Metraipolis › archive › dc_f480d4aa707e

Thinking effort aligns between humans and reasoning models in abductive reasoning — arXiv:2609.01867

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.01867
License (as archived)
CC0-1.0
Archived
2026-10-17T03:26:57.841Z
SHA-256 of the archived text
94c933175d956fbbe12666af5ae41046f36cdd299c124dbb42976cd456eeb7df
Stable URL of this page
https://metraipolis.com/documento/dc_f480d4aa707e

How to cite

Thinking effort aligns between humans and reasoning models in abductive reasoning — arXiv:2609.01867 — arXiv (en), https://arxiv.org/abs/2609.01867 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f480d4aa707e (Metraipolis archive dc_f480d4aa707e, sha256 94c933175d956fbbe12666af5ae41046f36cdd299c124dbb42976cd456eeb7df)
BibTeX
@misc{dcf480d4aa707e,
  title = {Thinking effort aligns between humans and reasoning models in abductive reasoning — arXiv:2609.01867},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.01867},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f480d4aa707e; sha256 94c933175d956fbbe12666af5ae41046f36cdd299c124dbb42976cd456eeb7df},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.01867. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

A major question in cognitive modeling concerns the behavioral alignment between large language models and humans across linguistic and non-linguistic tasks. Unlike standard LLMs, large reasoning models (LRMs) are optimized with reinforcement learning from verifiable rewards, encouraging correct solutions to reasoning tasks rather than preference-aligned responses. Recent work (de Varda et al., 2025) investigates the cost of thinking in humans and LRMs by comparing human reaction times with model reasoning traces across a range of reasoning tasks. We isolate this alignment by turning to abductive reasoning: unlike deductive tasks, its difficulty cannot be inferred from formal structure and offers no shortcuts a model could exploit to mimic effort without genuine search, providing firmer ground for empirical claims of shared effort. We find further evidence of alignment between LRM and human reasoning effort, as well as evidence that models and humans tend to make similar errors. Finally, we show that decoding methods that let models explore multiple reasoning paths increase alignment in reasoning cost between humans and LRMs across the three models tested.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-03 in cs.AI).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.01867 — 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 2026-10-17T03:13:12.487Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-03?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).