Skip to content

@thu-ml TSAIL group

Change the repository type filter

All

    Repositories list

    72 repositories

    • RIFLEx

      Public
      Official implementation for "RIFLEx: A Free Lunch for Length Extrapolation in Video Diffusion Transformers" (ICML 2025)
      Python
      •
      Apache License 2.0
      •63•632•17•0•Updated May 1, 2025May 1, 2025

    • SpargeAttn

      Public
      SpargeAttention: A training-free sparse attention that can accelerate any model inference.
      attentionquantizationmlsysinference-accelerationai-infravision-transformersparse-attentionllmsageattention
      Cuda
      •
      Apache License 2.0
      •31•504•18•3•Updated Apr 29, 2025Apr 29, 2025

    • SageAttention

      Public
      Quantized Attention achieves speedup of 2-3x and 3-5x compared to FlashAttention and xformers, without lossing end-to-end metrics across language, image, and video models.
      cudatritonattentionvitquantizationvideo-generationmlsysinference-accelerationefficient-attentionllm
      Cuda
      •
      Apache License 2.0
      •99•1.4k•42•5•Updated Apr 28, 2025Apr 28, 2025

    • cond-image-leakage

      Public
      Official implementation for "Identifying and Solving Conditional Image Leakage in Image-to-Video Diffusion Model" (NeurIPS 2024)
      Apache License 2.0
      •22•250•8•0•Updated Apr 25, 2025Apr 25, 2025

    • tianshou

      Public
      An elegant PyTorch deep reinforcement learning library.
      pytorchdqnpolicy-gradientrlcqlatariddpgimitation-learningsacdrl
      Python
      •
      MIT License
      •1.2k•8.4k•147•5•Updated Apr 23, 2025Apr 23, 2025

    • RoboticsDiffusionTransformer

      Public
      RDT-1B: a Diffusion Foundation Model for Bimanual Manipulation
      Python
      •
      MIT License
      •116•1.1k•26•3•Updated Apr 4, 2025Apr 4, 2025

    • MMTrustEval

      Public
      A toolbox for benchmarking trustworthiness of multimodal large language models (MultiTrust, NeurIPS 2024 Track Datasets and Benchmarks)
      benchmarkprivacytoolboxsafetymulti-modalfairnessrobustnessclaudegpt-4trustworthy-ai
      Python
      •
      Creative Commons Attribution Share Alike 4.0 International
      •10•145•1•0•Updated Mar 26, 2025Mar 26, 2025

    • TetraJet-MXFP4Training

      Public
      Pytorch implementation of "Oscillation-Reduced MXFP4 Training for Vision Transformers" on DeiT Model Pre-training
      Python
      •
      Apache License 2.0
      •1•10•0•0•Updated Mar 22, 2025Mar 22, 2025

    • DiffusionBridge

      Public
      Official codebase for "Diffusion Bridge Implicit Models" (ICLR 2025) and "Consistency Diffusion Bridge Models" (NeurIPS 2024)
      Python
      •4•45•2•0•Updated Mar 10, 2025Mar 10, 2025

    • GFT

      Public
      Python
      •
      MIT License
      •0•30•3•0•Updated Mar 8, 2025Mar 8, 2025

    • i-DODE

      Public
      Official code for "Improved Techniques for Maximum Likelihood Estimation for Diffusion ODEs" (ICML 2023)
      Python
      •
      Apache License 2.0
      •1•18•1•0•Updated Mar 4, 2025Mar 4, 2025

    • STAIR

      Public
      Official codebase for "STAIR: Improving Safety Alignment with Introspective Reasoning"
      Python
      •
      MIT License
      •0•34•0•0•Updated Feb 26, 2025Feb 26, 2025

    • EffWRN-paddle

      Public
      Python
      •
      MIT License
      •0•0•0•0•Updated Feb 24, 2025Feb 24, 2025

    • RIFLEx.github.io

      Public
      JavaScript
      •0•0•0•0•Updated Feb 21, 2025Feb 21, 2025

    • oddefense

      Public
      Jupyter Notebook
      •
      MIT License
      •0•1•0•0•Updated Feb 14, 2025Feb 14, 2025

