# Rudy 焦点短名单（对话 / Agent / RAG / 主动问诊 / 指南 / 分诊 / 不确定）

- 生成：2026-10-19 13:03 Asia/Shanghai
- 从 jsonl 关键词筛出 **72** 篇（候选池 1781）
- 字段：title · venue · year · url_page · url_code · summary_zh（一行）

## 1. OccuTriage: An AI Agent Orchestration Framework for Occupational Health Triage Prediction
- **venue/year**: ACL · 2025
- **url_page**: https://aclanthology.org/2025.acl-industry.84/
- **url_code**: https://github.com/Lucanyc/TriageAgent
- **summary_zh**: 本篇（ACL）入选 Cardiology-Chat 心内优先阅读包，相关信号：多智能体/Agent；对话/问诊；∩Dialogue；Agent切片。英文摘要要点：Occupational Health (OH) triage is a systematic process for evaluating and prioritising workplac…

## 2. CliCARE: Grounding Large Language Models in Clinical Guidelines for Decision Support over Longitudinal Cancer Electronic Health Records
- **venue/year**: AAAI · 2026
- **url_page**: https://ojs.aaai.org/index.php/AAAI/article/view/40421
- **url_code**: https://github.com/sakurakawa1/CliCARE
- **summary_zh**: 本篇（AAAI）入选短名单，相关信号：主题命中；CliCARE·纵向病历指南接地决策支持；指南/CDSS/路径/治疗规划主题命中；顶会；指南接地加分；CDSS/决策支持加分；核心点名。英文摘要要点：Target: Decision Support over Longitudinal Cancer EHRs combined with KG and TK…

## 3. MedAgent-Pro: Towards Evidence-based Multi-modal Medical Diagnosis via Reasoning Agentic Workflow
- **venue/year**: ICLR · 2026
- **url_page**: https://openreview.net/forum?id=ZOuU0udyA4
- **url_code**: https://github.com/jinlab-imvr/MedAgent-Pro
- **summary_zh**: 本文面向心血管/心内科、心内×临床对话场景，提出/研究「MedAgent-Pro: Towards Evidence-based Multi-modal Medical Diagnosis via Reasoning Agentic Workflow」，技术焦点包括多智能体/Agent。论文核心工作可概括为：In modern medicine, cl…

## 4. MMedAgent-RL: Optimizing Multi-Agent Collaboration for Multimodal Medical Reasoning
- **venue/year**: ICLR · 2026
- **url_page**: https://openreview.net/forum?id=2awntLXwR6
- **url_code**: https://github.com/Deep-Agent/R1-V
- **summary_zh**: 本篇（ICLR）入选短名单，相关信号：Agent主题；MMedAgent；多Agent临床工作流；顶会。英文摘要要点：Medical Large Vision-Language Models (Med-LVLMs) have shown strong potential in multimodal diagnostic tasks. However, …

## 5. 3MDBench: Medical Multimodal Multi-agent Dialogue Benchmark
- **venue/year**: EMNLP · 2025
- **url_page**: https://aclanthology.org/2025.emnlp-main.1353/
- **url_code**: https://github.com/univanxx/3mdbench
- **summary_zh**: 本篇（EMNLP）入选短名单，相关信号：Agent主题；多模态多Agent对话基准；多Agent临床工作流；顶会。英文摘要要点：Though Large Vision-Language Models (LVLMs) are being actively explored in medicine, their ability to conduct com…

## 6. MDTeamGPT: Mitigating Context Collapse and Enabling Self-Evolution in Medical Multi-Agent Reasoning
- **venue/year**: Findings-ACL · 2026
- **url_page**: https://aclanthology.org/2026.findings-acl.1427/
- **url_code**: https://github.com/KaiChenNJ/MDTeamGPT
- **summary_zh**: 本篇（Findings-ACL）入选短名单，相关信号：Agent主题；MDTeamGPT；多Agent临床工作流；顶会；核心必读。英文摘要要点：A previously healthy 36-year-old man presents with yellow skin discoloration and dark urine for 2 weeks ……

## 7. StoryMI: Steerable Multi-Agent Therapeutic Dialogue Generation
- **venue/year**: Findings-ACL · 2026
- **url_page**: https://aclanthology.org/2026.findings-acl.468/
- **url_code**: https://github.com/Beren-sds/StoryMI
- **summary_zh**: 《StoryMI: Steerable Multi-Agent Therapeutic Dialogue Generation》发表于ACL系会场、ACL/EMNLP残余扫街，重点涉及多智能体/Agent、对话/问诊/追问。英文摘要要点：Abstract Large language models (LLMs) can generate fluent …

## 8. LLM-Based Multi-Agent Systems for Clinical Workflows: A Survey of AI Hospitals
- **venue/year**: ACL · 2026
- **url_page**: https://aclanthology.org/2026.acl-long.2123/
- **url_code**: （无）
- **summary_zh**: 本篇（ACL）入选短名单，相关信号：Agent主题；AI Hospital；AI Hospital综述；多Agent临床工作流；顶会；核心必读。英文摘要要点：This survey reviews LLM-based multi-agent systems for clinical and healthcare workflows, including…

