arXiv Machine Learning October 5, 2026 By Ambuj Mehrish, Sebastiano Vascon

Min-Cost Flow Routing for Evidence Assembly in Long Multimodal Documents

Read the original on arXiv Machine Learning →

The paper introduces <flowreader>, a method that casts evidence selection for long multimodal documents as a minimum‑cost flow problem over a multimodal content graph. It uses spectral decomposition to identify latent query‑relevant aspects and allocates a fixed evidence budget proportionally to their spectral energy, ensuring aspect coverage without a language‑model planning call. On the VisDoMBench benchmark with Qwen3‑VL‑32B, <flowreader> achieves the highest macro accuracy (68.9%) and outperforms prior systems by 2.7 points, while using fewer content nodes and reader tokens.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

arXiv Machine Learning

Jun 8

FLOWREADER: Min-Cost Flow Optimization for Multi-Modal Long Document Q&A

arXiv:2606. 07235v1 Announce Type: cross Abstract: Long, multimodal documents force retrieval-augmented systems to assemble answers from evidence fragmented across text, tables, and slides broken across cells in a long table, spread over multiple slides, or split between a figure and its discussion.

By Ambuj Mehrish, Sebatiano Vascon

rag multimodal

arXiv AI

Jul 8

Modality Relevance is not Modality Utility: Post-hoc Selective Modality Escalation for Cost-Aware Multimodal RAG

arXiv:2607. 05438v1 Announce Type: cross Abstract: Multimodal retrieval-augmented generation (RAG) grounds a generator in evidence drawn from heterogeneous modalities -- text, tables, and images.

By Xue Li, Yiming Gai

llms rag multimodal

arXiv AI

Sep 17

Transcribe, Then Reason: Two-Pass Decomposition for Multimodal Review

The paper demonstrates that a single-pass multimodal model struggles to produce faithful, comprehensive reviews of long recordings or documents, often omitting a third of the content and embellishing the rest. By splitting the task into two passes—first transcribing the source and then reviewing the transcript—the authors show improved faithfulness and coverage across a diverse set of 21 sources. The benefit is most pronounced for longer or weaker baseline cases, while the approach introduces new failure modes such as space constraints and memory confabulation.

By Bojie Li, Noah Shi

multimodal

arXiv AI

Jul 29

Salient Knowledge Pathways: Sparse Cross-Modal Routing for Efficient Knowledge-Intensive Multimodal Question Answering

arXiv:2607. 25422v1 Announce Type: new Abstract: Knowledge-intensive multimodal question answering (KI-MMQA) sits at the intersection of three expensive primitives: long visual token sequences, dense retrieval over large external corpora, and full cross-modal fusion.

By Noor Islam S. Mohammad, Ulu\u{g} Bayaz{\i}t

nlp efficiency multimodal benchmarks

arXiv AI

Aug 21

Question-Guided Evidence Acquisition for Multimodal Visual Question Answering

arXiv:2608. 19739v1 Announce Type: cross Abstract: Multimodal LLMs can see a document, but they often can't read it reliably.

By Alin-Ionut Popa

llms agents nlp multimodal

arXiv Computer Vision

Sep 10

Beyond Frame Selection: Rethinking Long-Video Understanding with MLLMs

arXiv:2608.05592v2 Announce Type: replace Abstract: Multimodal Large Language Models (MLLMs) have made strong progress in video understanding, yet long videos remain difficult: the visual token budge...

By Ziling Huang, Shin'ichi Satoh

llms multimodal benchmarks

One email a morning, machine-written

Related stories

FLOWREADER: Min-Cost Flow Optimization for Multi-Modal Long Document Q&A

Modality Relevance is not Modality Utility: Post-hoc Selective Modality Escalation for Cost-Aware Multimodal RAG

Transcribe, Then Reason: Two-Pass Decomposition for Multimodal Review

Salient Knowledge Pathways: Sparse Cross-Modal Routing for Efficient Knowledge-Intensive Multimodal Question Answering

Question-Guided Evidence Acquisition for Multimodal Visual Question Answering

Beyond Frame Selection: Rethinking Long-Video Understanding with MLLMs