Hugging Face Trending Papers

VideoMM: Adaptive Macro-Micro Inference for Efficient Video MLLMs

arXiv Computer Vision
Nov 9

AdaAlloc: Adaptive Visual Token Allocation for Long-Video Question Answering

AdaAlloc is a training‑free inference framework that adaptively allocates visual tokens between a global video overview and localized visual detail for long‑video question answering. It first plans a global/local allocation policy, uses temporal grounding to locate candidate evidence segments, and refines them into a compact temporal memory. The selected policy controls both frame sampling density and resolution, combining sparse, lower‑resolution global frames with denser, higher‑resolution local frames, and achieves up to 10.5 percentage‑point improvements over budget‑matched uniform sampling on several benchmarks.

By Joungbin An, Kristen Grauman
arXiv Computation and Language
Sep 3

ShallowStream: Index Shallow then Answer Deep for Streaming Video Understanding

ShallowStream is a framework for streaming video understanding that uses the shallow layers of a multimodal large language model (MLLM) to encode frames and build a lightweight index. During streaming, it maintains an always‑on index via the KV cache of shallow layers, and at query time it scores context frames using shallow‑layer attention and selects diverse evidence for answering. The approach matches the performance of leading streaming methods while cutting per‑frame prefill latency and 10‑second end‑to‑end latency by up to 52.1× and 11.9×, respectively.

By Jitai Hao, Ke Yang, Qiang Huang, Jun Yu
arXiv AI
Jul 29

Reasoning with Memory: A Temporal Granularity-Adaptive Framework for Training-Free Long Video Understanding

arXiv:2607. 24794v1 Announce Type: new Abstract: While Multimodal Large Language Models (MLLMs) demonstrate superior generalization in fundamental video tasks, restricted context windows limit their long video understanding.

By Linghao Meng, Qiankun Li, Junyuan Mao, Pujin Liao, Zhicheng He, Enbo Zhang, Kun Wang, Yang Liu, Huazhu Fu, Yueming Jin
arXiv Computer Vision
Oct 28

Diagnosing and Correcting Bias in MLLM for Long Video Understanding

The paper introduces S‑MME, a benchmark for diagnosing shortcut‑biased behavior in Multimodal Large Language Models (MLLMs) on long‑video question answering. It proposes CLEAR, a training‑free framework that estimates and mitigates bias by treating the full video‑question pair as a complete view and intervening on hidden states during inference. Experiments on five benchmarks demonstrate that CLEAR improves prediction accuracy and robustness across different MLLMs.

By Xusheng Liang, Jianqiao Sun, Hao Zhang, Yulei Niu, Hengshuang Zhao, Jiawei Ma
arXiv Computer Vision
Oct 19

LDDR: Linear-DPP Dynamic-Resolution Frame Sampling for Video MLLMs

LDDR (Linear DPP Dynamic Resolution) is a training‑free, budget‑aware video frame sampling framework designed for multimodal large language models (MLLMs). It uses query‑aware Determinantal Point Process (DPP) selection in a task‑conditioned feature space, achieving a 3× runtime speedup over standard DPP baselines. By assigning a non‑uniform token budget—allocating more tokens to informative, non‑redundant frames and downscaling or pruning less useful ones—LDDR improves performance by 2.5 points in low‑budget settings and 1.6 points in high‑budget scenarios across four video benchmarks and multiple MLLM families.

By Jingfeng Chen, Jiawen Qian, Wendi Deng, Yinuo Guo, Jiaqi Yu, Sicong Leng, Raghuveer Thirukovalluru, Bhuwan Dhingra
arXiv AI
Aug 28

PACE: A Unified Condense-and-Extract Paradigm for Fast VLM Inference

PACE introduces a training‑free Condense‑and‑Extract framework that speeds up Vision‑Language Model inference by first adaptively downsampling visual inputs before encoding and then selectively retaining essential tokens during decoding. The Adaptive Pixel Compressor (APC) reduces encoder workload while preserving global context, and the Dynamic Dual‑Attention Extractor (DDAE) keeps task‑critical details by fusing visual and language signals. Applied to Qwen2.5‑VL‑7B, PACE maintains 93.8% of performance using only 10% of visual tokens, achieving a 3.1× speedup in time to first token.

By Junjie Liu, Shengyuan Ye, Xu Chen
arXiv Computer Vision
Oct 9

Look Back, Think Ahead: Visual Memory on Demand for Efficient Multimodal Reasoning

The paper introduces ViMoD, a lightweight framework that dynamically manages visual memory for multimodal large language models. ViMoD uses Deformable Aggregation of Region-wise Tokens (DART) to create content‑adaptive coarse representations linked to recoverable fine tokens, and Temporal Routing for Adaptive Contextual Evidence (TRACE) to anticipate and adjust evidence needs during reasoning. On the Qwen3‑VL‑4B model, ViMoD achieves superior performance across eight reasoning benchmarks with only a 20% visual token budget and minimal additional parameters.

By Yicheng Xue, Han Wu, Jufeng Yang, Minjing Dong, Xinghao Chen, Hanting Chen, Jianyuan Guo
arXiv Computer Vision
Sep 10

VideoTIR: Accurate Understanding for Long Videos with Efficient Tool-Integrated Reasoning

VideoTIR introduces a reinforcement‑learning approach to improve long‑video understanding by encouraging multimodal large language models to use comprehensive multi‑level toolkits efficiently. It combines Zero‑RL and SFT cold‑starting strategies to help models retrieve and focus on meaningful video segments, images, and regions, thereby reducing hallucinations. The method includes Toolkit Action Grouped Policy Optimization (TAGPO) to streamline tool‑calling and a sandbox‑based trajectory synthesis framework for high‑quality data, achieving strong results on three long‑video QA benchmarks.

By Zhe Gao, Shiyu Shen, Taifeng Chai, Weinong Wang, Haotian Xu, Xing Wu, Wenbin Li, Qi Fan, Yang Gao, Dacheng Tao