    • CCA

      Public
      Codes accompanying the paper "Toward Guidance-Free AR Visual Generation via Condition Contrastive Alignment"
      Python
      •
      MIT License
      •0•31•0•0•Updated Feb 11, 2025Feb 11, 2025

    • Adaptive-Sparse-Trainer

      Public
      Official implementation for "Pruning Large Language Models with Semi-Structural Adaptive Sparse Training" (AAAI 2025)
      Python
      •
      Apache License 2.0
      •1•10•0•0•Updated Dec 22, 2024Dec 22, 2024

    • ReMoE

      Public
      [ICLR2025] Codebase for "ReMoE: Fully Differentiable Mixture-of-Experts with ReLU Routing", built on Megatron-LM.
      Python
      •
      Other
      •1•70•0•0•Updated Dec 20, 2024Dec 20, 2024

    • tianshou-docs-zh_CN

      Public
      天授中文文档
      TeX
      •10•58•0•0•Updated Dec 10, 2024Dec 10, 2024

    • 2by4-pretrain-acc-examples

      Public
      Code for "Accelerating Transformer Pre-training with 2:4 Sparsity"
      Jupyter Notebook
      •
      BSD 3-Clause "New" or "Revised" License
      •0•23•0•0•Updated Dec 8, 2024Dec 8, 2024

    • CRM

      Public
      [ECCV 2024] Single Image to 3D Textured Mesh in 10 seconds with Convolutional Reconstruction Model.
      generative-modelreconstruction3dmultiviewdiffusion-modelsaigc
      Python
      •
      MIT License
      •54•641•24•1•Updated Nov 28, 2024Nov 28, 2024

    • HiDe-PET

      Public
      IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
      Python
      •
      MIT License
      •2•4•0•0•Updated Nov 15, 2024Nov 15, 2024

    • HiDe-Prompt

      Public
      Hierarchical Decomposition of Prompt-Based Continual Learning: Rethinking Obscured Sub-optimality (NeurIPS 2023, Spotlight)
      Python
      •
      MIT License
      •9•82•3•0•Updated Nov 15, 2024Nov 15, 2024

    • Noise-Contrastive-Alignment

      Public
      Code accompanying the paper "Noise Contrastive Alignment of Language Models with Explicit Rewards" (NeurIPS 2024)
      Python
      •
      MIT License
      •5•51•2•0•Updated Nov 8, 2024Nov 8, 2024

    • Efficient-Diffusion-Alignment

      Public
      Official Codebase for "Aligning Diffusion Behaviors with Q-functions for Efficient Continuous Control" (NeurIPS 2024)
      Python
      •0•13•1•0•Updated Oct 29, 2024Oct 29, 2024

    • CEURL

      Public
      Official implementation for "PEAC: Unsupervised Pre-training for Cross-Embodiment Reinforcement Learning" (NeurIPS 2024)
      reinforcement-learningunsupervised-reinforcement-learningembodied-aicross-embodiment-reinforcement-learningneurips24
      Python
      •
      MIT License
      •0•13•0•0•Updated Oct 13, 2024Oct 13, 2024

    • 3D_Corruptions_AD

      Public
      Benchmarking Robustness of 3D Object Detection to Common Corruptions in Autonomous Driving, CVPR 2023
      point-cloudself-driving-carautonomous-vehiclescvpr2023
      Python
      •
      MIT License
      •16•132•7•0•Updated Oct 9, 2024Oct 9, 2024

    • Jetfire-INT8Training

      Public
      Jupyter Notebook
      •5•40•2•0•Updated Jul 21, 2024Jul 21, 2024

    • Bridge-TTS

      Public
      Official codebase for "Schrodinger Bridges Beat Diffusion Models on Text-to-Speech Synthesis" (https://arxiv.org/abs/2312.03491).
      MIT License
      •1•127•4•0•Updated Jul 12, 2024Jul 12, 2024

    • conditional-image-leakage

      Public
      Official implementation for "Identifying and Solving Conditional Image Leakage in Image-to-Video Diffusion Model"
      Apache License 2.0
      •0•1•0•0•Updated Jun 23, 2024Jun 23, 2024