## 9. SEMA-RAG: A Self-Evolving Multi-Agent Retrieval-Augmented Generation Framework for Medical Reasoning
- **venue/year**: Findings-ACL · 2026
- **url_page**: https://aclanthology.org/2026.findings-acl.917/
- **url_code**: （无）
- **summary_zh**: 本篇（Findings-ACL）入选短名单，相关信号：主题命中；SEMA-RAG多Agent医学；顶会；医学语境；核心点名。英文摘要要点：Retrieval-Augmented Generation (RAG) is widely employed to mitigate risks such as hallucinations and knowled…

## 10. From Conflict to Consensus: Boosting Medical Reasoning via Multi-Round Agentic RAG
- **venue/year**: ICML · 2026
- **url_page**: https://openreview.net/forum?id=S7lpdz7NAW
- **url_code**: https://github.com/NJU-RL/MA-RAG
- **summary_zh**: 《From Conflict to Consensus: Boosting Medical Reasoning via Multi-Round Agentic RAG》属于Agent工具×临床工作流相关工作，重点涉及多智能体/Agent、检索/知识增强。英文摘要要点：Large Language Models (LLMs) exhibit high r…

## 11. The Attachment Index: Auditing Attachment Language Cues and Relational Safety Risks in Human-LLM Dialogue
- **venue/year**: CLPSYCH · 2026
- **url_page**: https://aclanthology.org/2026.clpsych-1.26/
- **url_code**: https://github.com/animeshprasad/llm_attachment_index
- **summary_zh**: 《The Attachment Index: Auditing Attachment Language Cues and Relational Safety Risks in Human-LLM Dialogue》属于弃权/转诊×分诊相关工作，重点涉及对话/问诊/追问、安全/不确定。英文摘要要点：As conversational AI systems…

## 12. AI-Assisted Triage and Decision Support of Head and Neck Cancer Screening and Diagnosis in Low-Resourced Settings
- **venue/year**: IJCAI · 2025
- **url_page**: https://www.ijcai.org/proceedings/2025/1087
- **url_code**: （无）
- **summary_zh**: 本篇（IJCAI）入选短名单，相关信号：主题命中；H&N Triage·头颈癌筛查分诊决策支持；分诊/急救/院前/路由主题命中；顶会；核心点名。本地摘要暂缺，请直接打开 PDF。可与 CLARITY / InTriage / SOAPTriage / OccuTriage、AI Hospitals / Spanish Triage / H&N Tria…

## 13. Ask Patients with Patience: Enabling LLMs for Human-Centric Medical Dialogue with Grounded Reasoning
- **venue/year**: EMNLP · 2025
- **url_page**: https://aclanthology.org/2025.emnlp-main.142/
- **url_code**: （无）
- **summary_zh**: 本篇（EMNLP）入选 ProMed 相邻短名单，相关信号：医患对话；顶会NLP。英文摘要要点：The severe shortage of medical doctors limits access to timely and reliable healthcare, leaving millions underserved. Large langu…

## 14. DART: Leveraging Multi-Agent Disagreement for Tool Recruitment in Multimodal Reasoning
- **venue/year**: EACL · 2026
- **url_page**: https://aclanthology.org/2026.eacl-long.253/
- **url_code**: https://github.com/nsivaku/dart
- **summary_zh**: 《DART: Leveraging Multi-Agent Disagreement for Tool Recruitment in Multimodal Reasoning》属于ACL系会场相关工作，重点涉及多智能体/Agent、工具调用。英文摘要要点：Specialized visual tools can augment large langua…

## 15. From Sound to Symptom: Real-Time Respiratory Signal Understanding for Conversational Healthcare Agents
- **venue/year**: SIGDIAL · 2026
- **url_page**: https://aclanthology.org/2026.sigdial-1.39/
- **url_code**: https://github.com/paulleamy/AMI-Cough-Annotations
- **summary_zh**: 本文面向心血管/心内科、心内×临床对话场景，提出/研究「From Sound to Symptom: Real-Time Respiratory Signal Understanding for Conversational Healthcare Agents」，技术焦点包括多智能体/Agent、对话系统。论文核心工作可概括为：Cough events…

## 16. MIND: Unified Inquiry and Diagnosis RL with Criteria Grounded Clinical Supports for Psychiatric Consultation
- **venue/year**: EMNLP · 2026
- **url_page**: https://arxiv.org/abs/2603.03677v2
- **url_code**: https://github.com/Lingxi-mental-health/MIND
- **summary_zh**: 《MIND: Unified Inquiry and Diagnosis RL with Criteria Grounded Clinical Supports for Psychiatric Consultation》属于过程奖励×临床对话相关工作，重点涉及对话/问诊/追问、检索/知识增强。英文摘要要点：Psychiatric consultatio…

## 17. Dialogue to Question Generation for Evidence-based Medical Guideline Agent Development
- **venue/year**: ML4H · 2025
- **url_page**: https://proceedings.mlr.press/v297/ji26a.html
- **url_code**: https://github.com/Jerryji007/Dialogue2Questions-ML4H2025
- **summary_zh**: 《Dialogue to Question Generation for Evidence-based Medical Guideline Agent Development》发表于Workshop/其它会场，重点涉及多智能体/Agent、对话/问诊/追问、指南/决策支持。英文摘要要点：Evidence-based medicine ({EBM}) i…

## 18. MultiAgentESC: A LLM-based Multi-Agent Collaboration Framework for Emotional Support Conversation
- **venue/year**: EMNLP · 2025
- **url_page**: https://aclanthology.org/2025.emnlp-main.232/
- **url_code**: https://github.com/MindIntLab-HFUT/MultiAgentESC
- **summary_zh**: 《MultiAgentESC: A LLM-based Multi-Agent Collaboration Framework for Emotional Support Conversation》发表于ACL系会场、ACL/EMNLP残余扫街，重点涉及多智能体/Agent、对话/问诊/追问、大模型。英文摘要要点：Abstract The develo…

