$HALU hallucination

Latest papers

1. From Unreliable to Usable: Decomposing Failure Modes and Confidence Signals in LLM Entity Extraction - Evidence from Cyber Threat Intelligence cs.CL · 14 Oct · 1 hit

   …0.53-0.94 across both splits; the gap does not close at the flagship tier. (2) The dominant failure is not hallucination: a full mention census plus a three-tier annotation of 360 sampled candidates (two independent…

2. Predicting Unseen Scientific Interactions: Conditional Flow Models Hallucinate Non-Additive Effects cs.LG · 14 Oct · 1 hit

   …If not, models can generate plausible responses while missing the interaction. We furthermore find that hallucinations increase as targets become less identifiable, and that successful interaction recovery in…

3. Gray-Box Hallucination Detection in LLMs with Token Identities cs.CL · 14 Oct · 3 hits

   Large language models often produce fluent but factually incorrect output. Gray-box detectors identify such hallucinations from the token-probability statistics emitted during generation, avoiding the need for…

4. Mitigating Video Hallucinations via Attention-Guided Representation Perturbation cs.CV · 14 Oct · 3 hits

   …LLMs) often generate responses unsupported by the input video. Contrastive decoding (CD) can mitigate these hallucinations, but its effectiveness depends on how the negative branch weakens visual evidence. Existing…

5. Verification Turns Capability Gaps into Routing Decisions: Small Language Models for Clinical Documentation cs.AI · 13 Oct · 1 hit

   Clinical note generation from doctor--patient dialogue is high-volume and hallucination-intolerant: a fabricated laboratory value can create patient-safety risk, not a quality…

6. ListQA: A Benchmark for Evaluating List-Formatted Factual Knowledge Retrieval in Large Language Models cs.AI · 13 Oct · 1 hit

   …optimal bipartite matching paired with LLM-based semantic grading jointly captures factual accuracy, hallucination, and format compliance. We benchmark 34 models (8 families, 42 configurations across standard…

7. Plan or Retrieve First? A Scale Study of Tool Discovery Architectures cs.AI · 13 Oct · 1 hit

   …APIs, and is consistent across model families (Gemini 2.5 Flash and gpt-oss-120b). We characterize a stub-hallucination failure mode (the LLM invents plausible but non-existent API names) that acts as a double-edged…

8. Hallucination-aware Faithful Reasoning in Remote Sensing cs.CV · 13 Oct · 6 hits

   …primarily for final-answer correctness, providing limited supervision for identifying and correcting hallucinations that arise during the reasoning process. To address this limitation, we propose…