Metraipolis › archive › dc_a70080b293b7
TurnSight: Turn-Level Hindsight Self-Distillation for Tool-Integrated Reasoning — arXiv:2608.04007
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.04007
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T12:19:44.610Z
- SHA-256 of the archived text
6cda54712f57ebb0601d31cb9e56efbeb4475056f5f17b5c43cace6b9337a9e3- Stable URL of this page
https://metraipolis.com/documento/dc_a70080b293b7
How to cite
TurnSight: Turn-Level Hindsight Self-Distillation for Tool-Integrated Reasoning — arXiv:2608.04007 — arXiv (en), https://arxiv.org/abs/2608.04007 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a70080b293b7 (Metraipolis archive dc_a70080b293b7, sha256 6cda54712f57ebb0601d31cb9e56efbeb4475056f5f17b5c43cace6b9337a9e3)
BibTeX
@misc{dca70080b293b7,
title = {TurnSight: Turn-Level Hindsight Self-Distillation for Tool-Integrated Reasoning — arXiv:2608.04007},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.04007},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a70080b293b7; sha256 6cda54712f57ebb0601d31cb9e56efbeb4475056f5f17b5c43cace6b9337a9e3},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.04007. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Tool-Integrated Reasoning (TIR) enables LLMs to solve complex tasks through iterative tool interactions. However, existing reinforcement learning methods often rely on trajectory-level supervision, limiting fine-grained credit assignment in long-horizon TIR scenarios. On-policy self-distillation offers denser signals through teacher branches with privileged context, but existing approaches typically derive such context from ground-truth answers or retrieved skills, which may not reflect the states actually visited by the agent. Moreover, token-level supervision fails to capture the turn-level structure of tool interactions. To address this, we propose TurnSight, a turn-level hindsight self-distillation framework that derives supervision directly from execution-conditioned hindsight. It then constructs multiple hindsight views with different lookahead horizons and selects reliable supervision through cross-horizon directional agreement. Finally, the selected hindsight signal is normalized across sibling rollouts and used to adaptively modulate RL advantages while preserving their original optimization direction. Extensive experiments on three benchmarks demonstrate the effectiveness of TurnSight. Our codes are available at this https URL . Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-05 in cs.AI). Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2608.04007 — 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 2026-10-31T12:05:53.109Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).