Metraipolis › archive › dc_7b48584830dc

Failure Profiling and Reachable Trajectory Selection for Reasoning Distillation — arXiv:2611.09042

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.09042
License (as archived)
CC0-1.0
Archived
2026-11-09T22:34:46.619Z
SHA-256 of the archived text
102cbc2585bc6a1f2430d8cfac151a68ebb0075531cee9924e81eba1df202be7
Stable URL of this page
https://metraipolis.com/documento/dc_7b48584830dc

How to cite

Failure Profiling and Reachable Trajectory Selection for Reasoning Distillation — arXiv:2611.09042 — arXiv (en), https://arxiv.org/abs/2611.09042 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7b48584830dc (Metraipolis archive dc_7b48584830dc, sha256 102cbc2585bc6a1f2430d8cfac151a68ebb0075531cee9924e81eba1df202be7)
BibTeX
@misc{dc7b48584830dc,
  title = {Failure Profiling and Reachable Trajectory Selection for Reasoning Distillation — arXiv:2611.09042},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.09042},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7b48584830dc; sha256 102cbc2585bc6a1f2430d8cfac151a68ebb0075531cee9924e81eba1df202be7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.09042. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Training smaller language models on reasoning solutions generated by a larger teacher model is the prevailing approach for transferring mathematical reasoning capabilities. However, the student tends to overfit the output distribution of the teacher and generalizes poorly when its own behavior at inference time deviates from the training data. Recent methods attempt to bridge this gap by incorporating error data, yet the corrective trajectories still reflect the reasoning patterns of the teacher, which the student cannot reliably reproduce, leaving the distribution mismatch largely unresolved. We propose a method, Failure Profiling and Reachable Trajectory Selection (FPRT), built on the insight that effective training data should be jointly determined by what the student fails on and what it can realistically learn. Our method first profiles the failure behavior of the student on each problem through repeated sampling, characterizing it in terms of error rate and error consistency to reveal qualitatively distinct failure modes. Each mode demands a fundamentally different training signal. The diagnosed mode then guides candidate generation from the teacher via distinct prompting strategies tailored to each failure type. Among the correct candidates, our method selects the one whose initial reasoning steps after the point of divergence are most reproducible by the student. These first corrective steps constitute the primary bottleneck, while the remaining steps follow an already corrected context and can therefore be continued more readily. This contrasts with standard reranking approaches, which score the trajectory as a whole and thereby dilute the signal at the critical reasoning transition. Experiments demonstrate that our method achieves the SOTA performance across the five benchmarks on both student backbones.

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); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2611.09042 — 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 2026-11-09T22:30:47.315Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).