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

• URL: http://arxiv.org/abs/2605.26680v1
• Date: Tue, 26 May 2026 08:16:16 GMT
• Title: DynFrame: Adaptive Reasoning-Driven Multimodal Framework with Dynamic Frame Augmentation for Complex Video Understanding
• Abstract summary: We present DynFrame, a framework that emits the temporal window and the sampling density as native tokens within a single autoregressive pass.<n>This learnable span-density retrieval enables acquiring multi-granularity evidence with a single retrieval step.
• Score: 19.492762364136393
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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.

Related papers

• RIDGE: Region-Informed Derivative-Guided Evidence Selection for Long Video Understanding [55.75126327155788]
  RIDGE is a frame selection framework that reads the frame-query similarity curve as a temporal signal.<n>It partitions the timeline into structural regions and applies region-specific selection to preserve event cores, transitions, buildup, aftermath, and contextual frames under a fixed budget.<n>It achieves the best performance in most settings and remains competitive in the others.
  arXiv  Detail & Related papers  (2026-08-30T18:40:51Z)
• Rethinking RAG in Long Videos: What to Retrieve and How to Use It? [56.38819694781005]
  We introduce V-RAGBench, a benchmark of $langle$query, evidence chunk, answer$rangle$ triplets that enables faithful, decoupled evaluation of retrieval and generation.<n>We also present CARVE, a method that runs parallel retrievers across configurations and employs chunk-adaptive reranking to identify the winning configuration for each chunk.
  arXiv  Detail & Related papers  (2026-06-11T10:05:49Z)
• Going Down Memory Lane: Scaling Tokens for Video Stream Understanding with Dynamic KV-Cache Memory [50.30283773196725]
  Existing approaches rely on key-value caching to accumulate frame-level details over time, but use a limited number of tokens per frame.<n>We propose scaling the token budget to enable more granular-temporal understanding and reasoning.
  arXiv  Detail & Related papers  (2026-02-20T18:59:50Z)
• CFVBench: A Comprehensive Video Benchmark for Fine-grained Multimodal Retrieval-Augmented Generation [29.58444236508143]
  Multimodal Retrieval-Augmented Generation (MRAG) enables Large Language Models (MLLMs) to generate responses with external multimodal evidence.<n>Existing benchmarks remain limited in modality coverage and format diversity.<n>We introduce CFVBench, a large-scale, manually verified benchmark constructed from 599 publicly available videos.
  arXiv  Detail & Related papers  (2025-10-10T11:05:37Z)
• FrameMind: Frame-Interleaved Video Reasoning via Reinforcement Learning [65.42201665046505]
  Current video understanding models rely on fixed frame sampling strategies, processing predetermined visual inputs regardless of the specific reasoning requirements of each question.<n>This static approach limits their ability to adaptively gather visual evidence, leading to suboptimal performance on tasks that require broad temporal coverage or fine-grained spatial detail.<n>We introduce FrameMind, an end-to-end framework trained with reinforcement learning that enables models to dynamically request visual information during reasoning through Frame-Interleaved Chain-of-Thought (FiCOT)<n>Unlike traditional approaches, FrameMind operates in multiple turns where the model alternates between textual reasoning and active visual perception, using tools to extract
  arXiv  Detail & Related papers  (2025-09-28T17:59:43Z)
• Less is More: Token-Efficient Video-QA via Adaptive Frame-Pruning and Semantic Graph Integration [24.337139909108117]
  "Less is more" phenomenon where excessive frames can paradoxically degrade performance due to context dilution.<n>"Visual echoes" yield significant temporal redundancy, which we term 'visual echoes'<n>"AFP" employs an adaptive hierarchical clustering algorithm on a fused ResNet-50 and CLIP feature space to identify and merge these echoes into single representatives.<n>Our full approach demonstrates a drastic reduction in required frames by up to 86.9% and total input tokens by up to 83.2%.
  arXiv  Detail & Related papers  (2025-08-05T11:31:55Z)
• Towards Universal Modal Tracking with Online Dense Temporal Token Learning [66.83607018706519]
  We propose a universal video-level modality-awareness tracking model with online dense temporal token learning.<n>We expand the model's inputs to a video sequence level, aiming to see a richer video context from a near-global perspective.
  arXiv  Detail & Related papers  (2025-07-27T08:47:42Z)
• MAGNET: A Multi-agent Framework for Finding Audio-Visual Needles by Reasoning over Multi-Video Haystacks [67.31276358668424]
  We introduce a novel task named AV-HaystacksQA, where the goal is to identify salient segments across different videos in response to a query and link them together to generate the most informative answer.<n> AVHaystacks is an audio-visual benchmark comprising 3100 annotated QA pairs designed to assess the capabilities of LMMs in multi-video retrieval and temporal grounding task.<n>We propose a model-agnostic, multi-agent framework to address this challenge, achieving up to 89% and 65% relative improvements over baseline methods on BLEU@4 and GPT evaluation scores in QA task on our proposed AVHaystack
  arXiv  Detail & Related papers  (2025-06-08T06:34:29Z)
• Chain-of-Frames: Advancing Video Understanding in Multimodal LLMs via Frame-Aware Reasoning [37.86612817818566]
  We propose to obtain video LLMs whose reasoning steps are grounded in, and explicitly refer to, the relevant video frames.<n>Our approach is simple and self-contained, and, unlike existing approaches for video CoT, does not require auxiliary networks to select or caption relevant frames.<n>This, in turn, leads to improved performance across multiple video understanding benchmarks.
  arXiv  Detail & Related papers  (2025-05-31T00:08:21Z)
• VidCtx: Context-aware Video Question Answering with Image Models [15.1350316858766]
  We introduce VidCtx, a novel training-free VideoQA framework which integrates both visual information from input frames and textual descriptions of others frames.<n>Experiments show that VidCtx achieves competitive performance among approaches that rely on open models.
  arXiv  Detail & Related papers  (2024-12-23T09:26:38Z)
• Frame-wise Cross-modal Matching for Video Moment Retrieval [32.68921139236391]
  Video moment retrieval targets at retrieving a moment in a video for a given language query. The challenges of this task include 1) the requirement of localizing the relevant moment in an untrimmed video, and 2) bridging the semantic gap between textual query and video contents. We propose an Attentive Cross-modal Relevance Matching model which predicts the temporal boundaries based on an interaction modeling.
  arXiv  Detail & Related papers  (2020-09-22T10:25:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.