Metraipolis › archive › dc_94007df2b410

Semantic Reasoning Denoising: Correcting Language Model Reasoning with Semantic Operators — arXiv:2608.22090

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.22090
License (as archived)
CC0-1.0
Archived
2026-10-18T07:42:37.666Z
SHA-256 of the archived text
f666cbf984e680a0181d4b220d5ed9ced686c19c938aa24f8464fe9375f98e61
Stable URL of this page
https://metraipolis.com/documento/dc_94007df2b410

How to cite

Semantic Reasoning Denoising: Correcting Language Model Reasoning with Semantic Operators — arXiv:2608.22090 — arXiv (en), https://arxiv.org/abs/2608.22090 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_94007df2b410 (Metraipolis archive dc_94007df2b410, sha256 f666cbf984e680a0181d4b220d5ed9ced686c19c938aa24f8464fe9375f98e61)
BibTeX
@misc{dc94007df2b410,
  title = {Semantic Reasoning Denoising: Correcting Language Model Reasoning with Semantic Operators — arXiv:2608.22090},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.22090},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_94007df2b410; sha256 f666cbf984e680a0181d4b220d5ed9ced686c19c938aa24f8464fe9375f98e61},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.22090. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language models can produce fluent reasoning traces whose local semantic errors propagate to an incorrect conclusion, while unconstrained self-correction may preserve, amplify, or introduce errors. Existing diffusion language models provide iterative refinement, but usually define noise as token masking or replacement rather than as errors in the reasoning process. We present Semantic Reasoning Denoising (SRD), an operatorized Markov denoising method for natural-language reasoning trajectories. SRD represents semantic noise with executable error operators that describe the error type, its location, and the corrupted and repaired propositions. Composing these operators constructs progressively noisier states. During training, the model learns to identify the semantic noise active in the current trajectory and to reconstruct the paired adjacent lower-noise state. During inference, noise-level-aware denoising repeatedly predicts an inverse operator and checks whether it is applicable, so each executed update makes a localized move toward a stable trajectory. Across six in-domain benchmarks spanning mathematics, code, knowledge, and commonsense, SRD improves the strongest same backbone baseline by 3.2 points on average. On seven cross-dataset transfer targets, it remains competitive with Llama-3-8B-Instruct and improves the strongest Qwen3-8B baseline average by 2.9 points. Analyses of noise sources, objectives, and denoising depth further show that structured semantic-noise prediction and iterative operator execution are central to the improvement.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-25 in cs.AI).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.22090 — 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 2026-10-18T07:37:52.854Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).