{"slug": "mode-rag-manifold-outlier-diagnosis-and-energy-based-retrieval-augmented", "title": "MODE-RAG: Manifold Outlier Diagnosis and Energy-based Retrieval-Augmented Generation Evaluation", "summary": "Researchers propose MODE-RAG, a multi-agent system using Variational Free Energy and attention states to dynamically gate interventions in multimodal retrieval-augmented generation, reducing hallucinations and logical fabrications. The system routes high-risk queries to five agents integrating Monte Carlo Tree Search and logit perturbations, and introduces the ModeVent evaluation subset from MultiVent. Experiments show improved robustness in M-RAG systems.", "body_md": "arXiv:2606.17449v1 Announce Type: new\nAbstract: While Multimodal Retrieval-Augmented Generation (M-RAG) enhances Large Vision-Language Models, it remains highly susceptible to cross-modal hallucinations, causal fabrications, and sycophancy. Furthermore, existing mitigation pipelines often face an intervention paradox: static rules tend to unnecessarily disrupt accurate generations, whereas leaving the multi-modal reasoning completely unguided allows existing mismatches to cascade into severe logical fabrications. To quantify and mitigate these hallucinations, we propose a Multi-Agent system, MODE-RAG, driven by Variational Free Energy (VFE) and internal attention states to dynamically gate interventions. High-risk queries are routed to five stage-specific agents, integrating Monte Carlo Tree Search (MCTS) for rigorous causal derivation and logit perturbations to penalize sycophancy. Dedicated Correction and Overseer agents ensure formatting stability and perform post-hoc factual verification. To objectively evaluate our approach, we introduce ModeVent, a challenging subset derived from the MultiVent dataset. Extensive experiments indicate that our system effectively reduces hallucination rates and logical fabrication, significantly improving the robustness of M-RAG systems.", "url": "https://wpnews.pro/news/mode-rag-manifold-outlier-diagnosis-and-energy-based-retrieval-augmented", "canonical_source": "https://arxiv.org/abs/2606.17449", "published_at": "2026-06-17 04:00:00+00:00", "updated_at": "2026-06-17 04:27:43.721229+00:00", "lang": "en", "topics": ["large-language-models", "ai-research", "ai-safety", "generative-ai", "artificial-intelligence"], "entities": ["MODE-RAG", "MultiVent", "ModeVent", "Monte Carlo Tree Search", "Variational Free Energy"], "alternates": {"html": "https://wpnews.pro/news/mode-rag-manifold-outlier-diagnosis-and-energy-based-retrieval-augmented", "markdown": "https://wpnews.pro/news/mode-rag-manifold-outlier-diagnosis-and-energy-based-retrieval-augmented.md", "text": "https://wpnews.pro/news/mode-rag-manifold-outlier-diagnosis-and-energy-based-retrieval-augmented.txt", "jsonld": "https://wpnews.pro/news/mode-rag-manifold-outlier-diagnosis-and-energy-based-retrieval-augmented.jsonld"}}