Metraipolis › archive › dc_2018f3b48ef5
Learning from Failures: Heterogeneous Graph Memory for Small Language Model Tool-Using Agents — arXiv:2609.28003
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.28003
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T10:57:32.581Z
- SHA-256 of the archived text
3b5034ac4d55c13457222792778b7c2a8a04fc55c63987e051baa11878c6c752- Stable URL of this page
https://metraipolis.com/documento/dc_2018f3b48ef5
How to cite
Learning from Failures: Heterogeneous Graph Memory for Small Language Model Tool-Using Agents — arXiv:2609.28003 — arXiv (en), https://arxiv.org/abs/2609.28003 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2018f3b48ef5 (Metraipolis archive dc_2018f3b48ef5, sha256 3b5034ac4d55c13457222792778b7c2a8a04fc55c63987e051baa11878c6c752)
BibTeX
@misc{dc2018f3b48ef5,
title = {Learning from Failures: Heterogeneous Graph Memory for Small Language Model Tool-Using Agents — arXiv:2609.28003},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.28003},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2018f3b48ef5; sha256 3b5034ac4d55c13457222792778b7c2a8a04fc55c63987e051baa11878c6c752},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.28003. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Small and medium-sized language models offer cost-effective executors for tool-using agents, making them attractive for local and large-scale deployment. However, in long-horizon and stateful environments, they often make structural errors such as missing required observations, performing premature writes, repeating failed calls, and violating action preconditions. These errors can lead to incorrect state updates, policy violations, and costly or irreversible consequences, making reliable tool execution a critical deployment challenge. Existing fine-tuning approaches require substantial data and computation, while flat memory may retrieve failed actions without preserving their causal context or safety conditions. In this paper, we propose FRESH, a Failure-aware Retrieval framework over Experience-Structured Heterogeneous graphs, which transforms historical successes and failures into structured external experience for tool-using agents. By explicitly modeling the dependencies among tasks, actions, errors, repairs, and execution conditions, FRESH helps frozen language models reuse reliable strategies, avoid recurring failures, and make safer decisions in stateful tool interactions. Experiments on $\tau$-Bench and AppWorld with multiple open-source models show that FRESH consistently improves task success and tool-use reliability over no-memory agents and representative memory-based baselines. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-24 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.28003 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-08T10:06:16.767Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-24?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).