## 19. CoVaPh: A Vision-Language Multi-Agent Dialogue System for Tool-Augmented Pharmacogenetic Reasoning and Personalized Guidance
- **venue/year**: IWSDS · 2026
- **url_page**: https://aclanthology.org/2026.iwsds-1.42/
- **url_code**: （无）
- **summary_zh**: 本篇（IWSDS）入选短名单，相关信号：工具增强；工具调用/增强。英文摘要要点：The post-pandemic healthcare labor crisis has intensified the demand for accessible, high-precision pharmaceutical care. To meet this cha…

## 20. D$^2$-RAG: Dual-Decision Retrieval-Augmented Generation via Multi-Dimensional Uncertainty and Utility-Aware Decoding
- **venue/year**: Findings-ACL · 2026
- **url_page**: https://aclanthology.org/2026.findings-acl.754/
- **url_code**: https://github.com/zakelawen/d--rag
- **summary_zh**: 《D$^2$-RAG: Dual-Decision Retrieval-Augmented Generation via Multi-Dimensional Uncertainty and Utility-Aware Decoding》发表于ACL系会场、ACL/EMNLP残余扫街，重点涉及检索/知识增强、安全/不确定/隐私。英文摘要要点：Abstra…

## 21. GraphDx: A Cost-Aware Knowledge-Enhanced Multi-Agent Framework for Sequential Diagnosis
- **venue/year**: Findings-ACL · 2026
- **url_page**: https://aclanthology.org/2026.findings-acl.1092/
- **url_code**: https://github.com/ossiver-GEL/GraphDx
- **summary_zh**: 本篇（Findings-ACL）入选 ProMed 相邻短名单，相关信号：交互式诊断；诊断Agent/医院；顶会NLP。英文摘要要点：Sequential diagnosis requires balancing diagnostic accuracy against resource costs through iterative informati…

## 22. Faithful Medical Dialogue Generation Using Homo-Heterogeneous Exemplar-based In-Context Knowledge Grounding
- **venue/year**: LREC · 2026
- **url_page**: https://aclanthology.org/2026.lrec-1.4/
- **url_code**: https://github.com/Georgetown-IR-Lab/QuickUMLS
- **summary_zh**: 本文面向主动追问/预问诊场景，提出/研究「Faithful Medical Dialogue Generation Using Homo-Heterogeneous Exemplar-based In-Context Knowledge Grounding」，技术焦点包括对话系统。论文核心工作可概括为：The growing reliance on t…

## 23. Med-SRAF: A Multi-Agent Framework for Medical Reasoning via Semantic Routing and Agentic Fusion
- **venue/year**: Findings-ACL · 2026
- **url_page**: https://aclanthology.org/2026.findings-acl.1895/
- **url_code**: （无）
- **summary_zh**: 本篇（Findings-ACL）入选短名单，相关信号：Agent主题；语义路由多Agent；多Agent临床工作流；顶会。英文摘要要点：While Retrieval-Augmented Generation (RAG) has become a standard paradigm for mitigating hallucinations in La…

## 24. SOAPTriage: SOAP-Guided Multi-View Clinical Text Modeling Framework for Automated ESI Prediction
- **venue/year**: ACL · 2026
- **url_page**: https://aclanthology.org/2026.acl-long.1608/
- **url_code**: https://github.com/xiaoyaoiii/SOAPTriage
- **summary_zh**: 本篇（ACL）入选短名单，相关信号：主题命中；SOAPTriage·SOAP结构→急诊ESI；顶会；急诊分诊加分；核心点名。英文摘要要点：Emergency departments (ED) rely on the Emergency Severity Index (ESI) to assess patient acuity and prioritiz…

## 25. ASD-iLLM:An Intervention Large Language Model for Autistic Children based on Real Clinical Dialogue Intervention Dataset
- **venue/year**: Findings-EMNLP · 2025
- **url_page**: https://aclanthology.org/2025.findings-emnlp.427/
- **url_code**: https://github.com/Shuzhong-Lai/ASD-iLLM
- **summary_zh**: 本文面向主动追问/预问诊场景，提出/研究「ASD-iLLM:An Intervention Large Language Model for Autistic Children based on Real Clinical Dialogue Intervention Dataset」，技术焦点包括对话系统。论文核心工作可概括为：Currently, l…

## 26. MIND: Towards Immersive Psychological Healing with Multi-Agent Inner Dialogue
- **venue/year**: Findings-EMNLP · 2025
- **url_page**: https://aclanthology.org/2025.findings-emnlp.499/
- **url_code**: https://github.com/X-D-Lab/MindChat
- **summary_zh**: 《MIND: Towards Immersive Psychological Healing with Multi-Agent Inner Dialogue》属于ACL系会场相关工作，重点涉及多智能体/Agent、对话/问诊/追问。英文摘要要点：Mental health issues are worsening in today’s competit…

