CiteVQA requires models to cite specific document regions with bounding boxes alongside answers and finds that even the strongest MLLMs frequently cite the wrong region, with top SAA scores of only 76.0 for closed models and 22.5 for open-source ones.
hub
M3docrag: Multi-modal retrieval is what you need for multi-page multi-document understanding
22 Pith papers cite this work. Polarity classification is still indexing.
22
Pith papers citing it
hub tools
citation-role summary
background 1
method 1
citation-polarity summary
representative citing papers
PixelRAG shows that operating RAG entirely over web screenshots outperforms text-based retrieval on NQ, SimpleQA, MMSearch, LiveVQA, and MoNaCo, with up to 18.1% accuracy gains and 3x token savings via image compression.
MemLens benchmark shows long-context LVLMs lose accuracy with length while memory agents lose visual fidelity, with multi-session reasoning below 30% for most systems and neither approach solving the task alone.
DocPrune is a training-free token pruning method that removes background and irrelevant tokens from document images using question and comprehension signals, yielding 3x encoder and 3.3x decoder throughput gains plus +1 F1 on M3DocRAG.
PaperMind is a new benchmark that evaluates integrated multimodal reasoning and critique over scientific papers through four complementary task families across seven domains.
Kamera stores a low-rank patch with each position-free KV chunk to restore cross-chunk conditioning lost in naive reuse, enabling cheap reordering, sliding windows, and recall across attention mechanisms.
EviProp improves evidence-page retrieval in long multimodal documents by combining dense visual priors with sparse chunk seeds and running Personalized PageRank on a multimodal Chunk-Page graph.
Constrained Dominant Sets on query-augmented graphs select complementary evidence for long multimodal document QA, claiming new SOTA of 66.99 on VisDoMBench and gains of 37.1 and 4.8 points over baselines.
Introduces a four-axis difficulty taxonomy integrated into an enterprise RAG benchmark to systematically diagnose multi-dimensional challenges like reasoning complexity and retrieval difficulty.
A realistic scene synthesis strategy and document-aware training recipe enable a 1B-parameter MLLM to achieve superior accuracy and robustness in end-to-end parsing of real-world captured documents.
MoDora introduces local-alignment aggregation, a Component-Correlation Tree, and question-type-aware retrieval to improve accuracy on semi-structured document QA by 5.97-61.07% over baselines.
Stellar reduces memory and latency by 1-2 orders of magnitude for multimodal document retrieval via lexical filtering with a fine-tuned MLLM and disk-backed late interaction without loss in effectiveness.
LensVLM trains VLMs to scan compressed rendered text images and selectively expand task-relevant regions, achieving 4.3x compression with near full-text accuracy and outperforming baselines up to 10.1x on text QA benchmarks.
DocSeeker improves long-document understanding in MLLMs via a two-stage training process that combines supervised fine-tuning from distilled data with evidence-aware group relative policy optimization and memory-efficient resolution allocation.
A multimodal RAG framework with ColPali retrieval and task-specific reasoning variants reports 32.6% relative improvement over prior RAG baselines on DesignQA, but the gain is inflated by test-set-fitted routing and an oracle baseline.
R3G improves vision-centric VQA by generating a reasoning plan before retrieval and reranking candidate images with an MLLM judge on relevance, target match, and answerability.
AGREE boosts visual document retrieval by adding local relevance signals from MLLM attention maps to global document labels during retriever training.
LLM planning agent with dynamic KG state achieves 81.5% accuracy on 200 multi-hop questions from NuScale FSAR documents, outperforming non-planning RAG baselines by up to 38pp.
DocArena automates creation of multimodal document QA training data via MLLM-based structuring and cross-page reasoning pairs, yielding agents with top retrieval and QA performance in unified tests.
Unveil proposes a visual-textual embedding model for multi-modal documents that is distilled into an efficient visual-only retriever.
The EReL@MIR 2025 Track 1 challenge evaluates single systems on two multimodal retrieval tasks and finds that Qwen2-VL decoder-based embedders dominate, with a training-free entry within 0.1 points of the fine-tuned winner.
citing papers explorer
-
CiteVQA: Benchmarking Evidence Attribution for Trustworthy Document Intelligence
CiteVQA requires models to cite specific document regions with bounding boxes alongside answers and finds that even the strongest MLLMs frequently cite the wrong region, with top SAA scores of only 76.0 for closed models and 22.5 for open-source ones.
-
PIXELRAG: Web Screenshots Beat Text for Retrieval-Augmented Generation
PixelRAG shows that operating RAG entirely over web screenshots outperforms text-based retrieval on NQ, SimpleQA, MMSearch, LiveVQA, and MoNaCo, with up to 18.1% accuracy gains and 3x token savings via image compression.
-
MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language Models
MemLens benchmark shows long-context LVLMs lose accuracy with length while memory agents lose visual fidelity, with multi-session reasoning below 30% for most systems and neither approach solving the task alone.
