Reason-Then-Retrieve for CoVR-R with Structured Edit Prompts and Dense-Sparse Fusion

• URL: http://arxiv.org/abs/2606.02450v1
• Date: Mon, 01 Jun 2026 16:21:01 GMT
• Title: Reason-Then-Retrieve for CoVR-R with Structured Edit Prompts and Dense-Sparse Fusion
• Abstract summary: We build a zero-shot reason-then-retrieve pipeline around Qwen3.5-27B.<n>For each gallery video, the model generates a retrieval-oriented structured description and a dense embedding.<n>On validation, the current best submission reaches 80.81 at R@1, 94.86 at R@5, 97.11 at R@10, and 98.59 at R@50.
• Score: 13.20120200276798
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: CoVR-R studies reason-aware composed video retrieval: given a reference video and an edit instruction, the system must retrieve the target video that satisfies the edit. The main difficulty is that the target is not described directly; it must be inferred from fine-grained changes in object identity, action order, final state, hand interaction, and scene transition. We build a zero-shot reason-then-retrieve pipeline around Qwen3.5-27B. For each gallery video, the model generates a retrieval-oriented structured description and a dense embedding by pooling generated-token hidden states with token-dependent weights. For each query, the model first performs edit reasoning over the reference video and instruction, then generates a target-video description whose hidden states serve as the query embedding. We complement dense retrieval with a TF-IDF branch over the generated texts and fuse the two rankings with split-specific weights. On validation, the current best submission reaches 80.81 at R@1, 94.86 at R@5, 97.11 at R@10, and 98.59 at R@50. On the blind test split, it reaches 89.73 at R@1, 95.79 at R@5, 96.63 at R@10, and 97.98 at R@50.

Related papers

• Beyond Similarity: Foundation Models as an Efficient Backbone for Training-Free Composed Video Retrieval [10.716619269996677]
  Composed video retrieval (CoVR) searches a gallery for the target video that realizes a natural-language modification of a source clip.<n>We propose that frozen foundation models should instead occupy complementary roles, with inference depth adapted to query difficulty.<n>Our method reaches state-of-the-art performance among training-free approaches, with 89.55 and 93.43 R@1 on-WebVid-CoVR and CoVR-R, respectively.
  arXiv  Detail & Related papers  (2026-09-09T10:38:25Z)
• VisDocAgentBench: Benchmarking Agents for Visually Rich Document Retrieval [65.89247522243959]
  VisDocAgentBench is a closed-corpus benchmark comparing static and agentic retrieval under a ranked-output contract.<n>It contains 2,375 pages from 100 documents and 120 unique-target queries balanced across direct, one-bridge, and two-bridge evidence structures.<n>A strong late-interaction visual retriever reaches 97.50% Recall@1 on direct items but 2.50% on two-bridge items, exposing the limits of query--target matching when relevance depends on corpus context.
  arXiv  Detail & Related papers  (2026-08-18T15:23:06Z)
• R^3: Composed Video Retrieval via Reasoning-Guided Recalling and Re-ranking [73.0537447183962]
  We present a zero-shot composed video retrieval pipeline built around Reasoning-guided Recalling and Reranking.<n>The model generates a reasoning trace that describes the expected target video after applying the edit.<n>Experiments have demonstrated the effectiveness of our method in addressing this challenge.
  arXiv  Detail & Related papers  (2026-05-31T09:20:53Z)
• Reason, Retrieve, Re-rank: A Zero-Shot Reasoning-Aware Framework for Composed Video Retrieval [0.0]
  Composed Video Retrieval (CoVR) seeks the target video that results from applying a free-form textual modification to a reference video.<n>We present textbfR3-CoVR (emphReason, Retrieve, Re-rank), a training-free pipeline built entirely from frozen foundation models.
  arXiv  Detail & Related papers  (2026-05-30T22:21:42Z)
• CRAFT: Critic-Refined Adaptive Key-Frame Targeting for Multimodal Video Question Answering [2.3450855956637913]
  Grounded multi-video question answering over real-world news events requires systems to surface query-relevant evidence across heterogeneous archives.<n>We introduce CRAFT, a query-conditioned pipeline that combines dynamic selection, per-video ASR with multilingual fallback, and a hybrid critic loop.<n>On MAGMaR 2026, CRAFT achieves the best overall average (0.739), reference recall (0.810), and citation F1 (0.635)
  arXiv  Detail & Related papers  (2026-05-18T20:01:05Z)
• Beyond Simple Edits: Composed Video Retrieval with Dense Modifications [96.46069692338645]
  We introduce a novel dataset that captures both fine-grained and composed actions across diverse video segments.<n>Dense-WebVid-CoVR consists of 1.6 million samples with dense modification text that is around seven times more than its existing counterpart.<n>We develop a new model that integrates visual and textual information through Cross-Attention (CA) fusion.
  arXiv  Detail & Related papers  (2025-08-19T17:59:39Z)
• Video Referring Expression Comprehension via Transformer with Content-conditioned Query [68.06199031102526]
  Video Referring Expression (REC) aims to localize a target object in videos based on the queried natural language. Recent improvements in video REC have been made using Transformer-based methods with learnable queries.
  arXiv  Detail & Related papers  (2023-10-25T06:38:42Z)
• GMMFormer: Gaussian-Mixture-Model Based Transformer for Efficient Partially Relevant Video Retrieval [59.47258928867802]
  Given a text query, partially relevant video retrieval (PRVR) seeks to find videos containing pertinent moments in a database. This paper proposes GMMFormer, a Gaussian-Mixture-Model based Transformer which models clip representations implicitly. Experiments on three large-scale video datasets demonstrate the superiority and efficiency of GMMFormer.
  arXiv  Detail & Related papers  (2023-10-08T15:04:50Z)
• DiffusionRet: Generative Text-Video Retrieval with Diffusion Model [56.03464169048182]
  Existing text-video retrieval solutions focus on maximizing the conditional likelihood, i.e., p(candidates|query) We creatively tackle this task from a generative viewpoint and model the correlation between the text and the video as their joint probability p(candidates,query) This is accomplished through a diffusion-based text-video retrieval framework (DiffusionRet), which models the retrieval task as a process of gradually generating joint distribution from noise.
  arXiv  Detail & Related papers  (2023-03-17T10:07:19Z)

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.