AdaFocus: Adaptive Relevance-Diversity Sampling with Zero-Cache Look-back for Efficient Long Video Understanding

• URL: http://arxiv.org/abs/2605.12954v1
• Date: Wed, 13 May 2026 03:40:21 GMT
• Title: AdaFocus: Adaptive Relevance-Diversity Sampling with Zero-Cache Look-back for Efficient Long Video Understanding
• Abstract summary: Long video understanding is heavily bottlenecked by a rigid one-shot paradigm.<n>We propose AdaFocus, an efficient framework that rethinks long-video understanding as progressive evidence acquisition rather than one-pass encoding.
• Score: 4.238932162084854
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Long video understanding is heavily bottlenecked by a rigid one-shot paradigm: existing methods either densely encode videos at prohibitive memory and latency costs, or aggressively compress them into sparse frame sets that irreversibly discard fine-grained evidence needed for downstream reasoning. Consequently, current models struggle to simultaneously balance temporal coverage, visual details, and computational efficiency. We propose AdaFocus, an efficient framework that rethinks long-video understanding as progressive evidence acquisition rather than one-pass encoding. AdaFocus relies on two tightly coupled components. First, a Query-Aware Adaptive Relevance-Diversity sampler (AdaRD) produces a compact yet informative video preview, adaptively switching to global clustering when the query lacks reliable local grounding. Second, instead of caching exhaustive frame sequences in memory, AdaFocus introduces an uncertainty-triggered refinement mechanism. It performs targeted look-back only when the model is not confident, retrieving high-resolution evidence directly from disk via a zero-cache I/O design. This turns discarded visual details from an irreversible loss into on-demand recoverable evidence without paying the cost of exhaustive preloading. Experiments on seven standard long-video benchmarks show that AdaFocus delivers a substantially better efficiency-accuracy trade-off than strong baselines. Compared with conventional dense encoding, AdaFocus achieves improved task performance (e.g., +2.59 accuracy on VideoMME, +8.39 mIoU on Charades-STA over single-pass inference) while reducing visual token consumption by ~33x and eliminating the need for in-memory frame pre-caching through its zero-cache disk retrieval design. These findings suggest that progressive preview combined with zero-cache evidence refinement is a highly effective paradigm for scalable multimedia reasoning.

Related papers

• CoFiE: Coarse-to-Fine Evidence Selection for Efficient Streaming Video Understanding [9.22946015931432]
  Streaming video understanding requires Vision Language Models (VLLMs) to process growing video streams and answer user questions under tight latency constraints.<n>We propose CoFiE, a Coarse-to-Fine Evidence Selection framework that decouples evidence selection into a coarse, query-agnostic filtering stage before the vision encoder and a fine, query-specific refinement stage during LLM prefill.<n>CoFiE establishes a new state-of-the-art accuracy-efficiency trade-off across multiple video understanding benchmarks, reaching 78.86% accuracy on StreamingBench and 68.72% on OvO-Bench, with improvements of up to 3
  arXiv  Detail & Related papers  (2026-09-03T11:14:21Z)
• TetherCache: Stabilizing Autoregressive Long-Form Video Generation with Gated Recall and Trusted Alignment [51.33418612284208]
  We propose TetherCache, a training-free and plug-and-play cache management strategy for drift-resistant long video generation.<n>Gated Recall with Attention-Diversity Balancing selects long-range memory frames using a gated score.<n>TAME lightly edits newly recalled memory tokens by aligning their statistics to a trusted context distribution.
  arXiv  Detail & Related papers  (2026-06-11T08:16:08Z)
• Progressive Online Video Understanding with Evidence-Aligned Timing and Transparent Decisions [75.23170605943457]
  textbfmodel is a framework that decouples reasoning control from memory integration.<n>emphActive Thinking Decision Maker (ATDM) is a transparent reasoning controller that externalizes its decision process.<n>emphHierarchical Progressive Semantic Integration (HPSI) module acts as an efficient memory system.
  arXiv  Detail & Related papers  (2026-04-20T16:15:33Z)
• SenCache: Accelerating Diffusion Model Inference via Sensitivity-Aware Caching [75.02865981328509]
  Caching reduces computation by reusing previously computed model outputs across timesteps.<n>We propose Sensitivity-Aware Caching (SenCache), a dynamic caching policy that adaptively selects caching timesteps on a per-sample basis.<n>SenCache achieves better visual quality than existing caching methods under similar computational budgets.
  arXiv  Detail & Related papers  (2026-02-27T17:36:09Z)
• WeaveTime: Stream from Earlier Frames into Emergent Memory in VideoLLMs [37.61875409530676]
  We present WeaveTime, a simple, efficient, and model agnostic framework that first teaches order and then uses order.<n>At inference, a Past-Current Dynamic Focus Cache performs uncertainty triggered, coarse-to-fine retrieval, expanding history only when needed.<n>These results establish WeaveTime as a practical path toward time aware stream Video-LLMs under strict online, time causal constraints.
  arXiv  Detail & Related papers  (2026-02-25T17:45:45Z)
• 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)
• ERTACache: Error Rectification and Timesteps Adjustment for Efficient Diffusion [30.897215456167753]
  Diffusion models suffer from substantial computational overhead due to their inherently iterative inference process.<n>We propose ERTACache, a principled caching framework that jointly rectifies both error types.<n>ERTACache achieves up to 2x inference speedup while consistently preserving or even improving visual quality.
  arXiv  Detail & Related papers  (2025-08-27T10:37:24Z)
• PromptTea: Let Prompts Tell TeaCache the Optimal Threshold [1.0665410339553834]
  A common acceleration strategy involves reusing model outputs via caching mechanisms at fixed intervals.<n>We propose Prompt-Complexity-Aware (PCA) caching, a method that automatically adjusts reuse thresholds based on scene complexity estimated directly from the input prompt.
  arXiv  Detail & Related papers  (2025-07-09T10:53:05Z)
• Less is Enough: Training-Free Video Diffusion Acceleration via Runtime-Adaptive Caching [57.7533917467934]
  EasyCache is a training-free acceleration framework for video diffusion models.<n>We conduct comprehensive studies on various large-scale video generation models, including OpenSora, Wan2.1, and HunyuanVideo.<n>Our method achieves leading acceleration performance, reducing inference time by up to 2.1-3.3$times$ compared to the original baselines.
  arXiv  Detail & Related papers  (2025-07-03T17:59:54Z)
• FasterCache: Training-Free Video Diffusion Model Acceleration with High Quality [58.80996741843102]
  FasterCache is a training-free strategy designed to accelerate the inference of video diffusion models with high-quality generation.<n>We show that FasterCache can significantly accelerate video generation while keeping video quality comparable to the baseline.
  arXiv  Detail & Related papers  (2024-10-25T07:24:38Z)

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.