Metraipolis › archive › dc_9ba38bf43eeb

Automatic or Controlled? Repetition Priming Reveals Divergent Processing in Base LLMs, Instruct LLMs, and Humans — arXiv:2608.14681

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.14681
License (as archived)
CC0-1.0
Archived
2026-10-18T17:52:02.279Z
SHA-256 of the archived text
9f388a167baf16ffe48aea32bc1e8dd3a40edc4d9f73b4c8ed81e8bf3b3a8373
Stable URL of this page
https://metraipolis.com/documento/dc_9ba38bf43eeb

How to cite

Automatic or Controlled? Repetition Priming Reveals Divergent Processing in Base LLMs, Instruct LLMs, and Humans — arXiv:2608.14681 — arXiv (en), https://arxiv.org/abs/2608.14681 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9ba38bf43eeb (Metraipolis archive dc_9ba38bf43eeb, sha256 9f388a167baf16ffe48aea32bc1e8dd3a40edc4d9f73b4c8ed81e8bf3b3a8373)
BibTeX
@misc{dc9ba38bf43eeb,
  title = {Automatic or Controlled? Repetition Priming Reveals Divergent Processing in Base LLMs, Instruct LLMs, and Humans — arXiv:2608.14681},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.14681},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9ba38bf43eeb; sha256 9f388a167baf16ffe48aea32bc1e8dd3a40edc4d9f73b4c8ed81e8bf3b3a8373},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.14681. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Words recur constantly in natural language use, yet it remains unclear whether language models reactivate prior representations or re-evaluate repeated words afresh, and whether post-training changes this default behavior. We apply repetition priming (Shiffrin and Schneider, 1977) to 15 models across five model families (1.5B-14B parameters) in two tasks, semantic categorization and cloze completion, with matched human experiments using identical stimuli. We find that base models exhibit automatic processing: they show immediate facilitation that remains stable across lags, partially survives context removal, and correlates with attention to prior occurrences. Instruct models exhibit controlled processing: their facilitation decays with lag, collapses without expected context, and reverses to interference at larger scales. Within the Qwen 2.5 family, this dissociation increases monotonically with model scale, suggesting that post-training progressively alters repetition processing. Humans show a hybrid profile, with lag-sensitive facilitation resembling instruct models but without interference, suggesting that neither model type fully captures human cognition. Our findings reveal a qualitative shift in how language models process repeated information after post-training and provide mechanistic evidence for the divergence between model behaviors.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-18 in cs.AI).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.14681 — 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 2026-10-18T16:29:13.940Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).