Skip to content

@CASIA-IVA-Lab CASIA-IVA-Lab

Change the repository type filter

All

    Repositories list

    21 repositories

    • M3VQA

      Public
      Python
      •0•0•0•0•Updated Apr 20, 2026Apr 20, 2026

    • SciVQR

      Public
      This repo holds the official evaluation code for "SciVQR: A Multidisciplinary Multimodal Benchmark for Advanced Scientific Reasoning Evaluation".
      Python
      •
      MIT License
      •0•0•0•0•Updated Apr 18, 2026Apr 18, 2026

    • VisReason

      Public
      0•0•0•0•Updated Apr 16, 2026Apr 16, 2026

    • ThinkStream

      Public
      Python
      •
      MIT License
      •1•28•0•0•Updated Apr 9, 2026Apr 9, 2026

    • DeepSlyme

      Public
      Python
      •
      Apache License 2.0
      •0•0•0•0•Updated Apr 7, 2026Apr 7, 2026

    • UrbanNav

      Public
      [AAAI 2026] Official implementation of paper "UrbanNav: Learning Language-Guided Embodied Urban Navigation from Web-Scale Human Trajectories"
      Python
      •
      MIT License
      •4•60•3•0•Updated Mar 27, 2026Mar 27, 2026

    • S1-MMAlign

      Public
      S1-MMAlign: 科学多模态数据集（入口页，数据托管于Hugging Face）
      0•0•0•0•Updated Mar 19, 2026Mar 19, 2026

    • ChatSearch

      Public
      ChatSearch: a Dataset and a Generative Retrieval Model for General Conversational Image Retrieval
      Python
      •
      Apache License 2.0
      •0•9•1•0•Updated Jan 6, 2026Jan 6, 2026

    • VRoPE

      Public
      [EMNLP 2025 Main] Official implementation of VRoPE: Rotary Position Embedding for Video Large Language Models.
      Python
      •1•27•0•0•Updated Nov 18, 2025Nov 18, 2025

    • PrefixGrouper

      Public
      An efficient GRPO training util.
      Python
      •
      MIT License
      •3•55•0•0•Updated Jun 13, 2025Jun 13, 2025

    • VideoNIAH

      Public
      VideoNIAH: A Flexible Synthetic Method for Benchmarking Video MLLMs
      Python
      •0•57•4•0•Updated Mar 9, 2025Mar 9, 2025

    • COSA

      Public
      [ICLR2024] Codes and Models for COSA: Concatenated Sample Pretrained Vision-Language Foundation Model
      video-captioningvideo-qavideo-retrieval
      video-captioningvideo-qavideo-retrievalvision-language-pretrainingvideo-language-pretrainng
      Python
      •
      MIT License
      •3•43•3•0•Updated Dec 25, 2024Dec 25, 2024

    • VALOR

      Public
      [TPAMI2024] Codes and Models for VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset
      vision-language-pretrainingaudio-language-pretrainingaudiovisual-language-pretraining
      vision-language-pretrainingaudio-language-pretrainingaudiovisual-language-pretrainingmultimodal-representation-learning
      Python
      •
      MIT License
      •18•310•8•0•Updated Dec 25, 2024Dec 25, 2024

    • DANet

      Public
      Dual Attention Network for Scene Segmentation (CVPR2019)
      Python
      •
      MIT License
      •485•2.5k•61•1•Updated Dec 23, 2024Dec 23, 2024

    • MRES

      Public
      This repo holds the official code and data for "Unveiling Parts Beyond Objects: Towards Finer-Granularity Referring Expression Segmentation", accepted by CVPR 2…
      Apache License 2.0
      •0•73•5•0•Updated Jun 3, 2024Jun 3, 2024

    • SC-Tune

      Public
      Official code for CVPR 2024 paper, "SC-Tune: Unleashing Self-Consistent Referential Comprehension in Large Vision Language Models"
      Python
      •
      MIT License
      •1•16•1•0•Updated Apr 22, 2024Apr 22, 2024

    • VAST

      Public
      [NIPS2023] Code and Model for VAST: A Vision-Audio-Subtitle-Text Omni-Modality Foundation Model and Dataset
      datasetvision-languageaudio-language
      datasetvision-languageaudio-languagemultimodal-foundation-modelcross-modality-pretrainingvision-audio-subtitle-text
      Jupyter Notebook
      •
      MIT License
      •18•301•22•0•Updated Mar 14, 2024Mar 14, 2024

    • GLOBER

      Public
      Python
      •
      Other
      •0•9•1•0•Updated Jan 11, 2024Jan 11, 2024

    • ChatBridge

      Public
      ChatBridge, an approach to learning a unified multimodal model to interpret, correlate, and reason about various modalities without relying on all combinations…
      Python
      •
      BSD 3-Clause "New" or "Revised" License
      •1•55•6•0•Updated Sep 4, 2023Sep 4, 2023

    • OPT_Questioner

      Public
      Official PyTorch implementation of the paper "Enhancing Vision-Language Pre-Training with Jointly Learned Questioner and Dense Captioner"
      Python
      •
      MIT License
      •1•15•1•0•Updated Aug 9, 2023Aug 9, 2023

    • MOSO

      Public
      Python
      •2•35•5•0•Updated Jun 6, 2023Jun 6, 2023
    ProTip! When viewing an organization's repositories, you can use the props. filter to filter by custom property.