Scaling Multimodal Large Language Models (MLLMs) to long-form video understanding is bottlenecked by the explosion of visual tokens, which saturates context windows and incurs prohibitive costs. Curre...
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.
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.
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.
arXiv:2607. 25266v2 Announce Type: replace-cross Abstract: Multimodal large language models (MLLMs) have enabled long-form video understanding at a scale that was not previously possible.
arXiv:2609.37042v1 Announce Type: cross Abstract: Video Large Language Models (VideoLLMs) have achieved strong video understanding capabilities but incur substantial inference overhead due to the lar...
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.
arXiv:2608.20814v1 Announce Type: new Abstract: Though Multimodal Large Language Models (MLLMs) have shown impressive potential in video understanding, long video understanding (LVU) remains challeng...
VisionWeave introduces elastic visual representation weaving, a native capability for multimodal large language models that learns where and at what granularity to encode visual information. The method combines a gated spatial pooler for coarse representations with a granularity router that allocates content‑adaptive token usage, trained end‑to‑end on large‑scale data. Experiments on Qwen3.5‑4B and Qwen3.8‑27B show that VisionWeave can save 43.0% of tokens while preserving 98.9% of performance across eight benchmarks, and delivers significant throughput gains and latency reductions when deployed on the SGLang serving engine.
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.
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.
arXiv:2609.37426v1 Announce Type: cross Abstract: Modern vision-language models (VLMs) have shown promising results in long-video understanding due to the rich semantic information they can capture....