## 27. From Scanning Guidelines to Action: A Robotic Ultrasound Agent with LLM-Based Reasoning
- **venue/year**: MICCAI · 2026
- **url_page**: https://papers.miccai.org/miccai-2026/0406-Paper2814.html
- **url_code**: https://github.com/yuan-12138/RUSSAgent
- **summary_zh**: 《From Scanning Guidelines to Action: A Robotic Ultrasound Agent with LLM-Based Reasoning》发表于医学影像/生医NLP专会、MICCAI、医视语言，重点涉及多智能体/Agent、指南/决策支持、多模态/影像。英文摘要要点：Robotic ultrasound offe…

## 28. AI Diagnostic Assistant (AIDA): A Predictive Model for Diagnoses from Health Records in Clinical Decision Support Systems
- **venue/year**: IJCAI · 2025
- **url_page**: https://www.ijcai.org/proceedings/2025/1098
- **url_code**: （无）
- **summary_zh**: 本篇（IJCAI）入选短名单，相关信号：主题命中；AIDA·病历诊断决策助手；指南/CDSS/路径/治疗规划主题命中；顶会；CDSS/决策支持加分；核心点名。英文摘要要点：AI Diagnostic Assistant (AIDA), a machine learning model designed to predict ICD-10 diagnos…

## 29. CLARITY: Clinical Assistant for Routing, Inference, and Triage
- **venue/year**: EMNLP · 2025
- **url_page**: https://aclanthology.org/2025.emnlp-industry.127/
- **url_code**: https://github.com/AnonymousSubmission996/CLARITY
- **summary_zh**: 本篇（EMNLP）入选短名单，相关信号：主题命中；CLARITY·路由/推理/分诊一体助手；分诊/急救/院前/路由主题命中；顶会；急诊分诊加分；核心点名。英文摘要要点：We present CLARITY (Clinical Assistant for Routing, Inference and Triage), an AI-driven platf…

## 30. Case2Flow: Bridging Patient Cases and Guideline Flowcharts through Multimodal Retrieval
- **venue/year**: EMNLP · 2026
- **url_page**: https://arxiv.org/abs/2608.26414v1
- **url_code**: https://github.com/JialeWei/Case2Flow
- **summary_zh**: 《Case2Flow: Bridging Patient Cases and Guideline Flowcharts through Multimodal Retrieval》发表于ACL系会场、ACL/EMNLP残余扫街，重点涉及检索/知识增强、指南/决策支持、多模态/影像。英文摘要要点：Medical guidelines encode rich…

## 31. Greench-v1: distilling SLMs on Greenwashing Detection
- **venue/year**: NLP4ECOLOGY · 2026
- **url_page**: https://aclanthology.org/2026.nlp4ecology-1.8/
- **url_code**: （无）
- **summary_zh**: 《Greench-v1: distilling SLMs on Greenwashing Detection》属于过程奖励×临床对话、弃权/转诊×分诊相关工作，重点涉及医学AI方法。英文摘要要点：Validating greenwashing claims in environmental, social, and governance (ESG) r…

## 32. LLM-Based Multi-Agent System with Retrieval-Augmented Generation for Medical Care Planning Generation in Sickle Cell Disease
- **venue/year**: PROPOR · 2026
- **url_page**: https://aclanthology.org/2026.propor-2.16/
- **url_code**: （无）
- **summary_zh**: 《LLM-Based Multi-Agent System with Retrieval-Augmented Generation for Medical Care Planning Generation in Sickle Cell Disease》发表于Workshop/其它会场，重点涉及多智能体/Agent、检索/知识增强、大模型。英文摘要要点：…

## 33. TabMedQA: From Structured Data to Question-Answer Datasets in Early Clinical Decision-Making
- **venue/year**: CL4HEALTH · 2026
- **url_page**: https://aclanthology.org/2026.cl4health-1.6/
- **url_code**: https://github.com/iai-group/cl4health-tabmedqa
- **summary_zh**: 本篇（CL4HEALTH）入选 ProMed 相邻短名单，相关信号：MedQA交互对话；主动/交互诊断。英文摘要要点：The rising adoption of Large Language Models (LLMs) and Retrieval Augmented Generation (RAG) in clinical general pract…

## 34. ``What's Up, Doc?'': Analyzing How Users Seek Health Information in Large-Scale Conversational AI Datasets
- **venue/year**: Findings-EMNLP · 2025
- **url_page**: https://aclanthology.org/2025.findings-emnlp.125/
- **url_code**: https://github.com/yahskapar/HealthChat
- **summary_zh**: 本文面向心血管/心内科、心内×临床对话、主动追问/预问诊场景，提出/研究「``What's Up, Doc?'': Analyzing How Users Seek Health Information in Large-Scale Conversational AI Datasets」，技术焦点包括对话系统。论文核心工作可概括为：People are…

## 35. CoMMa: Contribution-Aware Medical Multi-Agents for Decentralized Oncology Decision Support
- **venue/year**: NeurIPS · 2026
- **url_page**: https://neurips.cc/virtual/2026/poster/153350
- **url_code**: （无）
- **summary_zh**: 本篇（NeurIPS）入选短名单，相关信号：主题命中；CoMMa·去中心化肿瘤多Agent决策；肿瘤学主题命中；顶会；肿瘤基准/Agent加分；核心点名。英文摘要要点：Recent multi-agent frameworks have shown promise for oncology decision sup port, yet most ass…

