Safety Geometry Collapse in Multimodal LLMs and Adaptive Drift Correction
The paper discusses the safety challenges faced by multimodal large language models (MLLMs) in transferring safety capabilities across different input modalities. It introduces the concept of Safety Geometry Collapse, which occurs when models struggle to refuse harmful inputs due to modality-induced drift. The authors propose a method called ReGap to correct this drift and enhance the safety of MLLMs without sacrificing their general capabilities.
- ▪Multimodal large language models often fail to transfer safety capabilities learned in text to non-text inputs.
- ▪The study identifies a failure mode termed Safety Geometry Collapse, which compresses usable separation along refusal directions.
- ▪The proposed ReGap method adaptively corrects modality drift and improves multimodal safety during inference.
arXiv cs.AI files mainly under ai research. We currently carry 1,128 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | arXiv cs.AI |
| Canonical URL | https://arxiv.org/abs/2605.18104 |
| Publication time | Tue, 19 May 2026 00:00:00 -0400 |
| Retrieval time | 2026-05-19T04:04:57.272Z |
| Last seen | 2026-05-19T04:04:57.272Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | ACqj4cqV-usl |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Artificial Intelligence arXiv:2605.18104 (cs) [Submitted on 18 May 2026] Title:Safety Geometry Collapse in Multimodal LLMs and Adaptive Drift Correction Authors:Jiahe Guo, Xiangran Guo, Jiaxuan Chen, Weixiang Zhao, Yanyan Zhao, Yutai Hou, Qianchao Wang, Dandan Tu, Bing Qin View a PDF of the paper titled Safety Geometry Collapse in Multimodal LLMs and Adaptive Drift Correction, by Jiahe Guo and 8 other authors View PDF HTML (experimental) Abstract:Multimodal large language models (MLLMs) often fail to transfer safety capabilities learned in the text modality to semantically equivalent non-text inputs, revealing a persistent multimodal safety gap.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
