Causal Evidence for Attention Head Imbalance in Modality Conflict Hallucination
The paper presents findings on modality-conflict hallucination in multimodal large language models (MLLMs). It identifies an imbalance in attention heads that either drive or resist hallucination during generation. The authors propose a new intervention method, MACI, which effectively reduces hallucinations while maintaining accuracy.
- ▪Modality-conflict hallucination occurs when MLLMs prioritize incorrect textual information over visual evidence.
- ▪The study reveals two groups of attention heads: those that drive hallucinations and those that resist them.
- ▪The proposed MACI intervention suppresses driving heads when conflicts are detected, leading to significant reductions in hallucinations.
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/2605.19250 |
| Publication time | Wed, 20 May 2026 00:00:00 -0400 |
| Retrieval time | 2026-05-20T04:04:59.484Z |
| Last seen | 2026-05-20T04:04:59.484Z |
| 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 | FIhRqm2St9Aa |
| 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:2605.19250 (cs) [Submitted on 19 May 2026] Title:Causal Evidence for Attention Head Imbalance in Modality Conflict Hallucination Authors:Jinrui Jiang, Zhangtai Wu, Zhen Wu, Xinyu Dai View a PDF of the paper titled Causal Evidence for Attention Head Imbalance in Modality Conflict Hallucination, by Jinrui Jiang and 2 other authors View PDF HTML (experimental) Abstract:Modality-conflict hallucination occurs when multimodal large language models (MLLMs) prioritize erroneous textual premises over contradictory visual evidence. To understand why visual evidence fails to prevail during generation, we take a mechanistic perspective and examine which internal components drive or resist this failure.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
