Metraipolis › archive › dc_5550fa77c4ee

Shared relations, specific choices: conditional sensitivity and competing pathways in language models — arXiv:2611.01809

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.01809
License (as archived)
CC0-1.0
Archived
2026-11-03T18:23:25.813Z
SHA-256 of the archived text
6b0fdfed7e8c1c04c9dff443b60d09a79d57d0f42a4fe903886b2e4fd6d31bd5
Stable URL of this page
https://metraipolis.com/documento/dc_5550fa77c4ee

How to cite

Shared relations, specific choices: conditional sensitivity and competing pathways in language models — arXiv:2611.01809 — arXiv (en), https://arxiv.org/abs/2611.01809 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5550fa77c4ee (Metraipolis archive dc_5550fa77c4ee, sha256 6b0fdfed7e8c1c04c9dff443b60d09a79d57d0f42a4fe903886b2e4fd6d31bd5)
BibTeX
@misc{dc5550fa77c4ee,
  title = {Shared relations, specific choices: conditional sensitivity and competing pathways in language models — arXiv:2611.01809},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.01809},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5550fa77c4ee; sha256 6b0fdfed7e8c1c04c9dff443b60d09a79d57d0f42a4fe903886b2e4fd6d31bd5},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.01809. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

How does a relational representation shared across queries drive a query-specific choice? Controlled entity-binding experiments with GPT-2 Small, Qwen2.5-1.5B and Llama-3.2-1B reveal a low-dimensional relational background that is predominantly shared across query roles and transfers to new names. Same-domain interactions persist when another attribute domain is queried. After removing this component, restoring independently trained relation directions brings output contrasts closer to their original values than sign-reversed or equal-norm random controls in all three models. Yet a transferable relational representation need not have a transferable functional sign. In Qwen2.5, fixing model weights, the training-mean relation vector and edit dose yields opposite selection effects across binary and lexical-mapping prompts with different answer vocabularies. A factorial contrast of downstream sensitivities across attribute-query combinations determines the shared direction's local signed selection gain. In a separate eight-condition protocol, mean local gain is positive in Qwen2.5 and negative in Llama3.2. Late-component clamps resolve opposing pathway contributions: in Llama3.2, negative transmission through selected attention-only paths is partially offset by positive transmission through paths traversing both selected attention and MLP sets. Source-coordinate analyses establish distributed factor coverage, hidden-receiver readouts reveal layout-dependent combination, and instance-geometry interventions separate response direction from amplitude. These findings show how conditional sensitivity and competing pathways give shared relational structure its effect on specific choices.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2611.01809 — 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 2026-11-03T18:10:07.941Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).