Metraipolis › archive › dc_ddbbc5648fdd

STEP: Soft Trajectory Estimation of Prediction Correctness via Per-Step Probability Scoring — arXiv:2610.34537

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.34537
License (as archived): CC0-1.0
Archived: 2026-10-27T08:11:35.263Z
SHA-256 of the archived text: f35b778645fad284820f06ecf6552d2db81a4ee0b0295ef70add14bf74f1f489
Stable URL of this page: https://metraipolis.com/documento/dc_ddbbc5648fdd

How to cite

STEP: Soft Trajectory Estimation of Prediction Correctness via Per-Step Probability Scoring — arXiv:2610.34537 — arXiv (en), https://arxiv.org/abs/2610.34537 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ddbbc5648fdd (Metraipolis archive dc_ddbbc5648fdd, sha256 f35b778645fad284820f06ecf6552d2db81a4ee0b0295ef70add14bf74f1f489)

BibTeX

@misc{dcddbbc5648fdd,
  title = {STEP: Soft Trajectory Estimation of Prediction Correctness via Per-Step Probability Scoring — arXiv:2610.34537},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.34537},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ddbbc5648fdd; sha256 f35b778645fad284820f06ecf6552d2db81a4ee0b0295ef70add14bf74f1f489},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.34537. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Estimating the correctness of large language model (LLM) predictions without access to ground truth is critical for reliable deployment in high-stakes settings. Existing approaches, such as entropy and perplexity, operate solely on final output statistics, discarding the rich information encoded in intermediate reasoning steps. To address this limitation, we propose STEP (Soft Trajectory Estimation of Prediction correctness), a training-free framework that scores each reasoning step via lightweight forward passes to obtain a soft probability distribution qi over candidate answers, from which we derive a set of soft trajectory features capturing per-step confidence and its progression toward the final answer. Extensive experiments on various benchmarks demonstrate consistent and substantial improvements, particularly in AUROC, over output-based baselines and hard-indicator trajectory methods. Notably, Temporally-Weighted Consistency, which assigns higher importance to later reasoning steps where answer distributions have stabilized, contributes the largest gains, highlighting the importance of temporal structure in reasoning trajectories. These results further validate the advantage of preserving continuous probability signals over binary discretization.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.34537 — 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 2026-10-27T07:56:00.230Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).