PulseAugur / Brief
EN
LIVE 08:31:52

Brief

last 24h
[3/3] 223 sources

Multi-source AI news clustered, deduplicated, and scored 0–100 across authority, cluster strength, headline signal, and time decay.

  1. TOOL · arXiv cs.IR (Information Retrieval) English(EN) ·

    EviProp: Seeded Relevance Diffusion on Chunk-Page Graphs for Long Multimodal Document Retrieval

    Researchers have developed EviProp, a novel method for retrieving relevant pages from long, visually rich documents. Unlike existing approaches that score pages independently, EviProp models documents as multimodal Chunk-Page graphs. It uses seeded relevance diffusion, combining query-page similarity with chunk-level signals to improve retrieval accuracy. Experiments on benchmark datasets show EviProp outperforms traditional methods and leads to better downstream question-answering performance. AI

    IMPACT Enhances retrieval accuracy for complex multimodal documents, potentially improving AI systems that rely on document understanding.

  2. RESEARCH · Hugging Face Daily Papers English(EN) · · [2 sources]

    MARDoc: A Memory-Aware Refinement Agent Framework for Multimodal Long Document QA

    Researchers have introduced MARDoc, a novel framework designed to improve question answering for long, multimodal documents. This system utilizes three specialized agents: an Explorer for retrieval, a Refiner for processing interactions into structured memories, and a Reflector for feedback. By employing a dynamic structured memory instead of a continuously growing context, MARDoc aims to reduce noise and preserve critical information for more effective multi-hop reasoning. AI

    IMPACT Introduces a new method for handling complex, multimodal documents, potentially improving AI's ability to process and reason over extensive information.

  3. RESEARCH · arXiv cs.IR (Information Retrieval) English(EN) · · [2 sources]

    Constrained Dominant Sets for Multimodal Document Question Answering

    Researchers have developed a new retrieval method called Constrained Dominant Sets (CDS) for multimodal document question answering. This technique addresses limitations in current systems that struggle with long documents by selecting complementary evidence rather than near-duplicates. CDS encodes the query as a structural constraint, automatically balances relevance and redundancy, and avoids greedy heuristics by achieving global equilibrium. When used with a Qwen3-VL-32B reader, CDS sets a new state-of-the-art on VisDoMBench and significantly improves performance on MMLongBench-Doc. AI

    IMPACT Establishes new SOTA on multimodal QA benchmarks, improving retrieval for long documents.