OmniHallu: Unified Hallucination Detection for Cross-Modal Comprehension and Generation in Multimodal Large Language Models

2026-09-10

AI Summary

OmniHallu框架通过多代理架构和模态特定专家验证，解决了多模态大语言模型在跨模态理解和生成任务中的幻觉检测问题。

Abstract

While Multimodal Large Language Models (MLLMs) have achieved remarkable progress across diverse tasks, they suffer from hallucinations where generated outputs contradict or misrepresent input semantics. Existing research typically addresses hallucination detection within a single modality or task type, limiting generalizability. We introduce OmniHallu, a unified hallucination detection framework spanning both comprehension and generation tasks across image, video, and audio modalities. We contribute OmniHallu-Bench, a 10,000-sample benchmark with claim-level human annotations covering six cross-modal tasks: image-to-text (I2T), video-to-text (V2T), audio-to-text (A2T), text-to-image (T2I), text-to-video (T2V), and text-to-audio (T2A). Our multi-agent architecture decomposes model outputs into atomic claims, verifies them through modality-specific experts, and aggregates evidence via structured reasoning. We further propose a preference-optimized trainable verifier that approximates the multi-agent decision boundary, reducing expert calls by 66% with minimal performance loss. Extensive experiments reveal a consistent modality-dependent performance gradient and provide fine-grained insights into cross-modal hallucination patterns.

Problem

• hallucination detection
• multimodal large language models
• cross-modal tasks

Innovation

• Unified Hallucination Detection
• Cross-Modal Comprehension and Generation
• Multi-Agent Architecture
• Preference-Optimized Verifier
• OmniHallu-Bench