PKU-YUAN-Lab (袁粒课题组-北大深研院) ORG

This project aim to reproduce Sora (Open AI T2V model), we wish the open source community contribute to this project.

ChatLaw：A Powerful LLM Tailored for Chinese Legal. 中文法律大模型

【EMNLP 2024🔥】Video-LLaVA: Learning United Visual Representation by Alignment Before Projection

【TMM 2025🔥】 Mixture-of-Experts for Large Vision-Language Models

[ICCV 2025] LLaVA-CoT, a visual language model capable of spontaneous, systematic reasoning

Helios: Real Real-Time Long Video Generation Model

[TPAMI 2025🔥] MagicTime: Time-lapse Video Generation Models as Metamorphic Simulators

[CVPR 2024 Highlight🔥] Chat-UniVi: Unified Visual Representation Empowers Large Language Models with Image and Video Understanding

UniWorld: High-Resolution Semantic Encoders for Unified Visual Understanding and Generation

【ICLR 2024🔥】 Extending Video-Language Pretraining to N-modality by Language-based Semantic Alignment

[CVPR 2025 Highlight🔥] Identity-Preserving Text-to-Video Generation by Frequency Decomposition

High-Fidelity 3D Shape Generation via Scalable Geometric Refinement

An MBTI Exploration of Large Language Models

[NeurIPS 2025 D&B🔥] ImgEdit: A Unified Image Editing Dataset and Benchmark

Edit-R1: Reinforce Image Editing with Diffusion Negative-Aware Finetuning and MLLM Implicit Feedback

Official implementation of Repaint123: Fast and High-quality One Image to 3D Generation with Progressive Controllable 2D Repainting (ECCV 2024)

9.9 元豆包API复刻 Claude Science

[NeurIPS 2025 D&B🔥] OpenS2V-Nexus: A Detailed Benchmark and Million-Scale Dataset for Subject-to-Video Generation

[AAAI 2025🔥] Official implementation of Cycle3D: High-quality and Consistent Image-to-3D Generation via Generation-Reconstruction Cycle

[ICML 2026🔥] WISE: A World Knowledge-Informed Semantic Evaluation for Text-to-Image Generation

[NeurIPS 2024 D&B Spotlight🔥] ChronoMagic-Bench: A Benchmark for Metamorphic Evaluation of Text-to-Time-lapse Video Generation

A Protein Large Language Model for Multi-Task Protein Language Processing

[CVPR 2025🔥] Enhancing Video VAE by Wavelet-Driven Energy Flow for Latent Video Diffusion Model

[AAAI 2026 🔥] Official implementation of "NeuralGS: Bridging Neural Fields and 3D Gaussian Splatting for Compact 3D Representation"

Official repository for the UAE paper, unified-GRPO, and unified-Bench

Attack to induce LLMs within hallucinations

[ACM MM 2025] HoloTime: Taming Video Diffusion Models for Panoramic 4D Scene Generation

A Comprehensive Benchmark and Toolkit for Evaluating Video-based Large Language Models!

No description.

[AAAI26] Next Patch Prediction

Envision3D: One Image to 3D with Anchor Views Interpolation

No description.

Official implementation of UniWorld-View

This repository is the official implementation of "Look-Back: Implicit Visual Re-focusing in MLLM Reasoning".

No description.

An official implementation of SwapAnyone.

The official code for "TaxDiff: Taxonomic-Guided Diffusion Model for Protein Sequence Generation"

No description.

Turning the TIDE: Cross-Architecture Distillation for Diffusion Large Language Models

OSP-Next

Does Understanding Inform Generation in Unified Multimodal Models? From Analysis to Path Forward

No description.

No description.

GPT as a Monte Carlo Language Tree: A Probabilistic Perspective

【COLING 2025🔥】Code for the paper "Is Parameter Collision Hindering Continual Learning in LLMs?".

Code for the paper "AsFT: Anchoring Safety During LLM Fune-Tuning Within Narrow Safety Basin".

[ICLR'25] PiCO: Peer Review in LLMs based on the Consistency Optimization, https://arxiv.org/pdf/2402.01830

An official implementation of FlashI2V.

Helios: Real Real-Time Long Video Generation Model

[NeurIPS 2025 D&B🔥] Implementation of "GS2E: Gaussian Splatting is an Effective Data Generator for Event Stream Generation"

LLMBind: A Unified Modality-Task Integration Framework

Spark-Wan: Few-Step Video Diffusion Transformer via Hybrid Adversarial Post-Training

DeblurNVS: Geometric Latent Diffusion for Novel View Synthesis from Motion-Blurred Images

A living arena for Human vs Agent and Agent vs Agent competition, testing OpenClaw intelligence through real-time gameplay.

No description.

Turning the TIDE: Cross-Architecture Distillation for Diffusion Large Language Models

No description.