arXiv:2606. 17449v1 Announce Type: cross Abstract: While Multimodal Retrieval-Augmented Generation (M-RAG) enhances Large Vision-Language Models, it remains highly susceptible to cross-modal hallucinations, causal fabrications, and sycophancy.
arXiv:2601.22984v3 Announce Type: replace Abstract: Diagnosing failure patterns in Deep Research Agents (DRAs) remains a critical challenge. Existing benchmarks predominantly rely on end-to-end evalu...
OmniHallu is a unified framework for detecting hallucinations in multimodal large language models across both comprehension and generation tasks involving image, video, and audio modalities. It introduces OmniHallu-Bench, a 10,000-sample benchmark with claim-level human annotations for six cross-modal tasks (I2T, V2T, A2T, T2I, T2V, T2A). The system uses a multi‑agent architecture that decomposes outputs into atomic claims, verifies them with modality‑specific experts, and aggregates evidence through structured reasoning, while a preference‑optimized verifier reduces expert calls by 66% with minimal performance loss.
arXiv:2607. 07507v1 Announce Type: cross Abstract: Hallucinations in vision language models (VLMs) are commonly treated as semantic errors, yet they often arise from partial or ambiguous visual evidence.
The paper examines six inference-time hallucination mitigation methods applied to three large vision-language models across four benchmarks, including MMStar. It finds that reducing hallucination rates often comes at the cost of lower informativeness—such as decreased object recall, visual coverage, and response detail—and that gains on hallucination benchmarks do not consistently translate to improved performance on fine-grained perception and reasoning tasks. The authors argue that current evaluation protocols may overstate progress by favoring conservative generation, and propose that hallucination mitigation should be assessed as a trade-off among faithfulness, informativeness, and overall capability.
The paper proposes using the temporal volatility of internal attention mechanisms—measured by an unsupervised attention dispersion metric—as a diagnostic signal for hallucinations in large language models. It demonstrates that spikes in attention entropy within intermediate layers correlate with reasoning breakdowns, and shows statistically significant AUC improvements of up to +0.076 over output-based baselines on GSM8K and MATH-500 benchmarks using the Qwen2.5 model family.
arXiv:2606. 27596v1 Announce Type: cross Abstract: Large Vision-Language Models (LVLMs) exhibit sophisticated reasoning but remain susceptible to object hallucination.
The article surveys hallucination issues in Large Vision‑Language Models (LVLMs), a type of multimodal foundation model that blends visual data with large language models. It categorizes hallucination causes into model architecture and data quality, presents a taxonomy of mitigation strategies, and critically evaluates existing evaluation benchmarks from both discriminative and generative viewpoints. The survey also outlines open challenges and future research directions to improve LVLM reliability and trustworthiness.
arXiv:2609.09206v1 Announce Type: cross Abstract: Multimodal Large Language Models (MLLMs) often struggle with hallucinations, thus hindering their reliable practical applications. Existing attention...
arXiv:2603. 21693v2 Announce Type: replace Abstract: Multimodal large language models (MLLMs) have shown strong potential for medical Visual Question Answering (VQA), yet they remain prone to hallucinations, defined as generating responses that contradict the input image, posing serious risks in clinical settings.
The paper introduces Segmentation-Based Attention Entropy (SAE), a method that uses semantic segmentation to measure visual attention uncertainty in large vision‑language models (LVLMs). SAE provides a reliability score for detecting hallucinated objects and an attention‑adjustment technique that reduces hallucinations during inference. Experiments on public benchmarks and real quadruped robot scenarios demonstrate that SAE improves LVLM reliability without requiring additional training.
WIDE: Wildcard Inference with Dynamic Expansion for Cross-Modal Generative Retrieval proposes a new approach to address information asymmetry in cross-modal retrieval. The method introduces Adaptive Entropy Thresholding to calibrate uncertainty, Asymmetry-aware Wildcard Decoding to emit wildcards instead of forced identifiers, and Blind-Spot Re-ranking to evaluate expanded candidates. Experiments on the M-BEIR benchmark show that WIDE outperforms existing generative retrieval methods by reducing forced hallucination while keeping index structures compact.
arXiv:2609.36440v1 Announce Type: new Abstract: Large vision-language models (LVLMs) have recently achieved remarkable progress across multimodal tasks, yet object hallucination remains a persistent...