Metraipolis › archive › dc_bf6eead03f7b
(V)LMs generalize beyond surface co-occurrence: Evidence from cross-modal number agreement — arXiv:2609.00443
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.00443
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T06:33:21.583Z
- SHA-256 of the archived text
2a8e1f538753e3be357ebfe6e177db6fa4b662b4c8a1ab429b6222d277079593- Stable URL of this page
https://metraipolis.com/documento/dc_bf6eead03f7b
How to cite
(V)LMs generalize beyond surface co-occurrence: Evidence from cross-modal number agreement — arXiv:2609.00443 — arXiv (en), https://arxiv.org/abs/2609.00443 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bf6eead03f7b (Metraipolis archive dc_bf6eead03f7b, sha256 2a8e1f538753e3be357ebfe6e177db6fa4b662b4c8a1ab429b6222d277079593)
BibTeX
@misc{dcbf6eead03f7b,
title = {(V)LMs generalize beyond surface co-occurrence: Evidence from cross-modal number agreement — arXiv:2609.00443},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.00443},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bf6eead03f7b; sha256 2a8e1f538753e3be357ebfe6e177db6fa4b662b4c8a1ab429b6222d277079593},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.00443. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Language models learn about grammatical number primarily from co-occurrence, and show frequency effects as a result---sometimes taken to indicate that they do not learn abstract ``rules'', and are instead dependent on specific lexical items. Testing generalization with text stimuli alone cannot settle this debate, since distributional cues (is/are, this/these) easily give number away. We instead use cross-modal generalization as a tool to investigate abstractions in LMs that can also accept visual inputs (VLMs), restricting the evidence that diagnoses number to an extra-linguistic modality. We teach VLMs pairs of new nouns by adding new embeddings and only updating them during learning, comparing conditions where number is diagnosed by visual cues alone against ones where it is disambiguated by text. Across behavior, representational dynamics, and causal mechanisms, we find non-trivial evidence for cross-modal generalization across both exposure conditions, and that linguistic vs. extra-linguistic cue conditions are treated in similar ways in the internal mechanisms of the model. This suggests that statistical learners like VLMs can generalize beyond surface-level co-occurrence and show genuine abstraction-compatible behavior. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-02 in cs.AI). Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.00443 — 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 2026-10-17T06:15:12.779Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).