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

arXiv: https://arxiv.org/abs/2604.24564 · HTML: https://mlsanity.com/paper/2604.24564/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Document understanding and multimodal RAG, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 13 of its 22 references that are in the map.
- First posted on arXiv 27 Apr 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Document understanding and multimodal RAG](https://mlsanity.com/mllm/document-understanding-and-multimodal-rag/)
- Retrieval-augmented generation / [Multimodal RAG](https://mlsanity.com/rag/multimodal-rag/)

## Builds on (most cited first, 13 of 13)

- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency](https://mlsanity.com/paper/2508.18265/) (2025)
- [Retrieving Multimodal Information for Augmented Generation: A Survey](https://mlsanity.com/paper/2303.10868/) (2023)
- [MLLM Is a Strong Reranker: Advancing Multimodal Retrieval-augmented Generation via Knowledge-enhanced Reranking and Noise-injected Training](https://mlsanity.com/paper/2407.21439/) (2024)
- [Ask in Any Modality: A Comprehensive Survey on Multimodal Retrieval-Augmented Generation](https://mlsanity.com/paper/2502.08826/) (2025)
- [Log Probabilities Are a Reliable Estimate of Semantic Plausibility in Base and Instruction-Tuned Language Models](https://mlsanity.com/paper/2403.14859/) (2024)
- [MM-R5: MultiModal Reasoning-Enhanced ReRanker via Reinforcement Learning for Document Retrieval](https://mlsanity.com/paper/2506.12364/) (2025)
- [Benchmarking Retrieval-Augmented Generation in Multi-Modal Contexts](https://mlsanity.com/paper/2502.17297/) (2025)
- [SceneAlign: Aligning Multimodal Reasoning to Scene Graphs in Complex Visual Scenes](https://mlsanity.com/paper/2601.05600/) (2026)
- [Embedding-Informed Adaptive Retrieval-Augmented Generation of Large Language Models](https://mlsanity.com/paper/2404.03514/) (2024)
- [MemWeaver: Weaving Hybrid Memories for Traceable Long-Horizon Agentic Reasoning](https://mlsanity.com/paper/2601.18204/) (2026)
- [PruneRAG: Confidence-Guided Query Decomposition Trees for Efficient Retrieval-Augmented Generation](https://mlsanity.com/paper/2601.11024/) (2026)
- [SpeechAgent: An End-to-End Mobile Infrastructure for Speech Impairment Assistance](https://mlsanity.com/paper/2510.20113/) (2025)

## Built on by (most cited first, 2 of 2)

- [GraphMemix: Query-Aware Evidence Forests for Long-Term Multimodal Agent Memory](https://mlsanity.com/paper/2608.26983/) (2026)
- [TAEC: Trajectory-Aware Evidence Coordination for Multi-Step Visual RAG](https://mlsanity.com/paper/2609.37349/) (2026)

## Abstract (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 M2RAG benchmark show that MEG-RAG consistently outperforms strong baselines and demonstrates robust generalization across different teacher models.