Metraipolis › archive › dc_4a18e45936cc

Reasoning-Aware Cross-Modal Association Unlearning in Multimodal Large Reasoning Models — arXiv:2611.06168

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.06168
License (as archived)
CC0-1.0
Archived
2026-11-05T18:17:46.612Z
SHA-256 of the archived text
47180017b6a4071621d19cadb2565e46a178b02d7fcba9763db40f3ac618f760
Stable URL of this page
https://metraipolis.com/documento/dc_4a18e45936cc

How to cite

Reasoning-Aware Cross-Modal Association Unlearning in Multimodal Large Reasoning Models — arXiv:2611.06168 — arXiv (en), https://arxiv.org/abs/2611.06168 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4a18e45936cc (Metraipolis archive dc_4a18e45936cc, sha256 47180017b6a4071621d19cadb2565e46a178b02d7fcba9763db40f3ac618f760)
BibTeX
@misc{dc4a18e45936cc,
  title = {Reasoning-Aware Cross-Modal Association Unlearning in Multimodal Large Reasoning Models — arXiv:2611.06168},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.06168},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4a18e45936cc; sha256 47180017b6a4071621d19cadb2565e46a178b02d7fcba9763db40f3ac618f760},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.06168. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Multimodal large reasoning models (MLRMs) integrate visual inputs with textual and parametric knowledge, allowing sensitive information to be accessed through cross-modal associations during reasoning. This raises privacy concerns when a visual input should no longer provide access to such information. Machine unlearning can remove this influence without retraining the model from scratch. However, multimodal unlearning methods mainly target samples, concepts, or knowledge content, without separating the knowledge from the cross-modal association through which it is accessed. In reasoning models, target information may disappear from the final answer while remaining in intermediate reasoning, or be replaced by incorrect substitute knowledge. We therefore study cross-modal association unlearning, where the forget target is a designated visual-to-knowledge association: the model should suppress target-bearing behavior under the designated visual context while preserving access to the same knowledge in permitted contexts. We propose Visual-Keyed Counterfactual Rebinding (VKCR), a parameter-efficient framework for cross-modal association editing. VKCR first localizes the edit to a compact association-conditioned parameter subspace. It then reduces support for the designated target at task outputs and counterfactually rebinds the affected reasoning behavior by discouraging target-bearing or unsupported continuations while favoring valid non-sensitive alternatives. We further introduce CLEAR-R and MLLMU-R to evaluate target leakage, hallucinated reassociation, reasoning preservation, and modality-specific utility. Experiments show that VKCR substantially reduces image-conditioned target leakage and hallucinated reassociation, including across heterogeneous task formulations, while preserving non-sensitive visual reasoning and permitted text-only knowledge.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.06168 — 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 2026-11-05T18:12:49.049Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).