Metraipolis › archive › dc_b2ecaead0683

From Parameters to Answers: How LLMs Retrieve and Use Their Internal Knowledge — arXiv:2609.11859

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.11859
License (as archived)
CC0-1.0
Archived
2026-10-16T05:24:31.149Z
SHA-256 of the archived text
3622f4145b18bceb083c2062e1f8aeb32a64c8dd4961863a9ce1b93989d54af6
Stable URL of this page
https://metraipolis.com/documento/dc_b2ecaead0683

How to cite

From Parameters to Answers: How LLMs Retrieve and Use Their Internal Knowledge — arXiv:2609.11859 — arXiv (en), https://arxiv.org/abs/2609.11859 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b2ecaead0683 (Metraipolis archive dc_b2ecaead0683, sha256 3622f4145b18bceb083c2062e1f8aeb32a64c8dd4961863a9ce1b93989d54af6)
BibTeX
@misc{dcb2ecaead0683,
  title = {From Parameters to Answers: How LLMs Retrieve and Use Their Internal Knowledge — arXiv:2609.11859},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.11859},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b2ecaead0683; sha256 3622f4145b18bceb083c2062e1f8aeb32a64c8dd4961863a9ce1b93989d54af6},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.11859. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

How does a language model's dependence on query-routing information and target knowledge change as it answers a question? We study this question through layerwise interventions on the hidden state at the end of the question. Across Qwen, Llama, and Gemma, we compare country-continent questions with noun, adjective, and code answers while keeping several fitted measurements distinct. A pair-conditioned request direction describes which country is queried in natural single-country questions; a global request direction describes first- versus second-country requests in paired questions; separate selection candidates test control among contents already available in the hidden state. A diagnostic reanalysis of frozen Qwen natural-question states shows that the pair-conditioned direction grows stronger before interventions on it begin to alter later fitted knowledge, with this causal window opening while answer-supporting content is still forming. The paired three-model trajectories are not uniform: Gemma shows a partially overlapping mid-layer routing-content profile, whereas Llama has no sustained routing-effect window under the same gates. In the paired protocol, dependence on the global request direction decreases from fixed earlier to later layer sets while dependence on fitted content persists. A matched Qwen comparison shows that the pair-conditioned direction retains a late effect, so this operational handoff concerns the global fitted direction rather than all request information. These results separate early readability, natural strength, causal steering, and later content dependence.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-11 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.11859 — 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 2026-10-16T05:17:52.053Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).