-
DocPrune:Efficient Document Question Answering via Background, Question, and Comprehension-aware Token Pruning
DocPrune is a training-free token pruning method that removes background and irrelevant tokens from document images using question and comprehension signals, yielding 3x encoder and 3.3x decoder throughput gains plus +1 F1 on M3DocRAG.
-
PaperMind: Benchmarking Agentic Reasoning and Critique over Scientific Papers in Multimodal LLMs
PaperMind is a new benchmark that evaluates integrated multimodal reasoning and critique over scientific papers through four complementary task families across seven domains.
-
Kamera: Unified Position-Invariant Multimodal KV Cache for Training-Free Reuse
Kamera stores a low-rank patch with each position-free KV chunk to restore cross-chunk conditioning lost in naive reuse, enabling cheap reordering, sliding windows, and recall across attention mechanisms.
-
EviProp: Seeded Relevance Diffusion on Chunk-Page Graphs for Long Multimodal Document Retrieval
EviProp improves evidence-page retrieval in long multimodal documents by combining dense visual priors with sparse chunk seeds and running Personalized PageRank on a multimodal Chunk-Page graph.
-
Constrained Dominant Sets for Multimodal Document Question Answering
Constrained Dominant Sets on query-augmented graphs select complementary evidence for long multimodal document QA, claiming new SOTA of 66.99 on VisDoMBench and gains of 37.1 and 4.8 points over baselines.
-
Overcoming the "Impracticality" of RAG: Proposing a Real-World Benchmark and Multi-Dimensional Diagnostic Framework
Introduces a four-axis difficulty taxonomy integrated into an enterprise RAG benchmark to systematically diagnose multi-dimensional challenges like reasoning complexity and retrieval difficulty.
-
Towards Real-World Document Parsing via Realistic Scene Synthesis and Document-Aware Training
A realistic scene synthesis strategy and document-aware training recipe enable a 1B-parameter MLLM to achieve superior accuracy and robustness in end-to-end parsing of real-world captured documents.
-
MoDora: Tree-Based Semi-Structured Document Analysis System
MoDora introduces local-alignment aggregation, a Component-Correlation Tree, and question-type-aware retrieval to improve accuracy on semi-structured document QA by 5.97-61.07% over baselines.
-
Stellar: Scalable Multimodal Document Retrieval for Natural Language Queries
Stellar reduces memory and latency by 1-2 orders of magnitude for multimodal document retrieval via lexical filtering with a fine-tuned MLLM and disk-backed late interaction without loss in effectiveness.
-
LensVLM: Selective Context Expansion for Compressed Visual Representation of Text
LensVLM trains VLMs to scan compressed rendered text images and selectively expand task-relevant regions, achieving 4.3x compression with near full-text accuracy and outperforming baselines up to 10.1x on text QA benchmarks.
-
DocSeeker: Structured Visual Reasoning with Evidence Grounding for Long Document Understanding
DocSeeker improves long-document understanding in MLLMs via a two-stage training process that combines supervised fine-tuning from distilled data with evidence-aware group relative policy optimization and memory-efficient resolution allocation.
-
MCERF: Advancing Multimodal LLM Evaluation of Engineering Documentation with Enhanced Retrieval
A multimodal RAG framework with ColPali retrieval and task-specific reasoning variants reports 32.6% relative improvement over prior RAG baselines on DesignQA, but the gain is inflated by test-set-fitted routing and an oracle baseline.
-
R3G: A Reasoning-Retrieval-Reranking Framework for Vision-Centric Answer Generation
R3G improves vision-centric VQA by generating a reasoning plan before retrieval and reranking candidate images with an MLLM judge on relevance, target match, and answerability.
-
Attention Grounded Enhancement for Visual Document Retrieval
AGREE boosts visual document retrieval by adding local relevance signals from MLLM attention maps to global document labels during retriever training.
-
LLM-Guided Planning for Multi-hop Reasoning over Multimodal Nuclear Regulatory Documents
LLM planning agent with dynamic KG state achieves 81.5% accuracy on 200 multi-hop questions from NuScale FSAR documents, outperforming non-planning RAG baselines by up to 38pp.
-
DocArena: Turning Raw Documents into Controllable Training Environments for Document Search Agents
DocArena automates creation of multimodal document QA training data via MLLM-based structuring and cross-page reasoning pairs, yielding agents with top retrieval and QA performance in unified tests.
-
Unveil: Unified Visual-Textual Integration and Distillation for Multi-modal Document Retrieval
Unveil proposes a visual-textual embedding model for multi-modal documents that is distilled into an efficient visual-only retriever.
-
Overview of the EReL@MIR 2025 Multimodal Document Retrieval Challenge (Track 1)
The EReL@MIR 2025 Track 1 challenge evaluates single systems on two multimodal retrieval tasks and finds that Qwen2-VL decoder-based embedders dominate, with a training-free entry within 0.1 points of the fine-tuned winner.
- The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modes