paper-with-me

Papers

DynFrame: Adaptive Reasoning-Driven Multimodal Framework with Dynamic Frame Augmentation for Complex Video Understanding

2026-05-26 · Peng Zhang, Guanghao Zhang, Wanggui He, Longxiang Zhang, Mushui Liu, Yan Xia, Zhenhao Peng, Weilong Dai, Jinlong Liu, Haobing Tang, Le Zhang, Hao Jiang, Pipei Huang arxiv

Recent video multimodal large language models (MLLMs) increasingly couple step-by-step reasoning with on-demand visual evidence retrieval, allowing models to revisit relevant video segments during inference. However, two structural gaps remain in existing thinking-with-video systems. (i) Sampling density is not a learnable decision: existing methods may let the model decide where to look, but the per-window frame rate is largely fixed. As a result, fine-grained evidence is often recovered through repeated retrieval calls, which increases inference context length and training difficulty. (ii) Retrieval and answer generation are usually optimized with a single trajectory-level advantage, so the "where to look" tokens and the "how to answer" tokens receive the same credit even when one is correct and the other is not. To address these gaps, we present DynFrame, a framework that emits the temporal window and the sampling density as native tokens within a single autoregressive pass. This learnable span-density retrieval enables acquiring multi-granularity evidence with a single retrieval step. Based on the above tokenized retrieval interface, we further introduce Segment-Decoupled GRPO (SD-GRPO), which splits each rollout at the retrieval boundary and assigns role-specific token-level advantages, separately crediting the sampling decision and the answer. Trained on the curated DM-CoT-74k and DM-RL-45k, DynFrame-4B is competitive with strong 7B-8B baselines across six benchmarks (NExT-GQA, Charades-STA, ActivityNet-MR, Video-MME, MLVU, LVBench), and DynFrame-8B sets new state-of-the-art on most metrics. Code is available at https://github.com/zhangguanghao523/DynFrame.

📄 PDF Abstract BibTeX arXiv:2605.26680

Code (0)

등록된 구현이 없습니다.

Tasks

Answer Generation

Similar Papers 제목 키워드 기반

Agentic Multimodal AI for Hyperpersonalized B2B and B2C Advertising in Competitive Markets: An AI-Driven Competitive Advertising Framework

2025-04-01 · Sakhinana Sagar Srinivas, Akash Das, Shivam Gupta, Venkataramana Runkana

The growing use of foundation models (FMs) in real-world applications demands adaptive, reliable, and efficient strategies for dynamic markets. In the chemical industry, AI-discovered materials drive innovation, but comm…

Decision MakingIn-Context LearningMarketingMultimodal Reasoning+3

Perception, Reason, Think, and Plan: A Survey on Large Multimodal Reasoning Models

2025-05-08 · Yunxin Li, Zhenyu Liu, Zitao Li, Xuanyu Zhang 외

Reasoning lies at the heart of intelligence, shaping the ability to make decisions, draw conclusions, and generalize across domains. In artificial intelligence, as systems increasingly operate in open, uncertain, and mul…

Multimodal Reasoning

Intend, Reflect, Refine: An Adaptive Multimodal Reflection Framework for Autonomous Driving

2026-06-22 · Zisheng Chen, Yuping Qiu, Jianhua Han, Tao Tang 외 arxiv

Recent Vision-Language-Action (VLA) models have advanced end-to-end autonomous driving by incorporating reasoning for better interpretability and planning quality. However, most existing approaches directly generate the …

Computational EfficiencyAutonomous Driving

Think When Needed: Adaptive Reasoning-Driven Multimodal Embeddings with a Dual-LoRA Architecture

2026-05-14 · Longxiang Zhang, Weilong Dai, Guanghao Zhang, Hao Jiang 외 arxiv

Multimodal large language models (MLLMs) have emerged as a powerful backbone for multimodal embeddings. Recent methods introduce chain-of-thought (CoT) reasoning into the embedding pipeline to improve retrieval quality, …

Omni-AutoThink: Adaptive Multimodal Reasoning via Reinforcement Learning

2025-12-03 · Dongchao Yang, Songxiang Liu, Disong Wang, Yuanyuan Wang 외 arxiv

Recent advances in Omni models have enabled unified multimodal perception and generation. However, most existing systems still exhibit rigid reasoning behaviors, either overthinking simple problems or failing to reason w…

Reinforcement LearningMultimodal Reasoning