This paper proposes an Intrinsic Temporal Adaptation (ITA) framework for PRVR, and introduces Affinity-Weighted Gradient Propagation to address the weakly supervised nature of PRVR, softly aggregating top-$k$ frames based on text-frame affinities and propagating learning signals to multiple query-relevant frames.
Abstract
Partially Relevant Video Retrieval (PRVR) aims to retrieve untrimmed videos that contain moments relevant to a text query. Since the target moment occupies only a portion of the video, PRVR requires retrieval based on fine-grained understanding beyond coarse video-level matching. However, existing methods often rely on frozen CLIP frame features, which lack temporal understanding. Even with recent progress in parameter-efficient CLIP adaptation, video-level predictions can still be supported by imprecise frame-level evidence. In this paper, we propose an Intrinsic Temporal Adaptation (ITA) framework for PRVR. First, our Backbone-Internal Temporal Adaptation allows the last few visual transformer layers to attend over groups of neighboring frames. This provides temporally aware frame embeddings while keeping CLIP frozen and training only adaptation parameters. Second, we introduce Affinity-Weighted Gradient Propagation to address the weakly supervised nature of PRVR, softly aggregating top-$k$ frames based on text-frame affinities and propagating learning signals to multiple query-relevant frames. Our method achieves state-of-the-art performance on PRVR benchmarks, demonstrates robust cross-dataset transfer, and retrieves substantially more accurate frame-level evidence within ground-truth query-relevant moments. Our code is available at github.com/hynnsk/ITA.
TF-PRVR is proposed, the first training-free framework for PRVR, which preserves the general-purpose alignment capability of pre-trained vision-language models and avoids dataset-specific overfitting.
Query-Aware Streaming Latent Reasoning (QASLR), a post-training framework that accumulates evidence across clips while keeping the embedding size fixed, and support streaming latent aggregation as an effective approach to integrating long-video evidence into fixed-dimensional representations.
Hao-Zhe Chi, Song Jin, Yang Jin et al.· 1 citation
Partially Relevant Video Retrieval (PRVR) seeks to retrieve untrim-med videos containing a moment that matches a text query, without temporal annotations. The relevant moment may last only seconds within a video spanning several minutes, creating an extremely low signal-to-noise ratio that makes PRVR more challenging t...
Zhen Liu, Le-Tian Li, Jinpeng Wang et al.· arXiv.org· 0 citations
Joint video moment retrieval and highlight detection requires identifying query-relevant temporal segments while estimating clip-level saliency, yet DETR-style pipelines do not explicitly preserve query-relevant evidence throughout encoding, decoding, and cross-task prediction. We propose EviDETR, an evidence-preservin...
STITCH is presented, a training-free method that divides a video into semantically meaningful temporal chunks that are computed once per video and reused across tasks, suggesting that reusable temporal abstraction is a promising direction for general video understanding.
Etienne Casanova, S. Brodjian, Pietro Perona· 0 citations
Streaming video understanding requires Vision Language Models (VLLMs) to process growing video streams and answer user questions under tight latency constraints. Existing methods improve efficiency through token pruning and memory-bank schemes, but mainly reduce visual tokens after visual encoding. Consequently, downst...
Jing-Chi Jiang, Yi-Ran Ling, Ruo-Nan Li et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.