# Better with Experience: Self-Evolving LLM Agents for Evidence-Grounded Health Community Notes

arXiv: https://arxiv.org/abs/2606.02215 · HTML: https://mlsanity.com/paper/2606.02215/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Medical and scientific domain agents, one of 16 directions in LLM agents and tool use.
- Builds on 20 of its 60 references that are in the map.
- First posted on arXiv 1 Jun 2026.

## Placed in

- LLM agents and tool use / [Medical and scientific domain agents](https://mlsanity.com/agents/medical-and-scientific-domain-agents/)
- Synthetic data and self-improvement for LLMs / [Medical synthetic data and agents](https://mlsanity.com/synth/medical-synthetic-data-and-agents/)

## Builds on (most cited first, 20 of 20)

- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [MemGPT: Towards LLMs as Operating Systems](https://mlsanity.com/paper/2310.08560/) (2023)
- [OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments](https://mlsanity.com/paper/2404.07972/) (2024)
- [ExpeL: LLM Agents Are Experiential Learners](https://mlsanity.com/paper/2308.10144/) (2023)
- [Zep: A Temporal Knowledge Graph Architecture for Agent Memory](https://mlsanity.com/paper/2501.13956/) (2025)
- [RAGEN: Understanding Self-Evolution in LLM Agents via Multi-Turn Reinforcement Learning](https://mlsanity.com/paper/2504.20073/) (2025)
- [Lingshu: A Generalist Foundation Model for Unified Multimodal Medical Understanding and Reasoning](https://mlsanity.com/paper/2506.07044/) (2025)
- [BALROG: Benchmarking Agentic LLM and VLM Reasoning On Games](https://mlsanity.com/paper/2411.13543/) (2024)
- [Evo-Memory: Benchmarking LLM Agent Test-time Learning with Self-Evolving Memory](https://mlsanity.com/paper/2511.20857/) (2025)
- [EvoTest: Evolutionary Test-Time Learning for Self-Improving Agentic Systems](https://mlsanity.com/paper/2510.13220/) (2025)
- [Supernotes: Driving Consensus in Crowd-Sourced Fact-Checking](https://mlsanity.com/paper/2411.06116/) (2024)
- [Birdwatch: Crowd Wisdom and Bridging Algorithms can Inform Understanding and Reduce the Spread of Misinformation](https://mlsanity.com/paper/2210.15723/) (2022)
- [Missci: Reconstructing Fallacies in Misrepresented Science](https://mlsanity.com/paper/2406.03181/) (2024)
- [Community notes reduce engagement with and diffusion of false information online](https://mlsanity.com/paper/2502.13322/) (2025)
- [Did the Roll-Out of Community Notes Reduce Engagement With Misinformation on X/Twitter?](https://mlsanity.com/paper/2307.07960/) (2023)
- [Beyond the Crowd: LLM-Augmented Community Notes for Governing Health Misinformation](https://mlsanity.com/paper/2510.11423/) (2025)
- [Understanding Fine-grained Distortions in Reports of Scientific Findings](https://mlsanity.com/paper/2402.12431/) (2024)
- [Community-based fact-checking reduces the spread of misleading posts on social media](https://mlsanity.com/paper/2409.08781/) (2024)
- [COMMUNITYNOTES: A Dataset for Exploring the Helpfulness of Fact-Checking Explanations](https://mlsanity.com/paper/2510.24810/) (2025)
- [GitSearch: Enhancing Community Notes Generation with Gap-Informed Targeted Search](https://mlsanity.com/paper/2602.08945/) (2026)

## Abstract (arXiv)

Large Language Model (LLM)-augmented Community Notes offer a scalable path for timely, evidence-grounded correction of health misinformation on social platforms. However, they still reset at every post, leaving useful correction experience from prior cases unused. We introduce EvoNote, an agentic framework that enables health Community Notes generation to self-evolve through an evolving experience memory of prior misinformation correction episodes. Its core is fine-grained credit assignment: EvoNote grounds trajectory-level feedback in health-specific note qualities and distills it into action-level memory for claim analysis, evidence acquisition, and note writing. We evaluate EvoNote on MM-HealthCN, a 1.2K-instance multimodal benchmark of user-flagged health posts with human-written Community Notes and crowd-derived helpfulness labels. Under a human-validated hierarchical utility judge, EvoNote-generated notes are preferred over corresponding human-written notes in 89.6% of cases; on a separate set of Needs More Ratings posts without a crowd helpfulness verdict, EvoNote produces helpful notes for 82.0% of cases. It also reduces time to an available candidate correction from multi-hour post-to-note delays to under 2 minutes on average. Analyses link these quality gains to stronger evidence use and reusable correction strategies, positioning self-evolving note generation as a promising paradigm for scalable health misinformation governance.