paper-with-me

홈 › Papers

QuoTA: Query-oriented Token Assignment via CoT Query Decouple for Long Video Comprehension

2025-03-11 · Yongdong Luo, Wang Chen, Xiawu Zheng, Weizhong Huang, Shukang Yin, Haojia Lin, Chaoyou Fu, Jinfa Huang, Jiayi Ji, Jiebo Luo, Rongrong Ji

Recent advances in long video understanding typically mitigate visual redundancy through visual token pruning based on attention distribution. However, while existing methods employ post-hoc low-response token pruning in decoder layers, they overlook the input-level semantic correlation between visual tokens and instructions (query). In this paper, we propose QuoTA, an ante-hoc training-free modular that extends existing large video-language models (LVLMs) for visual token assignment based on query-oriented frame-level importance assessment. The query-oriented token selection is crucial as it aligns visual processing with task-specific requirements, optimizing token budget utilization while preserving semantically relevant content. Specifically, (i) QuoTA strategically allocates frame-level importance scores based on query relevance, enabling one-time visual token assignment before cross-modal interactions in decoder layers, (ii) we decouple the query through Chain-of-Thoughts reasoning to facilitate more precise LVLM-based frame importance scoring, and (iii) QuoTA offers a plug-and-play functionality that extends to existing LVLMs. Extensive experimental results demonstrate that implementing QuoTA with LLaVA-Video-7B yields an average performance improvement of 3.2% across six benchmarks (including Video-MME and MLVU) while operating within an identical visual token budget as the baseline. Codes are open-sourced at https://github.com/MAC-AutoML/QuoTA.

📄 PDF Abstract BibTeX arXiv:2503.08689

Code (1)

mac-automl/quota 공식 구현 pytorch

Tasks

AutoMLDecoderMMEVideo MMEVideo Understanding

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Attention 설명 없음
Pruning 설명 없음

Similar Papers 제목 키워드 기반

Continuity of Topic, Interaction, and Query: Learning to Quote in Online Conversations

2021-06-18 · EMNLP 2020 11 · Lingzhi Wang, Jing Li, Xingshan Zeng, Haisong Zhang 외

Quotations are crucial for successful explanations and persuasions in interpersonal communications. However, finding what to quote in a conversation is challenging for both humans and machines. This work studies automati…

DecoderText Generation

Quotation Recommendation and Interpretation Based on Transformation from Queries to Quotations

2021-05-29 · ACL 2021 5 · Lingzhi Wang, Xingshan Zeng, Kam-Fai Wong

To help individuals express themselves better, quotation recommendation is receiving growing attention. Nevertheless, most prior efforts focus on modeling quotations and queries separately and ignore the relationship bet…

The Complexity of Interactively Learning a Stable Matching by Trial and Error

2020-02-18 · Ehsan Emamjomeh-Zadeh, Yannai A. Gonczarowski, David Kempe

In a stable matching setting, we consider a query model that allows for an interactive learning algorithm to make precisely one type of query: proposing a matching, the response to which is either that the proposed match…

Blocking

Who Speaks Matters: Authority-Aware Multi-View RAG over Italian Parliamentary Proceedings

2026-08-13 · Mirko Tritella, Riccardo Pozzi, Matteo Palmonari arxiv

Parliamentary proceedings are a primary record of democratic deliberation, yet their volume and fragmentation make multi-perspective access difficult for citizens, journalists, and researchers. Applying Retrieval-Augment…

D2Q-DETR: Decoupling and Dynamic Queries for Oriented Object Detection with Transformers

2023-03-01 · Qiang Zhou, Chaohui Yu, Zhibin Wang, Fan Wang

Despite the promising results, existing oriented object detection methods usually involve heuristically designed rules, e.g., RRoI generation, rotated NMS. In this paper, we propose an end-to-end framework for oriented o…

DecoderObjectobject-detectionObject Detection+2