Metraipolis › archive › dc_c4bb0e94073a
PG-SFT: Balancing Capability Acquisition and Retention in Offline Agent Fine-Tuning — arXiv:2610.00949
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.00949
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T12:03:09.973Z
- SHA-256 of the archived text
8e03c0cf196307eae5c3f182cc1ef389b22e66874eda497eeba034b63cde17e7- Stable URL of this page
https://metraipolis.com/documento/dc_c4bb0e94073a
How to cite
PG-SFT: Balancing Capability Acquisition and Retention in Offline Agent Fine-Tuning — arXiv:2610.00949 — arXiv (en), https://arxiv.org/abs/2610.00949 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c4bb0e94073a (Metraipolis archive dc_c4bb0e94073a, sha256 8e03c0cf196307eae5c3f182cc1ef389b22e66874eda497eeba034b63cde17e7)
BibTeX
@misc{dcc4bb0e94073a,
title = {PG-SFT: Balancing Capability Acquisition and Retention in Offline Agent Fine-Tuning — arXiv:2610.00949},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.00949},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c4bb0e94073a; sha256 8e03c0cf196307eae5c3f182cc1ef389b22e66874eda497eeba034b63cde17e7},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.00949. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Supervised fine-tuning (SFT) on offline agent trajectories is the standard approach for training specialized tool-using agents, but forcing models to imitate reasoning and actions token by token may harm other capabilities (e.g., general reasoning, tool calling, code generation) of the base model. In this work, we focus on studying \emph{how to better balance the trade-off between acquiring new capabilities and preserving existing ones during agent trace SFT}. By comparing several baselines in our setup, standard SFT improves the target benchmark while lowering several non-target benchmark scores; meanwhile, simply constraining distributional drift using KL penalty or limiting the update magnitude did not avoid this regression trend. Motivated by recent token-wise adaptive learning objectives, this work proposes \textbf{Privilege-Guided SFT (PG-SFT)} to leverage turn-level information gain of agent trajectories as an indicator to adjust supervision strength. PG-SFT yields a more favorable observed trade-off on the evaluated benchmarks, substantially reducing distributional drift and broad capability degradation at the cost of slight degradation in target-task performance. Our findings suggest that balancing the acquisition--retention trade-off depends not only on whether the model is anchored to its base behavior, but also on where and how strongly supervision should depart from that behavior.}
Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.00949 — 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 2026-10-06T10:21:27.897Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).