Lune

ACL Findings2026

TheraAgent: Self-Improving Therapeutic Agent for Precise and Comprehensive Treatment Planning

Junkai Li, Yunghwei Lai, Tianyi Zhu, Zheng Long Lee, Weizhi Ma, Yang Liu

Abstract

Formulating a treatment plan is inherently a complex reasoning and refinement task rather than a simple generation problem. However, existing large language models (LLMs) mainly rely on one-shot output without explicit verification, which may result in rough, incomplete, and potentially unsafe treatment plans. To address these limitations, we propose Ther-aAgent, an agentic framework that replaces one-shot generation with an iterative generatereflect-refine pipeline. Inspired by the iterative revision patterns commonly observed in expert clinical practice, our framework progressively reflects on clinical criteria and transforms coarse drafts into precise, comprehensive and safer therapeutic regimens. To facilitate the critical reflection component, we introduce Thera-Judge, a treatment-specific evaluation module integrated into the inference loop. Experiments show TheraAgent achieves state-of-the-art results on HealthBench in treatment planning task, leading in Accuracy and Completeness. In expert evaluations, it attains an 86% win rate against physicians, with superior Targeting and Harm Control. Moreover, the high consistency between TheraJudge and HealthBench evaluation indicates the reliability of our framework.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

Builds onCited byLeft to right by year; the more cited, the higher.

Builds on9

  1. Auto-Arena: Automating LLM Evaluations with Agent Peer Battles and Committee DiscussionsRuochen Zhao, Wenxuan Zhang, Yew Ken Chia, Weiwen Xu et al.ACL 2025 · 34 citations
  2. Contextual Experience Replay for Self-Improvement of Language AgentsYitao Liu, Chenglei Si, Karthik R. Narasimhan, Shunyu YaoACL 2025 · 22 citations
  3. Efficient and Accurate Prompt Optimization: the Benefit of Memory in Exemplar-Guided ReflectionCilin Yan, Jingyun Wang, Lin Zhang, Ruihui Zhao et al.ACL 2025 · 16 citations
  4. MEDDxAgent: A Unified Modular Agent Framework for Explainable Automatic Differential DiagnosisDaniel Philip Rose, Chia-Chien Hung, Marco Lepri, Israa Alqassem et al.ACL 2025 · 14 citations
  5. Towards Medical Complex Reasoning with LLMs through Medical Verifiable ProblemsJunying Chen, Zhenyang Cai, Ke Ji, Xidong Wang et al.ACL Findings 2025 · 14 citations

Related papers

  1. EHRAgent: Code Empowers Large Language Models for Few-shot Complex Tabular Reasoning on Electronic Health RecordsWenqi Shi, Ran Xu, Yuchen Zhuang, Yue Yu et al.EMNLP 2024 · 33 citations
  2. ReflecTool: Towards Reflection-Aware Tool-Augmented Clinical AgentsYusheng Liao, Shuyang Jiang, Yanfeng Wang, Yu WangACL 2025 · 14 citations
  3. DrAgent: Empowering Large Language Models as Medical Agents for Multi-hop Medical ReasoningFenglin Liu, Zheng Li, Hongjian Zhou, Qingyu Yin et al.EMNLP Findings 2025
  4. MedAgent-Pro: Towards Evidence-based Multi-modal Medical Diagnosis via Reasoning Agentic WorkflowZiyue Wang, Junde Wu, Linghan Cai, Chang Han Low et al.ICLR 2026 · 84 citations
  5. Exploring the Future of AI in Clinical Collaboration: A Study on Tumor Board Case PreparationJiachen Li, Amanda K. Hall, Ruican Rachel Zhong, Selin S. Everett et al.CHI 2026 · 1 citation
Dusk over the sea between two cliffs drawn in fine vertical lines