Evidence Selection

Follow this topic Ask about this topic

Momentum

5 papers in the last four weeks, against 1 the four weeks before. 0.0% of all new papers.

Jul 13Week of Sep 28

Latest papers 35

All topics

MEG-RAG: Quantifying Multi-modal Evidence Grounding for Evidence Selection in RAG
Apr 27, 2026Xihang Wang, Zihan Wang, Chengkai Huang +2Multimodal RAG Multimodal IR
Peking University Beijing, China · University of New South Wales, Macquarie University Sydney, Australia · University of New South Wales, CSIRO’s Data61 Sydney, Australia
Ensembles of Large Language Models for Identifying EQ-5D Studies in PubMed Based on Their Abstracts
Apr 24, 2026Zhyar Rzgar K. Rostam, Márta Péntek, János Tibor Czere +3Evidence Selection Language Model Ensembles
Doctoral School of Applied Informatics and Applied Mathematics, Obuda University, Budapest, Hungary · John von Neumann Faculty of Informatics, Obuda University, Budapest, Hungary · HECON Health Economics Research Center, University Research and Innovation Center, Obuda University, Budapest, Hungary +3
A Large-Scale, Cross-Disciplinary Corpus of Systematic Reviews
Apr 23, 2026Pierre Achkar, Tim Gollub, Arno Simons +2Evidence Selection Document Information Extraction
Leipzig University; Fraunhofer ISI · Bauhaus-Universität Weimar · TU Berlin +2
TiAb Review Plugin: A Browser-Based Tool for AI-Assisted Study Selection in Systematic Reviews
Apr 8, 2026Yuki Kataoka, Masahiro Banno, Michihito Kyo +8Evidence Selection AI-Assisted Scientific Research
Center for Postgraduate Clinical Training and Career Development, Nagoya University Hospital, 65, Tsurumai-cho, Showa-ku, Nagoya-city, Aichi, Japan · Center for Medical Education, Graduate School of Medicine, Nagoya University, 65, Tsurumai-cho, Showa-ku, Nagoya-city, Aichi, Japan · Scientific Research WorkS Peer Support Group (SRWS-PSG), Osaka, Japan +16
Relevance Is Not Permission: Localizing and Controlling Metric-Facing Attention Contributions
Date pendingMinwoo Yu, Young-guk HaFeature Attribution Evidence Selection
Smart Computing Laboratory, Department of Computer Science and Engineering, Konkuk University, 120 Neungdong-ro, Gwangjin-gu, Seoul 05029, Republic of Korea

Apr 27, 2026, cs.CLJ/K move · Enter open · S save

MEG-RAG: Quantifying Multi-modal Evidence Grounding for Evidence Selection in RAG

Xihang Wang, Zihan Wang, Chengkai Huang, Quan Z. Sheng+1

Peking University Beijing, China · University of New South Wales, Macquarie University Sydney, Australia · University of New South Wales, CSIRO’s Data61 Sydney, Australia

Multimodal RAG Multimodal IR Multimodal Grounding Multimodal Reranking Evidence Selection

Open paper AskPDF arXiv

Multimodal Retrieval-Augmented Generation (MRAG) addresses key limitations of Multimodal Large Language Models (MLLMs), such as hallucination and outdated knowledge. However, current MRAG systems struggle to distinguish whether retrieved multimodal data truly supports the semantic core of an answer or merely provides superficial relevance. Existing metrics often rely on heuristic position-based confidence, which fails to capture the informational density of multimodal entities. To address this, we propose Multi-modal Evidence Grounding (MEG), a semantic-aware metric that quantifies the contribution of retrieved evidence. Unlike standard confidence measures, MEG utilizes Semantic Certainty Anchoring, focusing on high-IDF information-bearing tokens that better capture the semantic core of the answer. Building on MEG, we introduce MEG-RAG, a framework that trains a multimodal reranker to align retrieved evidence with the semantic anchors of the ground truth. By prioritizing high-value content based on semantic grounding rather than token probability distributions, MEG-RAG improves the accuracy and multimodal consistency of generated outputs. Extensive experiments on the M

^2

RAG benchmark show that MEG-RAG consistently outperforms strong baselines and demonstrates robust generalization across different teacher models.

Momentum

Related topics

Latest papers 35

MEG-RAG: Quantifying Multi-modal Evidence Grounding for Evidence Selection in RAG

Ensembles of Large Language Models for Identifying EQ-5D Studies in PubMed Based on Their Abstracts

A Large-Scale, Cross-Disciplinary Corpus of Systematic Reviews

TiAb Review Plugin: A Browser-Based Tool for AI-Assisted Study Selection in Systematic Reviews

Relevance Is Not Permission: Localizing and Controlling Metric-Facing Attention Contributions