITS researchers and practitioners. △ Less

Submitted 11 February, 2024; originally announced February 2024.