Metraipolis › archive › dc_6e38285abc8b

The Error You See Is Not the Error You Made: Progression-aware Reasoning Origin for Reasoning Error Localization — arXiv:2609.33297

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.33297
License (as archived)
CC0-1.0
Archived
2026-10-07T20:56:39.103Z
SHA-256 of the archived text
1874f6cd497851471c819fc384e06ebbf6e659f8a342d6c50438549c1fa8c787
Stable URL of this page
https://metraipolis.com/documento/dc_6e38285abc8b

How to cite

The Error You See Is Not the Error You Made: Progression-aware Reasoning Origin for Reasoning Error Localization — arXiv:2609.33297 — arXiv (en), https://arxiv.org/abs/2609.33297 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6e38285abc8b (Metraipolis archive dc_6e38285abc8b, sha256 1874f6cd497851471c819fc384e06ebbf6e659f8a342d6c50438549c1fa8c787)
BibTeX
@misc{dc6e38285abc8b,
  title = {The Error You See Is Not the Error You Made: Progression-aware Reasoning Origin for Reasoning Error Localization — arXiv:2609.33297},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.33297},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6e38285abc8b; sha256 1874f6cd497851471c819fc384e06ebbf6e659f8a342d6c50438549c1fa8c787},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33297. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Verifying multi-step LLM reasoning requires more than determining whether a trace is correct: a useful verifier should identify where the reasoning first goes wrong. However, existing holistic methods provide little positional evidence, while forward sequential verification often treats the first rejected step as the error source. Under error propagation, this assumption can fail, since an earlier mistake may remain locally plausible and become observable only through its downstream consequences. We therefore rethink reasoning verification as a progression-aware error-source localization problem: rather than asking only where a reasoning trace first appears inconsistent, we ask which earlier step best explains how that inconsistency emerges along the trajectory. Based on this view, we propose Progression-aware Reasoning Origin (PRO), a training-free framework for first-error localization. PRO jointly models incoming support from the preceding context and outgoing compatibility with subsequent reasoning, selectively refines regions where these signals disagree, and finally performs detector-conditioned source attribution with intervention-based evidence to distinguish the true error origin from its propagated manifestations. We further formalize the gap between forward rejection and structural exposure, showing why incoming-side evidence alone is insufficient for reliable localization under error propagation. Experiments across open-form, medical, and structured reasoning tasks demonstrate consistent improvements over strong verification baselines, supporting progression-aware source attribution as a more faithful formulation of reasoning verification.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.33297 — 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 2026-10-07T20:29:01.210Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).