# DENSE: Distilling Agent Trajectories into Evidence-Grounded Shortcut Trees for Self-Refinement

arXiv: https://arxiv.org/abs/2609.21423 · HTML: https://mlsanity.com/paper/2609.21423/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 30 of its 38 references that are in the map.
- First posted on arXiv 18 Sep 2026.

## Builds on (most cited first, 30 of 30)

- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [WebArena: A Realistic Web Environment for Building Autonomous Agents](https://mlsanity.com/paper/2307.13854/) (2023)
- [SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering](https://mlsanity.com/paper/2405.15793/) (2024)
- [τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains](https://mlsanity.com/paper/2406.12045/) (2024)
- [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)
- [Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces](https://mlsanity.com/paper/2601.11868/) (2026)
- [When Can LLMs Actually Correct Their Own Mistakes? A Critical Survey of Self-Correction of LLMs](https://mlsanity.com/paper/2406.01297/) (2024)
- [Which Agent Causes Task Failures and When? On Automated Failure Attribution of LLM Multi-Agent Systems](https://mlsanity.com/paper/2505.00212/) (2025)
- [AgentRewardBench: Evaluating Automatic Evaluations of Web Agent Trajectories](https://mlsanity.com/paper/2504.08942/) (2025)
- [Tongyi DeepResearch Technical Report](https://mlsanity.com/paper/2510.24701/) (2025)
- [The MiniMax-M2 Series: Mini Activations Unleashing Max Real-World Intelligence](https://mlsanity.com/paper/2605.26494/) (2026)
- [AgentRx: Diagnosing AI Agent Failures from Execution Trajectories](https://mlsanity.com/paper/2602.02475/) (2026)
- [Reducing Cost of LLM Agents with Trajectory Reduction](https://mlsanity.com/paper/2509.23586/) (2025)
- [AgentProcessBench: Diagnosing Step-Level Process Quality in Tool-Using Agents](https://mlsanity.com/paper/2603.14465/) (2026)
- [DoVer: Intervention-Driven Auto Debugging for LLM Multi-Agent Systems](https://mlsanity.com/paper/2512.06749/) (2025)
- [Who&When Pro: Can LLMs Really Attribute Failures in AI Agents?](https://mlsanity.com/paper/2607.09996/) (2026)
- [Trajectory-Informed Memory Generation for Self-Improving Agent Systems](https://mlsanity.com/paper/2603.10600/) (2026)
- [A Self-Evolving Framework for Efficient Terminal Agents via Observational Context Compression](https://mlsanity.com/paper/2604.19572/) (2026)
- [ToolPRMBench: Evaluating and Advancing Process Reward Models for Tool-using Agents](https://mlsanity.com/paper/2601.12294/) (2026)
- [RealCritic: Towards Effectiveness-Driven Evaluation of Language Model Critiques](https://mlsanity.com/paper/2501.14492/) (2025)
- [On the Reliability of Computer Use Agents](https://mlsanity.com/paper/2604.17849/) (2026)
- [AgentDebugX: An Open-Source Toolkit for Failure Observability, Attribution, and Recovery in LLM Agents](https://mlsanity.com/paper/2607.18754/) (2026)
- [Training Language Model to Critique for Better Refinement](https://mlsanity.com/paper/2506.22157/) (2025)
- [GUIDE: Interpretable GUI Agent Evaluation via Hierarchical Diagnosis](https://mlsanity.com/paper/2604.04399/) (2026)
- [PRACT: Optimizing Principled Reasoning and Acting of LLM Agent](https://mlsanity.com/paper/2410.18528/) (2024)
- [Conversation for Non-verifiable Learning: Self-Evolving LLMs through Meta-Evaluation](https://mlsanity.com/paper/2601.21464/) (2026)
- [Fail-Fast, Restart-Smart: Early Failure Prediction and Restart for SWE Agentic Tasks](https://mlsanity.com/paper/2608.03222/) (2026)
- [Falsification, Not Exposure: An Internally Preregistered Placebo-Controlled Decomposition of Self-Repair Feedback in Frozen Small Code Models](https://mlsanity.com/paper/2606.31511/) (2026)
- [Long-Horizon Agent Trajectory Attribution: A Unified Benchmark and Fine-Grained Annotation Framework](https://mlsanity.com/paper/2608.06909/) (2026)

## Abstract (arXiv)

Online agent deployments accumulate execution trajectories at massive scale and behavioral diversity, for which predefined annotation criteria hardly exist. Extracting useful evidence therefore demands costly manual annotation or verifier signals that fail to scale, leaving valuable evidence buried among redundant, incomplete, and failed executions. This raises a question: without post-execution rewards or correctness labels, how can reusable experience be distilled from the trajectories themselves? To address this challenge, we introduce DENSE (Distilling Evidence from Nested Subtask Executions), which organizes trajectory-derived evidence into nested shortcut trees. By consolidating redundant attempts, identifying resolved subtasks, and retaining useful steps alongside outstanding requirements, DENSE transforms noisy execution traces into structured and reusable task-solving feedback. To evaluate whether such feedback helps agents retry the same task, we design REFIT, which measures success-rate changes between the initial attempt and feedback-guided retries. Among feedback methods without external outcome supervision, DENSE achieves the highest strict pass rate across four agent models on Terminal-Bench 2.1, improving over initial attempts by 7.12-21.81 percentage points with 19.0-43.6% fewer agent tokens on retries. In addition, on hard tasks DENSE consistently outperforms self-reflection in cumulative pass rate across multiple feedback iterations on all four models, demonstrating its strong potential for continual agent self-improvement.