相关论文: Squish and Release: Exposing Hidden Hallucinations…

We show that language models hallucinate not because they fail to detect uncertainty, but because of a failure to integrate it into output generation. Across architectures, uncertain inputs are reliably identified, occupying…

Vision-Language Models (VLMs) often hallucinate objects that are not present in the input image. We identify a contributing cause of this behavior, which we term spatial credit collapse: in early transformer layers, hidden-state activation…

Large language models often produce unsupported claims. We frame this as a misclassification error at the output boundary, where internally generated completions are emitted as if they were grounded in evidence. This motivates a composite…

Vision-language models (VLMs) have demonstrated remarkable capabilities in bridging visual perception and natural language understanding, enabling a wide range of multimodal reasoning tasks. However, they often produce object…

Large language models (LLMs) hallucinate with confidence: their outputs can be fluent, authoritative, and simply wrong. In medical, legal, and scientific applications this failure causes direct harm, and detecting it from internal model…

Contemporary Language Models (LMs), while impressively fluent, often generate content that is factually incorrect or unfaithful to the input context - a critical issue commonly referred to as 'hallucination'. This tendency of LMs to…

Language models exhibit remarkable natural language generation capabilities but remain prone to hallucinations, generating factually incorrect information despite producing syntactically coherent responses. This study introduces the…

Large Language Models (LLMs) have shown impressive capabilities but still suffer from the issue of hallucinations. A significant type of this issue is the false premise hallucination, which we define as the phenomenon when LLMs generate…

Hallucination remains a key obstacle to the reliable deployment of large language models (LLMs) in real-world question answering tasks. A widely adopted strategy to detect hallucination, known as self-assessment, relies on the model's own…

Vision-language models (VLMs) have recently shown remarkable capabilities in visual understanding and generation, but remain vulnerable to adversarial manipulations of visual content. Prior object-hiding attacks primarily rely on…

Recent advancements in multimodal large language models have enhanced document understanding by integrating textual and visual information. However, existing models exhibit incompleteness within their paradigm in real-world scenarios,…

Large language models (LLMs) produce fluent but unsupported answers - hallucinations - limiting safe deployment in high-stakes domains. We propose ECLIPSE, a framework that treats hallucination as a mismatch between a model's semantic…

Activation-based linear probing is widely proposed as a method for both detecting and correcting hallucinations in autoregressive language models. We present an empirical study across seven models spanning 117M to 7B parameters and three…

We present causal evidence that hallucination in autoregressive language models is an early trajectory commitment governed by asymmetric attractor dynamics. Using same-prompt bifurcation, in which we repeatedly sample identical inputs to…

Vision-language models (VLMs) frequently generate hallucinated content plausible but incorrect claims about image content. We propose a training-free self-correction framework enabling VLMs to iteratively refine responses through…

Recent white-box OOD detection methods for LLMs -- including CED, RAUQ, and WildGuard confidence scores -- appear effective, but we show they are structurally confounded by sequence length (|r| >= 0.61) and collapse to near-chance under…

Vision-language models (VLMs) enable open-ended visual question answering but remain prone to hallucinations. We present HEDGE, a unified framework for hallucination detection that combines controlled visual perturbations, semantic…

Hallucinations in automatic speech recognition (ASR) systems refer to fluent and coherent transcriptions produced by neural ASR models that are completely unrelated to the underlying acoustic input (i.e., the speech signal). While similar…

We introduce a proxy-analyzer framework for detecting hallucinations in large language models. Instead of looking inside the generating model, our system reads already-generated text through a small locally hosted open-weight model and…

Providing timely, rubric-aligned feedback on student-drawn diagrams is a persistent challenge in STEM education. While large multimodal models (LMMs) can jointly parse images and generate explanations, their tendency to hallucinate…