Metraipolis › archive › dc_034e88038153

When History Fails to Become Experience: Action Calibration in Language Agents — arXiv:2610.02769

Source
arXiv
Original
https://arxiv.org/abs/2610.02769
License (as archived)
CC0-1.0
Archived
2026-10-05T22:50:15.153Z
SHA-256 of the archived text
7de683b9e450dbab10afc9558441c752be2609eaa374c650a96151a53554c2ca
Stable URL of this page
https://metraipolis.com/documento/dc_034e88038153

How to cite

When History Fails to Become Experience: Action Calibration in Language Agents — arXiv:2610.02769 — arXiv, https://arxiv.org/abs/2610.02769 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_034e88038153 (Metraipolis archive dc_034e88038153, sha256 7de683b9e450dbab10afc9558441c752be2609eaa374c650a96151a53554c2ca)
BibTeX
@misc{dc034e88038153,
  title = {When History Fails to Become Experience: Action Calibration in Language Agents — arXiv:2610.02769},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.02769},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_034e88038153; sha256 7de683b9e450dbab10afc9558441c752be2609eaa374c650a96151a53554c2ca},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.02769. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Language agents should draw on prior attempts and environmental feedback to improve subsequent decisions within the same task. However, providing additional interaction history can sometimes reduce task success, suggesting that agents do not consistently use this information effectively. To investigate this limitation, we examine how agents use history. We find that history improves task completion overall, yet much of this benefit persists even when past actions are shuffled. Disrupting the correspondence between actions and observations causes only a modest decline in task success. We therefore hypothesize that agents do not reliably connect past actions with their outcomes when deciding how to proceed. To test this hypothesis, we explicitly label each returned observation as the outcome of the preceding action. This simple annotation improves task success and reduces next-action repetition without introducing new environmental information. Building on this insight, we introduce a learned calibrator that explicitly reassesses past actions and selectively records experience to guide subsequent decisions, improving task success beyond outcome labeling alone.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.02769 — 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 2026-10-05T22:45:55.795Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).