Metraipolis › archive › dc_df32a4bb3417
Learning a Fact Is Not Learning How to Retrieve It — arXiv:2610.03251
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.03251
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-05T22:49:33.539Z
- SHA-256 of the archived text
25f619acc6d153d566b7dd28d078d8c223e5954628226adc9e1bee7ca5b5b842- Stable URL of this page
https://metraipolis.com/documento/dc_df32a4bb3417
How to cite
Learning a Fact Is Not Learning How to Retrieve It — arXiv:2610.03251 — arXiv, https://arxiv.org/abs/2610.03251 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_df32a4bb3417 (Metraipolis archive dc_df32a4bb3417, sha256 25f619acc6d153d566b7dd28d078d8c223e5954628226adc9e1bee7ca5b5b842)
BibTeX
@misc{dcdf32a4bb3417,
title = {Learning a Fact Is Not Learning How to Retrieve It — arXiv:2610.03251},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.03251},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_df32a4bb3417; sha256 25f619acc6d153d566b7dd28d078d8c223e5954628226adc9e1bee7ca5b5b842},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.03251. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
A model trained on "The capital of X is Y" may produce "Y" after "The capital of X is" but fail after "The capital of X:". We call these different ways of eliciting the same fact request forms. To separate learning a fact from retrieving it, we train two models in two stages. In the first stage (request-form training), one model sees each fact in five forms and the other sees the same facts only as statements. In the second stage (target-fact training), both receive identical training on new facts, all as statements. Both then retrieve the new facts almost equally well from statements, but differ sharply on other request forms. Thus, a model can learn how to retrieve through a request form before it learns the facts. To understand this difference, we examine the hidden state immediately before the answer, which we call the context state. When given two different request forms for the same fact, the model trained on five forms in stage one produces more similar context states than the model trained on statements alone in that stage. Changing this state at retrieval time can enable or prevent retrieval of an already learned fact, and the same effect transfers across facts and factual relations, such as capitals and currencies. To test its role during learning, we change the context state only during target-fact training. This intervention changes later retrieval without intervention at test time. Together, these results show that later retrieval depends on earlier request-form experience and the context state during fact learning. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.03251 — 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 2026-10-05T22:45:55.683Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).