Metraipolis › archive › dc_cd11dff021a6

Mostly Reading: What Open-Book Scores Measure in Document-Level LoRA Adapters — arXiv:2610.31017

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.31017
License (as archived): CC0-1.0
Archived: 2026-10-23T15:15:16.377Z
SHA-256 of the archived text: 2e351e09c1f91785613a758ff635f2adc54271a608d8ffabde89cbc00fe73c36
Stable URL of this page: https://metraipolis.com/documento/dc_cd11dff021a6

How to cite

Mostly Reading: What Open-Book Scores Measure in Document-Level LoRA Adapters — arXiv:2610.31017 — arXiv (en), https://arxiv.org/abs/2610.31017 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cd11dff021a6 (Metraipolis archive dc_cd11dff021a6, sha256 2e351e09c1f91785613a758ff635f2adc54271a608d8ffabde89cbc00fe73c36)

BibTeX

@misc{dccd11dff021a6,
  title = {Mostly Reading: What Open-Book Scores Measure in Document-Level LoRA Adapters — arXiv:2610.31017},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.31017},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cd11dff021a6; sha256 2e351e09c1f91785613a758ff635f2adc54271a608d8ffabde89cbc00fe73c36},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.31017. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Single-document LoRA adapters have been proposed as units of knowledge to be merged and retrieved; yet open-book scoring, with the document in the prompt, can credit a document's own adapter (its specialist) for reading that document as readily as for holding it. Training one adapter for each of 150 Wikipedia articles on Llama-3.1-8B-Instruct, we find that open-book scores mostly measure a reading behaviour every adapter shares: a random adapter trained on another document recovers 82 % of the specialist's open-book advantage over the base model but only 2 % of its advantage on a cloze probe that asks for the same document's facts without providing the source text, a share we cannot distinguish from zero. Content does reach the weights, since the specialists beat the base model in cloze recall by 0.094, and the geometry of their updates follows the documents they read; yet, in a test of 30 queries, adapters chosen by weight-space proximity or text embedding are not distinguishable from random ones on either probe, although that test cannot exclude an advantage of about four fifths of what a query's own adapter gains over a random one. At 16 documents or more, correctly scaled merges recover 87 % to 89 % of the specialists' open-book advantage but only 12 % to 16 % of their cloze advantage, with intervals that run from below zero to at most 32 %, against 94 % for joint this http URL also report four silent errors, each of which produced a clean, wrong result.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.31017 — 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 2026-10-23T15:09:08.580Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).