Benchmarking plasticity optimization in deep reinforcement learning.

My GitHub profile.

YOLO-MS: Rethinking Multi-Scale Representation Learning for Real-Time Object Detection

Janus-Series: Unified Multimodal Understanding and Generation Models

Rethinking Step-by-step Visual Reasoning in LLMs

[CVPR2025]UniScene: Unified Occupancy-centric Driving Scene Generation

DIAMOND (DIffusion As a Model Of eNvironment Dreams) is a reinforcement learning agent trained in a diffusion world model. NeurIPS 2024 Spotlight.

[NeurIPS 2024] NeuMA: Neural Material Adaptor for Visual Grounding of Intrinsic Dynamics

[ICLR 2024] Adaptive Replay Ratio implementation from 'Revisiting Plasticity in Visual RL: Data, Modules and Training Stages'.

Original reference implementation of "3D Gaussian Splatting for Real-Time Radiance Field Rendering"

[ECCV 2024, Oral] DynamiCrafter: Animating Open-domain Images with Video Diffusion Priors

Code for the ICLR 2024 spotlight paper: "Learning to Act without Actions" (introducing Latent Action Policies)

[ICLR 2024] Latent Intuitive Physics: Learning to Transfer Hidden Physics from a 3D Video

An index of algorithms for offline reinforcement learning (offline-rl)

Awesome Papers related to Mamba.

本人的科研经验

SJTU Canvas 视频 (批量) 下载器

A curated list of awesome model based RL resources (continually updated)

This project aim to reproducing Sora (Open AI T2V model), but we only have limited resource. We deeply wish the all open source community can contribute to this project.

Pre-Training Goal-based Models for Sample-Efficient Reinforcement Learning