"VideoRAG: Retrieval-Augmented Generation with Extreme Long-Context Videos"

[CVPR 2024] MovieChat: From Dense Token to Sparse Memory for Long Video Understanding

Official code for Goldfish model for long video understanding and MiniGPT4-video for short video understanding

🔥🔥MLVU: Multi-task Long Video Understanding Benchmark

This is the official implementation of our paper "Video-RAG: Visually-aligned Retrieval-Augmented Long Video Comprehension"

Multi-granularity Correspondence Learning from Long-term Noisy Videos [ICLR 2024, Oral]

[EMNLP 2023] TESTA: Temporal-Spatial Token Aggregation for Long-form Video-Language Understanding

LongVALE: Vision-Audio-Language-Event Benchmark Towards Time-Aware Omni-Modal Perception of Long Videos. (CVPR 2025))

[ICLR 2025] TimeSuite: Improving MLLMs for Long Video Understanding via Grounded Tuning

Language Repository for Long Video Understanding

Winner solution to Generic Event Boundary Captioning task in LOVEU Challenge (CVPR 2023 workshop)

Official implementation for paper Learning Grounded Vision-Language Representation for Versatile Understanding in Untrimmed Videos

This is the official implementation of ReVisionLLM: Recursive Vision-Language Model for Temporal Grounding in Hour-Long Videos

Official Implementation (Pytorch) of the "VidChain: Chain-of-Tasks with Metric-based Direct Preference Optimization for Dense Video Captioning", AAAI 2025

A benchmark that focuses on the sampling dilemma in long-video tasks. Through well-designed tasks, it evaluates the sampling efficiency of long-video VLMs.

Official Repo for CVPR 2025 Paper -- DeCafNet: Delegate and Conquer for Efficient Temporal Grounding in Long Videos

HoPE: Hybrid of Position Embedding for Length Generalization in Vision-Language Models