AgenticSum: An Agentic Inference-Time Framework for Faithful Clinical Text Summarization

2026-02-23

AI Summary

This work addresses the challenge of factual inconsistency and hallucination in clinical text summarization, which commonly arise due to the length, noise, and heterogeneity of source documents. To tackle this issue, the paper introduces an agent-based architecture—the first of its kind for this task—proposing a zero-shot, inference-time framework that decouples generation from verification through four coordinated stages: context selection, draft generation, attention-guided fact verification, and targeted revision. By leveraging internal attention signals to precisely identify weakly supported segments, the framework effectively guides corrective refinements. Comprehensive evaluations on two public datasets—including automatic metrics, LLM-as-a-judge assessments, and human judgments—demonstrate that the proposed method significantly outperforms the base large language model and strong baselines, substantially improving the factual faithfulness of generated summaries.

Technology Category

• Natural Language Processing: Summarization
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Multiagent Systems: Adversarial Agents

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Large language models (LLMs) offer substantial promise for automating clinical text summarization, yet maintaining factual consistency remains challenging due to the length, noise, and heterogeneity of clinical documentation. We present AgenticSum, an inference-time, agentic framework that separates context selection, generation, verification, and targeted correction to reduce hallucinated content. The framework decomposes summarization into coordinated stages that compress task-relevant context, generate an initial draft, identify weakly supported spans using internal attention grounding signals, and selectively revise flagged content under supervisory control. We evaluate AgenticSum on two public datasets, using reference-based metrics, LLM-as-a-judge assessment, and human evaluation. Across various measures, AgenticSum demonstrates consistent improvements compared to vanilla LLMs and other strong baselines. Our results indicate that structured, agentic design with targeted correction offers an effective inference time solution to improve clinical note summarization using LLMs.

Problem

• clinical text summarization
• factual consistency
• hallucination
• large language models
• inference-time framework

Innovation

• agentic framework
• clinical text summarization
• inference-time reasoning
• hallucination reduction
• attention grounding