## 36. EMSDialog: Synthetic Multi-person Emergency Medical Service Dialogue Generation from Electronic Patient Care Reports via Multi-LLM Agents
- **venue/year**: Findings-ACL · 2026
- **url_page**: https://aclanthology.org/2026.findings-acl.1751/
- **url_code**: （无）
- **summary_zh**: 本篇（Findings-ACL）入选 Cardiology-Chat 心内优先阅读包，相关信号：心内信号(PDF)；LLM；多智能体/Agent；∩Dialogue；Agent切片。英文摘要要点：Conversational diagnosis prediction requires models to track evolving evidence …

## 37. From Medical Records to Diagnostic Dialogues: A Clinical-Grounded Approach and Dataset for Psychiatric Comorbidity
- **venue/year**: ICLR · 2026
- **url_page**: https://openreview.net/forum?id=sWWAZVHtke
- **url_code**: https://github.com/X-LANCE/PsyCo-Diagnosis
- **summary_zh**: 本文面向主动追问/预问诊场景，提出/研究「From Medical Records to Diagnostic Dialogues: A Clinical-Grounded Approach and Dataset for Psychiatric Comorbidity」，技术焦点包括对话系统。对医学大模型、临床对话或心血管相关应用具有参考价值，可结合…

## 38. Reliable Automated Triage in Spanish Clinical Notes: A Hybrid Framework for Risk-Aware HIV Suspicion Identification
- **venue/year**: BioNLP · 2026
- **url_page**: https://aclanthology.org/2026.bionlp-1.8/
- **url_code**: https://github.com/romorale/mil_uq_public
- **summary_zh**: 本篇（BioNLP）入选短名单，相关信号：主题命中；Spanish Triage·西语临床笔记风险分诊；分诊/急救/院前/路由主题命中；顶会；核心点名。英文摘要要点：Standard clinical Natural Language Processing (NLP) benchmarks often yield inflated metrics by…

## 39. Confidence-Based Response Abstinence: Improving LLM Trustworthiness via Activation-Based Uncertainty Estimation
- **venue/year**: UNCERTAINLP · 2025
- **url_page**: https://aclanthology.org/2025.uncertainlp-main.16/
- **url_code**: （无）
- **summary_zh**: 《Confidence-Based Response Abstinence: Improving LLM Trustworthiness via Activation-Based Uncertainty Estimation》属于弃权/转诊×分诊相关工作，重点涉及安全/不确定。英文摘要要点：We propose a method for confide…

## 40. Listening to Patients: Detecting and Mitigating Patient Misreport in Medical Dialogue System
- **venue/year**: Findings-ACL · 2025
- **url_page**: https://aclanthology.org/2025.findings-acl.135/
- **url_code**: （无）
- **summary_zh**: 本文面向心血管/心内科、心内×临床对话、主动追问/预问诊场景，提出/研究「Listening to Patients: Detecting and Mitigating Patient Misreport in Medical Dialogue System」，技术焦点包括对话系统。论文核心工作可概括为：Medical Dialogue Systems…

## 41. MedAgentBoard: Benchmarking Multi-Agent Collaboration with Conventional Methods for Diverse Medical Tasks
- **venue/year**: NeurIPS · 2025
- **url_page**: https://proceedings.neurips.cc/paper_files/paper/2025/hash/d59aa09699530c00d4b875a883876641-Abstract-Datasets_and_Benchmarks_Track.html
- **url_code**: https://github.com/yhzhu99/MedAgentBoard
- **summary_zh**: 本篇（NeurIPS）入选短名单，相关信号：Agent主题；MedAgentBoard；多Agent临床工作流；顶会。英文摘要要点：The rapid advancement of Large Language Models (LLMs) has stimulated interest in multi-agent collaboration for …

## 42. A Self-Evolving LLM Agent Framework for Role-Based Norm Compliance in Healthcare
- **venue/year**: Findings-ACL · 2026
- **url_page**: https://aclanthology.org/2026.findings-acl.1133/
- **url_code**: （无）
- **summary_zh**: 《A Self-Evolving LLM Agent Framework for Role-Based Norm Compliance in Healthcare》属于弃权/转诊×分诊相关工作，重点涉及多智能体/Agent。英文摘要要点：Large language models (LLMs) are increasingly proposed as …

## 43. ATPO: ADAPTIVE TREE POLICY OPTIMIZATION FOR MULTI-TURN MEDICAL DIALOGUE
- **venue/year**: ICLR · 2026
- **url_page**: https://openreview.net/forum?id=2bv3B8B9bl
- **url_code**: https://github.com/Quark-Medical/ATPO
- **summary_zh**: 本篇（ICLR）入选 ProMed 相邻短名单，相关信号：医患对话；交互式诊断。英文摘要要点：Effective information seeking in multi-turn medical dialogues is critical for accurate diagnosis, especially when dealing with inc…

## 44. DermAgent: A Multi-Tool Agentic System for Evidence-Grounded and Traceable Reasoning in Dermatology
- **venue/year**: MICCAI · 2026
- **url_page**: https://papers.miccai.org/miccai-2026/0263-Paper0286.html
- **url_code**: https://github.com/YizeezLiu/DermAgent
- **summary_zh**: 本篇（MICCAI）入选短名单，相关信号：工具增强；DermAgent；多工具Agent；非医学降权。英文摘要要点：. Dermatological diagnosis requires integrating fine-grained vi sual perception with expert clinical knowledge. Althoug…

