Configurable Semantic Chunking for Biomedical Information Extraction in Retrieval-Augmented Generation

• URL: http://arxiv.org/abs/2608.31139v1
• Date: Mon, 31 Aug 2026 17:44:54 GMT
• Title: Configurable Semantic Chunking for Biomedical Information Extraction in Retrieval-Augmented Generation
• Abstract summary: BioMedRAG introduced retrieval-augmented generation with a learned chunk scorer for biomedical information extraction.<n>We propose a semantic chunking framework that combines entity-preserving windows, trigger-centered chunking, proposition-first extraction, tiered trigger prioritization, and hierarchical relation resolution.<n>We evaluate the framework on biomedical relation extraction benchmarks (GM-CIHT, DDI, ChemProt) and adverse event classification (ADE)
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: BioMedRAG introduced retrieval-augmented generation with a learned chunk scorer for biomedical information extraction. However, it relies on fixed-size chunking which can fragment semantic evidence. We propose a configurable semantic chunking framework that addresses this limitation by combining entity-preserving windows, trigger-centered chunking, proposition-first extraction, tiered trigger prioritization, and hierarchical relation resolution. The framework integrates with BioMedRAG by replacing only the chunk construction stage while preserving the embedding model, learned chunk scorer, generator, and evaluation protocol. We evaluate the framework on biomedical relation extraction benchmarks (GM-CIHT, DDI, ChemProt) and adverse event classification (ADE). On GM-CIHT, the full hybrid configuration achieves 82.6% F1, improving over the fixed-size baseline (74.2% F1) by 8.4 points under our experimental setup. Cross-dataset analysis shows that semantic chunking improves extraction datasets with explicit relation cues, such as GM-CIHT and DDI, while fixed chunking remains competitive or stronger for dense biochemical extraction and binary classification settings such as ChemProt and ADE. By externalizing chunking logic into configuration files, the framework provides an interpretable and adaptable alternative to rigid fixed-size chunking for biomedical RAG pipelines.

Related papers

• BioM-JEPA: joint-embedding prediction of graph-connected gene blocks in single cells [64.21963170814706]
  BioM-JEPA predicts aggregate representations of graph-connected gene blocks.<n>A student network infers each target-block representation from the remaining genes in a cell.<n>Across CellBench tasks, BioM-JEPA embeddings retained expression, pathway and neighbourhood information.
  arXiv  Detail & Related papers  (2026-08-06T11:58:29Z)
• Synergistic Perception-Reasoning Governance: Grounding Medical MLLMs with Verifiable Anatomical Evidence [45.78409219344429]
  Multimodal large language models (MLLMs) show strong promise for clinical VQA and radiology report generation, yet inference-time hallucinations undermine trustworthy use.<n>We propose a holistic, training-free evidence-injection framework that systematically mitigates hallucinations through dual-side evidence injection.<n>We conduct systematic evaluations on 2 tasks and 5 datasets using textttLLaVA-1.5-7B, texttLLaVA-Med-1.5-7B, texttQwen3-VL-8B/32B, and textttInternVL
  arXiv  Detail & Related papers  (2026-06-30T08:07:25Z)
• Unlocking Biological Workflows for Robust Protein-Text Question Answering: A Dual-Dimensional RAG Framework [60.82334952881798]
  Protein-Text Question Answering (QA) is crucial for interpreting biological sequences through natural language.<n>We propose 2D-ProteinRAG, a novel framework that empowers Large Language Models (LLMs) to operate within the gold-standard biological research workflow.<n>We show that 2D-Protein consistently achieves state-of-the-art performance, outperforming fine-tuned baselines and other RAG methods.
  arXiv  Detail & Related papers  (2026-05-17T05:03:24Z)
• Rethinking Molecular OOD Generalization via Target-Aware Source Selection [10.102858221967791]
  SCOPE-BENCH is a benchmark built on cluster-level partitioning in an explicit physicochemical descriptor space.<n>POMA is a framework that formulates knowledge transfer as a retrieve-compose-adapt pipeline.<n> Evaluations show that prediction errors of state-of-the-art 3D molecular models surge by up to 8.0x on SCOPE-BENCH with a mean of 5.9x.
  arXiv  Detail & Related papers  (2026-05-13T16:09:46Z)
• Graph-Aware Late Chunking for Retrieval-Augmented Generation in Biomedical Literature [0.0]
  GraLC-RAG is a framework that unifies late chunking with graph-aware structural intelligence.<n>We evaluate six strategies on 2,359 IMRaD-filtered Central articles using 2,033 cross-section questions.
  arXiv  Detail & Related papers  (2026-03-23T23:22:58Z)
• SHAPE: Structure-aware Hierarchical Unsupervised Domain Adaptation with Plausibility Evaluation for Medical Image Segmentation [7.174788000625294]
  Unsupervised Domain Adaptation (UDA) is essential for deploying medical segmentation models across diverse clinical environments.<n>Structure-aware Hierarchical Unsupervised Domain Adaptation with Plausibility Evaluation (SHAPE) is a framework that reframes adaptation towards global anatomical plausibility.
  arXiv  Detail & Related papers  (2026-03-23T12:25:56Z)
• Breaking It Down: Domain-Aware Semantic Segmentation for Retrieval Augmented Generation [5.0491491564528515]
  This paper introduces two efficient semantic chunking methods, Projected Similarity Chunking (PSC) and Metric Fusion Chunking (MFC)<n>Our results show substantial retrieval improvements (24x with PSC) in MRR and higher Hits@k on PubMedQA.<n>Despite being trained on a single domain, PSC and MFC also generalize well, achieving strong out-of-domain generation performance across multiple datasets.
  arXiv  Detail & Related papers  (2025-11-29T07:30:37Z)
• Automatic Fine-grained Segmentation-assisted Report Generation [3.6341072547314037]
  We present ASaRG, an extension of the popular LLaVA architecture for report generation.<n>Our approach achieves a +0.89% performance gain in CE F1 score compared to the LLaVA baseline.<n>Our code will be made publicly available at a later date.
  arXiv  Detail & Related papers  (2025-07-22T14:16:20Z)
• Hybrid X-Linker: Automated Data Generation and Extreme Multi-label Ranking for Biomedical Entity Linking [45.16091578348614]
  State-of-the-art deep learning entity linking methods rely on extensive human-labelled data. Current datasets are limited in size, leading to inadequate coverage of biomedical concepts. We propose to automatically generate data to create large-scale training datasets.
  arXiv  Detail & Related papers  (2024-07-08T18:04:22Z)
• ConDiSR: Contrastive Disentanglement and Style Regularization for Single Domain Generalization [42.810247034149214]
  Medical data often exhibits distribution shifts, which cause test-time performance degradation for deep learning models trained using standard pipelines. This study highlights the importance and challenges of exploring Single Domain Generalization frameworks in the context of the classification task.
  arXiv  Detail & Related papers  (2024-03-14T13:50:44Z)
• ISTR: End-to-End Instance Segmentation with Transformers [147.14073165997846]
  We propose an instance segmentation Transformer, termed ISTR, which is the first end-to-end framework of its kind. ISTR predicts low-dimensional mask embeddings, and matches them with ground truth mask embeddings for the set loss. Benefiting from the proposed end-to-end mechanism, ISTR demonstrates state-of-the-art performance even with approximation-based suboptimal embeddings.
  arXiv  Detail & Related papers  (2021-05-03T06:00:09Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.