Metraipolis › archive › dc_6408fb65601a
Interrupting the Loop: Periodic Subject Changes Raise Judged Surprise and Connection in Base Language Models — arXiv:2608.19893
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.19893
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T11:39:19.122Z
- SHA-256 of the archived text
4a09fdd704512cc77ad5e93be5d7ac7bfa59a8e7acb20643cf2465e7db99196b- Stable URL of this page
https://metraipolis.com/documento/dc_6408fb65601a
How to cite
Interrupting the Loop: Periodic Subject Changes Raise Judged Surprise and Connection in Base Language Models — arXiv:2608.19893 — arXiv (en), https://arxiv.org/abs/2608.19893 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6408fb65601a (Metraipolis archive dc_6408fb65601a, sha256 4a09fdd704512cc77ad5e93be5d7ac7bfa59a8e7acb20643cf2465e7db99196b)
BibTeX
@misc{dc6408fb65601a,
title = {Interrupting the Loop: Periodic Subject Changes Raise Judged Surprise and Connection in Base Language Models — arXiv:2608.19893},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.19893},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6408fb65601a; sha256 4a09fdd704512cc77ad5e93be5d7ac7bfa59a8e7acb20643cf2465e7db99196b},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.19893. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Where does the novelty a base language model produces with no task come from, and what can an LLM judge of a long stream actually see? We dismantle a cognitively inspired generation loop over 24 conditions on three base models. Most of its effect lives in one operation: a new subject injected every few hundred tokens (an interruption) into a stream whose literal repetition is damped (habituation). We judge windows of generated text only, with the premise as the unit (n=10) and a judge measured for repeatability, against a second judge family and against human readers. Under that protocol the interruption raises judged surprise by 1.2 to 1.4 points and connection by 0.8 over habituation alone. A connective that asks for continuity hurts; a bare paragraph break adds nothing detectable on fresh text; a reset context does at least as well as a kept one; and a pre-registered replication on new premises confirms the primary contrast. Three things the window judge could not see changed the first version of this study, and we think they are of general use. The judge scores the experimenter's injected sentence as the model's own. A fixed rotation of injected sentences makes the model replay its earlier segments from beyond the judge's horizon, and the judge scores the replay as surprise and connection (65-80% of post-interruption windows at periods 150-300). And the local gains do not compose: no arm produces an integrated document. The salience monitor, the in-loop judge, memory across interruptions and a judge-gated Review run with a gate that opens add nothing. On a problem with a verifier (online bin packing), the interruption multiplies valid, distinct candidate heuristics three- to fourfold without raising the quality of the best. We report an evaluation protocol for long generation and a controlled characterization of a simple intervention, not a mechanism of creativity. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-21 in cs.CL). Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2608.19893 — 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 2026-10-18T11:38:27.915Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-21?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).