## 45. DialogGuard: Multi-Agent Psychosocial Safety Evaluation Interface of Sensitive LLM Responses
- **venue/year**: ACL · 2026
- **url_page**: https://aclanthology.org/2026.acl-demo.19/
- **url_code**: https://github.com/lhannnn/dialogguard-web
- **summary_zh**: 《DialogGuard: Multi-Agent Psychosocial Safety Evaluation Interface of Sensitive LLM Responses》发表于ACL系会场、ACL/EMNLP残余扫街，重点涉及多智能体/Agent、对话/问诊/追问、评测基准/共享任务。英文摘要要点：Abstract LLM-based…

## 46. Experience is the Teacher: Reusing Atomic Thoughts from LLMs to Improve Medical Dialogue
- **venue/year**: Findings-ACL · 2026
- **url_page**: https://aclanthology.org/2026.findings-acl.957/
- **url_code**: https://github.com/VioletAmethystLunar/Atomic-Thoughts-Medical-Dialogue
- **summary_zh**: 本文面向主动追问/预问诊场景，提出/研究「Experience is the Teacher: Reusing Atomic Thoughts from LLMs to Improve Medical Dialogue」，技术焦点包括对话系统。论文核心工作可概括为：With the remarkable performance of large lan…

## 47. Knowing When to Abstain: Medical LLMs Under Clinical Uncertainty
- **venue/year**: EACL · 2026
- **url_page**: https://aclanthology.org/2026.eacl-long.291/
- **url_code**: https://github.com/sravanthi6m/MedAbstain
- **summary_zh**: 本篇（EACL）入选短名单，相关信号：主题命中；临床不确定时知何时弃答；校准/不确定性/弃答/共形/转诊；顶会；弃答/转诊加分；核心点名。英文摘要要点：Current evaluation of large language models (LLMs) overwhelmingly prioritizes accuracy; however, in r…

## 48. Teaching LLMs to Recommend and Defer in Underrepresented Epilepsy Care
- **venue/year**: NeurIPS · 2026
- **url_page**: https://neurips.cc/virtual/2026/poster/154539
- **url_code**: https://github.com/roychowdhuryresearch/Manana
- **summary_zh**: 本篇（NeurIPS）入选短名单，相关信号：主题命中；Epilepsy-Defer·代表性不足癫痫护理LLM；顶会；脑血管/癫痫加分；核心点名。英文摘要要点：Specialist epilepsy expertise is scarce in resource-constrained settings, making LLM-based decisio…

## 49. AI Psychiatrist Assistant: An LLM-based Multi-Agent System for Depression Assessment from Clinical Interviews
- **venue/year**: ML4H · 2025
- **url_page**: https://proceedings.mlr.press/v297/greene26a.html
- **url_code**: https://github.com/trendscenter/ai-psychiatrist
- **summary_zh**: 《AI Psychiatrist Assistant: An LLM-based Multi-Agent System for Depression Assessment from Clinical Interviews》发表于Workshop/其它会场，重点涉及多智能体/Agent、精神健康、大模型。英文摘要要点：Depression is one …

## 50. MAM: Modular Multi-Agent Framework for Multi-Modal Medical Diagnosis via Role-Specialized Collaboration
- **venue/year**: Findings-ACL · 2025
- **url_page**: https://aclanthology.org/2025.findings-acl.1298/
- **url_code**: https://github.com/yczhou001/MAM
- **summary_zh**: 本篇（Findings-ACL）入选短名单，相关信号：Agent主题；多Agent临床工作流；模块化多Agent诊断；顶会。英文摘要要点：Recent advancements in medical Large Language Models (LLMs) have showcased their powerful reasoning and diag…

## 51. A Clinical Guideline-Grounded Hybrid Agentic Framework for Holistic Epilepsy Management
- **venue/year**: MICCAI · 2026
- **url_page**: https://papers.miccai.org/miccai-2026/0004-Paper2852.html
- **url_code**: https://github.com/khoapham154/epi_guide
- **summary_zh**: 本篇（MICCAI）入选短名单，相关信号：主题命中；Epilepsy-Guideline·癫痫指南混合智能体；指南/CDSS/路径/治疗规划主题命中；顶会；指南接地加分；核心点名。英文摘要要点：. Epilepsy is a chronic neurological disorder requiring multi faceted management…

## 52. EmoRes: Toward Adaptive Psychological Support via User-Agnostic Benchmark and Topic-Mining Agent
- **venue/year**: Findings-ACL · 2026
- **url_page**: https://aclanthology.org/2026.findings-acl.416/
- **url_code**: （无）
- **summary_zh**: 《EmoRes: Toward Adaptive Psychological Support via User-Agnostic Benchmark and Topic-Mining Agent》属于过程奖励×临床对话相关工作，重点涉及多智能体/Agent、评测基准。英文摘要要点：Large language models exhibit signif…

## 53. EmplifAI: a Fine-grained Dataset for Japanese Empathetic Medical Dialogues in 28 Emotion Labels
- **venue/year**: IJCNLP · 2025
- **url_page**: https://aclanthology.org/2025.ijcnlp-long.61/
- **url_code**: （无）
- **summary_zh**: 本文面向主动追问/预问诊场景，提出/研究「EmplifAI: a Fine-grained Dataset for Japanese Empathetic Medical Dialogues in 28 Emotion Labels」，技术焦点包括对话系统。论文核心工作可概括为：This paper introduces EmplifAI, a Jap…

