# On the Fragility of Self-Improving Agents: Variance, Task Order, and Underspecification

arXiv: https://arxiv.org/abs/2608.18066 · HTML: https://mlsanity.com/paper/2608.18066/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Self-evolving LLM agents, one of 13 directions in Synthetic data and self-improvement for LLMs.
- Builds on 52 of its 71 references that are in the map.
- First posted on arXiv 18 Aug 2026.

## Placed in

- Synthetic data and self-improvement for LLMs / [Self-evolving LLM agents](https://mlsanity.com/synth/self-evolving-llm-agents/)

## Builds on (most cited first, 30 of 52)

- [Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks](https://mlsanity.com/paper/1703.03400/) (2017)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [WebArena: A Realistic Web Environment for Building Autonomous Agents](https://mlsanity.com/paper/2307.13854/) (2023)
- [τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains](https://mlsanity.com/paper/2406.12045/) (2024)
- [Calibrate Before Use: Improving Few-Shot Performance of Language Models](https://mlsanity.com/paper/2102.09690/) (2021)
- [MemGPT: Towards LLMs as Operating Systems](https://mlsanity.com/paper/2310.08560/) (2023)
- [Mind2Web: Towards a Generalist Agent for the Web](https://mlsanity.com/paper/2306.06070/) (2023)
- [OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments](https://mlsanity.com/paper/2404.07972/) (2024)
- [Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivity](https://mlsanity.com/paper/2104.08786/) (2021)
- [A-MEM: Agentic Memory for LLM Agents](https://mlsanity.com/paper/2502.12110/) (2025)
- [Large Language Models Cannot Self-Correct Reasoning Yet](https://mlsanity.com/paper/2310.01798/) (2023)
- [STaR: Bootstrapping Reasoning With Reasoning](https://mlsanity.com/paper/2203.14465/) (2022)
- [Evaluating Very Long-Term Conversational Memory of LLM Agents](https://mlsanity.com/paper/2402.17753/) (2024)
- [Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formatting](https://mlsanity.com/paper/2310.11324/) (2023)
- [Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory](https://mlsanity.com/paper/2504.19413/) (2025)
- [Large Language Models Can Self-Improve](https://mlsanity.com/paper/2210.11610/) (2022)
- [τ2-Bench: Evaluating Conversational Agents in a Dual-Control Environment](https://mlsanity.com/paper/2506.07982/) (2025)
- [Executable Code Actions Elicit Better LLM Agents](https://mlsanity.com/paper/2402.01030/) (2024)
- [MemoryBank: Enhancing Large Language Models with Long-Term Memory](https://mlsanity.com/paper/2305.10250/) (2023)
- [Automated Design of Agentic Systems](https://mlsanity.com/paper/2408.08435/) (2024)
- [Fine-Tuning Pretrained Language Models: Weight Initializations, Data Orders, and Early Stopping](https://mlsanity.com/paper/2002.06305/) (2020)
- [Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscation](https://mlsanity.com/paper/2503.11926/) (2025)
- [Agent Workflow Memory](https://mlsanity.com/paper/2409.07429/) (2024)
- [Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agents](https://mlsanity.com/paper/2505.22954/) (2025)
- [Titans: Learning to Memorize at Test Time](https://mlsanity.com/paper/2501.00663/) (2024)
- [State of What Art? A Call for Multi-Prompt LLM Evaluation](https://mlsanity.com/paper/2401.00595/) (2023)
- [How to train your MAML](https://mlsanity.com/paper/1810.09502/) (2018)
- [Revisiting Few-sample BERT Fine-tuning](https://mlsanity.com/paper/2006.05987/) (2020)
- [SkillWeaver: Web Agents can Self-Improve by Discovering and Honing Skills](https://mlsanity.com/paper/2504.07079/) (2025)
- [Synapse: Trajectory-as-Exemplar Prompting with Memory for Computer Control](https://mlsanity.com/paper/2306.07863/) (2023)

## Built on by (most cited first, 4 of 4)

- [S3Gym: Can LLMs Turn Self-Testing and Self-Judging into Self-Improvement?](https://mlsanity.com/paper/2608.31100/) (2026)
- [Not All Experience Belongs in the Weights: Component Routing for Self-Improving GUI Agents](https://mlsanity.com/paper/2610.01787/) (2026)
- [The Winner's Curse in LLM Self-Improvement Loops: Selection Noise, Lock-in, and Acceptance Rules](https://mlsanity.com/paper/2610.09239/) (2026)
- [Reels-RSI: Program-Level Self-Improvement for Code-Rendered Explainer Videos](https://mlsanity.com/paper/2610.14495/) (2026)

## Abstract (arXiv)

Memory-based self-improving agents--those that learn from an online stream of tasks and improve over time by maintaining a textual memory bank--have shown great promise in recent literature. However, the reliability aspects of these methods have been critically overlooked. In this work, we conduct a comprehensive re-evaluation of two memory-based methods, broadening the scope of evaluation along two axes: (1) including multiple self-improving runs to quantify variance, and (2) shuffling the tasks to investigate the effect of task order. Through these experiments, we make two observations that expose the fragility of current methods: First, agent evaluation is inherently noisy in complex environments and on multi-step tasks, and stacking a self-improving loop on top can further amplify this noise. Second, the agent's improvement is highly dependent on task order. Prior works often adopt default orderings that impose an implicit curriculum, acting as a hidden prerequisite for success. To better understand this fragility, we manually examine the agents' memory and hypothesize that task and environment underspecification contribute to this fragility. We validate this hypothesis by incorporating information that enables better specification, such as detailed rubrics and environment feedback, into the memory construction process. While this added information partially closes the performance degradation in previous experiments, significant gaps still remain, suggesting that other uncharacterized factors contribute to this fragility. Looking ahead, our work advocates for more rigorous evaluation protocols for self-improving agents by reporting results across multiple runs and stress-testing them under challenging conditions. Moreover, our findings on underspecification call for systems and interfaces that enable effective human oversight, preventing agents from failing in unforeseeable ways.