# Counterfactual Self-Evolving Agents for Evidence-Grounded Reasoning

arXiv: https://arxiv.org/abs/2609.32870 · HTML: https://mlsanity.com/paper/2609.32870/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 29 of its 36 references that are in the map.
- First posted on arXiv 26 Sep 2026.

## Builds on (most cited first, 29 of 29)

- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Self-Refine: Iterative Refinement with Self-Feedback](https://mlsanity.com/paper/2303.17651/) (2023)
- [MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI](https://mlsanity.com/paper/2311.16502/) (2023)
- [FinQA: A Dataset of Numerical Reasoning over Financial Data](https://mlsanity.com/paper/2109.00122/) (2021)
- [MedGemma Technical Report](https://mlsanity.com/paper/2507.05201/) (2025)
- [Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agents](https://mlsanity.com/paper/2505.22954/) (2025)
- [MedAgents: Large Language Models as Collaborators for Zero-shot Medical Reasoning](https://mlsanity.com/paper/2311.10537/) (2023)
- [MEM1: Learning to Synergize Memory and Reasoning for Efficient Long-Horizon Agents](https://mlsanity.com/paper/2506.15841/) (2025)
- [HoVer: A Dataset for Many-Hop Fact Extraction And Claim Verification](https://mlsanity.com/paper/2011.03088/) (2020)
- [A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligence](https://mlsanity.com/paper/2507.21046/) (2025)
- [MDAgents: An Adaptive Collaboration of LLMs for Medical Decision-Making](https://mlsanity.com/paper/2404.15155/) (2024)
- [Probabilities of Causation: Bounds and Identification](https://mlsanity.com/paper/1301.3898/) (2013)
- [Context-faithful Prompting for Large Language Models](https://mlsanity.com/paper/2303.11315/) (2023)
- [MedAgent-Pro: Towards Evidence-based Multi-modal Medical Diagnosis via Reasoning Agentic Workflow](https://mlsanity.com/paper/2503.18968/) (2025)
- [FuseMoE: Mixture-of-Experts Transformers for Fleximodal Fusion](https://mlsanity.com/paper/2402.03226/) (2024)
- [CORAL: Towards Autonomous Multi-Agent Evolution for Open-Ended Discovery](https://mlsanity.com/paper/2604.01658/) (2026)
- [DiagnosisArena: Benchmarking Diagnostic Reasoning for Large Language Models](https://mlsanity.com/paper/2505.14107/) (2025)
- [Zero-shot LLM-guided Counterfactual Generation: A Case Study on NLP Model Evaluation](https://mlsanity.com/paper/2405.04793/) (2024)
- [On Expert Estimation in Hierarchical Mixture of Experts: Beyond Softmax Gating Functions](https://mlsanity.com/paper/2410.02935/) (2024)
- [Data Augmentations for Improved (Large) Language Model Generalization](https://mlsanity.com/paper/2310.12803/) (2023)
- [Propose, Solve, Verify: Self-Play Through Formal Verification](https://mlsanity.com/paper/2512.18160/) (2025)
- [GSEM: Graph-based Self-Evolving Memory for Experience Augmented Clinical Reasoning](https://mlsanity.com/paper/2603.22096/) (2026)
- [Improving Clinical Diagnosis with Counterfactual Multi-Agent Reasoning](https://mlsanity.com/paper/2603.27820/) (2026)
- [Quadratic Gating Mixture of Experts: Statistical Insights into Self-Attention](https://mlsanity.com/paper/2410.11222/) (2024)
- [Massively Multimodal Foundation Models: A Framework for Capturing Interactions with Specialized Mixture-of-Experts](https://mlsanity.com/paper/2509.25678/) (2025)
- [Counterfactual Modeling with Fine-Tuned LLMs for Health Intervention Design and Sensor Data Augmentation](https://mlsanity.com/paper/2601.14590/) (2026)
- [MILM: Large Language Models for Multimodal Irregular Time Series with Informative Sampling](https://mlsanity.com/paper/2605.13711/) (2026)

## Abstract (arXiv)

Self-play proposer--solver methods improve reasoning by generating tasks and learning from verified solutions. However, for evidence-identifiable tasks, where case-specific evidence and domain knowledge determine a checkable answer, self-play requires generating plausible cases whose answers can be independently verified. We introduce counterfactual self-evolution, which generates counterfactual context for reconsidering the original case. A trainable Proposer constructs targeted evidence edits and describes potential outcome changes with causal explanations. We handcraft an expert-verified counterfactual instruction-tuning dataset to teach the Proposer to generate high-quality counterfactuals across a broad range of action--outcome scenarios. Each counterfactual instruction-tuning example specifies an edit within a defined category and explains its hypothesized causal effect on the decision, teaching the Proposer to reason systematically about what changes and why. We instruction-tune the Proposer on these examples, then formulate a fine-tuning reward that integrates feedback from the Solver and Verifier. Across diverse counterfactual scenarios, this reward favors high-quality counterfactuals and warranted revisions, while penalizing changes that overturn correct decisions. The counterfactual context aims to correct errors and strengthen confidence in correct decisions. Accepted counterfactuals accumulate in memory that supplies in-context evidence to the frozen Solver; the Solver adapts through evolving context rather than weight updates. We apply the framework to clinical reasoning, fact verification, and business reasoning. Our evaluation tracks performance over successive rounds as counterfactual memory grows, including transfer to harder cases. Our method achieves superior results across diverse frontier models.