## 54. NLP4Health: Multilingual Clinical Dialogue Summarization and QA with mT5 and LoRA
- **venue/year**: NLPAI4HEALTH · 2025
- **url_page**: https://aclanthology.org/2025.nlpai4health-main.10/
- **url_code**: https://github.com/roymoutushi/NLP4Health
- **summary_zh**: 本文面向主动追问/预问诊场景，提出/研究「NLP4Health: Multilingual Clinical Dialogue Summarization and QA with mT5 and LoRA」，技术焦点包括对话系统。论文核心工作可概括为：In this work, we present NLP4Health, a unified and …

## 55. Benchmarking Reinforcement Learning Algorithms for ICU Ventilator Settings: An Interpretable and Probabilistic Patient Environment for Doctor Agents
- **venue/year**: AAAI · 2026
- **url_page**: https://ojs.aaai.org/index.php/AAAI/article/view/39081
- **url_code**: https://github.com/timothy0203/aaai-26-rl-benchmark-icu-ventilator-patient-environment
- **summary_zh**: 本文面向心血管/心内科、心内×临床对话场景，提出/研究「Benchmarking Reinforcement Learning Algorithms for ICU Ventilator Settings: An Interpretable and Probabilistic Patient Environment for Doctor Agents」…

## 56. Evaluating the Pre-Consultation Ability of LLMs using Diagnostic Guidelines
- **venue/year**: EACL · 2026
- **url_page**: https://aclanthology.org/2026.eacl-industry.6/
- **url_code**: https://github.com/seemdog/EPAG
- **summary_zh**: 本篇（EACL）入选 ProMed 相邻短名单，相关信号：预问诊；病史采集/预问诊；顶会NLP。英文摘要要点：We introduce EPAG, a benchmark dataset and framework designed for evaluating the pre-consultation ability of LLMs using di…

## 57. MED-COPILOT: A Medical Assistant Powered by GraphRAG and Similar Patient Case Retrieval
- **venue/year**: ACL · 2026
- **url_page**: https://aclanthology.org/2026.acl-demo.49/
- **url_code**: https://huggingface.co/spaces/shuhengc/MED-COPILOT
- **summary_zh**: 本篇（ACL）入选短名单，相关信号：主题命中；MED-COPILOT医学助手；草稿/门户/Copilot/医患沟通；顶会；Copilot加分；核心点名。英文摘要要点：Clinical decision-making requires synthesizing heterogeneous evidence, including patient histo…

## 58. MedAgentGym: A Scalable Agentic Training Environment for Code-Centric Reasoning in Biomedical Data Science
- **venue/year**: ICLR · 2026
- **url_page**: https://openreview.net/forum?id=jHDZEUgS4r
- **url_code**: https://github.com/wshi83/MedAgentGym
- **summary_zh**: 《MedAgentGym: A Scalable Agentic Training Environment for Code-Centric Reasoning in Biomedical Data Science》属于Agent工具×临床工作流相关工作，重点涉及多智能体/Agent。英文摘要要点：We introduce MedAgentGym, a…

## 59. SQUCS at MEDIQA-SYNUR 2026: A Multi-Agent Open Source LLM System for Nursing Observation Extraction
- **venue/year**: ClinicalNLP · 2026
- **url_page**: https://aclanthology.org/2026.clinicalnlp-1.14/
- **url_code**: （无）
- **summary_zh**: 本文面向心血管/心内科、心内×临床对话场景，提出/研究「SQUCS at MEDIQA-SYNUR 2026: A Multi-Agent Open Source LLM System for Nursing Observation Extraction」，技术焦点包括多智能体/Agent。论文核心工作可概括为：Clinical nursing doc…

## 60. AI Hospital: Benchmarking Large Language Models in a Multi-agent Medical Interaction Simulator
- **venue/year**: COLING · 2025
- **url_page**: https://aclanthology.org/2025.coling-main.680/
- **url_code**: https://github.com/LibertFan/AI_Hospital
- **summary_zh**: 本篇（COLING）入选短名单，相关信号：Agent主题；AI Hospital；多Agent临床工作流；顶会；核心必读。英文摘要要点：Artificial intelligence has significantly revolutionized healthcare, particularly through large language mode…

## 61. An Offline Mobile Conversational Agent for Mental Health Support: Learning from Emotional Dialogues and Psychological Texts with Student-Centered Evaluation
- **venue/year**: IJCNLP · 2025
- **url_page**: https://aclanthology.org/2025.ijcnlp-long.191/
- **url_code**: https://huggingface.co/SIR-Lab/EmoSApp
- **summary_zh**: 《An Offline Mobile Conversational Agent for Mental Health Support: Learning from Emotional Dialogues and Psychological Texts with Student-Centered Evaluation》发表于Workshop/其它会场，重点…

## 62. BioDiscoveryAgent: An AI Agent for Designing Genetic Perturbation Experiments
- **venue/year**: ICLR · 2025
- **url_page**: https://openreview.net/forum?id=HAwZGLcye3
- **url_code**: https://github.com/snap-stanford/BioDiscoveryAgent
- **summary_zh**: 《BioDiscoveryAgent: An AI Agent for Designing Genetic Perturbation Experiments》发表于通才ML会议、ICLR，重点涉及多智能体/Agent。英文摘要要点：Agents based on large language models have shown great potent…

## 63. Can an Individual Manipulate the Collective Decisions of Multi-Agents?
- **venue/year**: EMNLP · 2025
- **url_page**: https://aclanthology.org/2025.emnlp-main.611/
- **url_code**: （无）
- **summary_zh**: 《Can an Individual Manipulate the Collective Decisions of Multi-Agents?》发表于ACL系会场、ACL/EMNLP残余扫街，重点涉及多智能体/Agent。英文摘要要点：Abstract Individual Large Language Models (LLMs) have demon…

## 64. LLMs Can Simulate Standardized Patients via Agent Coevolution
- **venue/year**: ACL · 2025
- **url_page**: https://aclanthology.org/2025.acl-long.846/
- **url_code**: https://github.com/ZJUMAI/EvoPatient
- **summary_zh**: 本篇（ACL）入选 Cardiology-Chat 心内优先阅读包，相关信号：心内信号(PDF)；多智能体/Agent；∩Dialogue；Agent切片。英文摘要要点：Training medical personnel using standardized patients (SPs) remains a complex challenge, re…

## 65. MEDSAGE: Enhancing Robustness of Medical Dialogue Summarization to ASR Errors with LLM-generated Synthetic Dialogues
- **venue/year**: AAAI · 2025
- **url_page**: https://ojs.aaai.org/index.php/AAAI/article/view/34518
- **url_code**: （无）
- **summary_zh**: 本篇（AAAI）入选短名单，相关信号：主题命中；MEDSAGE抗ASR错误对话摘要；ASR错误临床影响；顶会；语音/ASR加分；核心点名。英文摘要要点：Guo et al. 2024), which limits their application in high stake domains where correctness of the summa…

## 66. OG-RAG: Ontology-grounded retrieval-augmented generation for large language models
- **venue/year**: EMNLP · 2025
- **url_page**: https://aclanthology.org/2025.emnlp-main.1674/
- **url_code**: https://github.com/microsoft/ograg2
- **summary_zh**: 本篇（EMNLP）入选短名单，相关信号：主题命中；OG-RAG本体 grounding RAG；医学KG/本体/实体链接；顶会；GraphRAG/KG-RAG加分；UMLS/SNOMED/本体加分；核心点名。英文摘要要点：While LLMs are widely used for generic tasks like question answeri…

## 67. TreeRAG: Unleashing the Power of Hierarchical Storage for Enhanced Knowledge Retrieval in Long Documents
- **venue/year**: Findings-ACL · 2025
- **url_page**: https://aclanthology.org/2025.findings-acl.20/
- **url_code**: https://github.com/gusye1234/nano-graphrag
- **summary_zh**: 《TreeRAG: Unleashing the Power of Hierarchical Storage for Enhanced Knowledge Retrieval in Long Documents》属于ACL系会场相关工作，重点涉及检索/知识增强。英文摘要要点：When confronting long document informat…

## 68. Agentic AI Architectures for SOAP Note Generation
- **venue/year**: BioNLP · 2026
- **url_page**: https://aclanthology.org/2026.bionlp-1.61/
- **url_code**: （无）
- **summary_zh**: 本篇（BioNLP）入选 Cardiology-Chat 心内优先阅读包，相关信号：心内信号(PDF)；多智能体/Agent；对话/问诊；∩Dialogue；Agent切片。英文摘要要点：Clinical documentation places significant time demands on medical professionals, co…

## 69. Beyond the Individual: Virtualizing Multi-Disciplinary Reasoning for Clinical Intake via Collaborative Agents
- **venue/year**: Findings-ACL · 2026
- **url_page**: https://aclanthology.org/2026.findings-acl.789/
- **url_code**: https://github.com/HovChen/Aegle
- **summary_zh**: 本篇（Findings-ACL）入选短名单，相关信号：Agent主题；多学科虚拟会诊；顶会。英文摘要要点：The initial outpatient consultation is critical for clinical decision-making, yet it is often conducted by a single physicia…

## 70. CAP: A Source-Grounded Proposition Scaffold for Faithful Clinical Dialogue-to-Note Generation
- **venue/year**: BioNLP · 2026
- **url_page**: https://aclanthology.org/2026.bionlp-1.46/
- **url_code**: https://github.com/sallyy1/CAP-ACL-2026
- **summary_zh**: 本篇（BioNLP）入选短名单，相关信号：主题命中；CAP·溯源命题支架对话→病历；顶会；对话摘要/SOAP加分；核心点名。英文摘要要点：Clinical dialogue-to-note generation is challenging because clinically salient evidence is noisy, distribute…

## 71. Strong Reasoning Isn't Enough: Evaluating Evidence Elicitation in Interactive Diagnosis
- **venue/year**: Findings-ACL · 2026
- **url_page**: https://aclanthology.org/2026.findings-acl.1372/
- **url_code**: https://github.com/zhao-zy15/RareArena
- **summary_zh**: 本篇（Findings-ACL）入选 ProMed 相邻短名单，相关信号：主动诊断；MedQA交互对话；交互式诊断；主动/交互诊断；顶会NLP。英文摘要要点：Interactive medical consultation requires an agent to proactively elicit missing clinical evidence…

## 72. Error Detection in Medical Note through Multi Agent Debate
- **venue/year**: BioNLP · 2025
- **url_page**: https://aclanthology.org/2025.bionlp-1.12/
- **url_code**: （无）
- **summary_zh**: 《Error Detection in Medical Note through Multi Agent Debate》发表于医学影像/生医NLP专会、BioNLP、ACL专会，重点涉及多智能体/Agent、病历/摘要/报告、分割/检测/分类。英文摘要要点：Large Language Models (LLMs) have ap proached hu…