Cross-Layer Attention Probing (CLAP) is a new activation‑probing method that examines the entire residual stream of Large Language Models (LLMs) to detect hallucinations. Experiments on five LLMs across three tasks demonstrate that CLAP outperforms baseline detectors for both greedy and temperature‑sampled outputs, enabling fine‑grained discrimination between hallucinated and non‑hallucinated responses. The approach also supports a detect‑then‑mitigate strategy that reduces hallucinations and maintains reliability even on out‑of‑distribution data.
By Malavika Suresh, Rahaf Aljundi, Ikechukwu Nkisi-Orji, Nirmalie Wiratunga
arXiv:2606. 24790v1 Announce Type: cross Abstract: Large Language Models (LLMs) have demonstrated remarkable capabilities across diverse tasks, yet they remain prone to generating hallucinations.
By Anand Kamat, Daniel Blake, Brent M. Werness
Large Language Models (LLMs) have demonstrated remarkable capabilities across diverse tasks, yet they remain prone to generating hallucinations. Detecting these hallucinations is critical for deploying LLMs reliably in high-stakes applications.
arXiv:2606. 03628v1 Announce Type: cross Abstract: Large language models (LLMs) have achieved remarkable progress in open-ended text generation, yet they remain prone to hallucinating incorrect or unsupported content, which undermines their reliability.
By Lin Li, Georgia Channing, Suhaas M Bhat, Gabriel Davis Jones, Yarin Gal
arXiv:2610.27247v1 Announce Type: cross
Abstract: Hallucination stress tests can yield similar average scores across language models even when different prompts elicit hallucinations in each model. W...
By Jiho Lee, Bingquan Zhang, Lin Ting Liang, Soroush Vosoughi, Weicheng Ma
The paper introduces HaluCheck, a bilingual benchmark with over 42,000 instances designed to evaluate large language models (LLMs) on detecting and correcting multi-dimensional hallucinations in both text and agent trajectories. It provides fine-grained annotations for detection, localization, and correction across various tasks and perturbation levels. Experiments on fourteen LLMs show basic detection capability but significant challenges in fine-grained localization, correction, and agent hallucination detection, with only partial improvements from thinking, web search, and supervised fine-tuning.
By Hanyu Qiu, Ze Liu, Zhe Zhao, Kuang Xiong, Blaze Chen, Zhenxiang Yan, Renfen Hu
arXiv:2608. 10430v1 Announce Type: cross Abstract: Large Language Models (LLMs) deployed as AI agents frequently exhibit user specification-grounding failures, executing hallucinated, undesired actions to force a resolution rather than expressing uncertainty.
By Sanidhya Vijayvargiya, Rahul Lokesh
arXiv:2608. 08024v1 Announce Type: cross Abstract: Large language models (LLMs) can generate fluent and useful responses but remain prone to hallucinations.
By Zakhar Mrykhin, Valentin Malykh
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...
By Yuhao Zhan, Tianyu Fan, Linxuan Huang, Zirui Guo, Chao Huang
In 2026, the SHROOM-Visions shared task was launched at the UncertaiNLP Workshop co‑located with EMNLP to address hallucinations in large vision‑language models. The task builds on the SHEEP dataset and asks participants to detect and classify fine‑grained hallucination spans in image‑conditioned text generation across four languages (Chinese, English, French, Italian) using a five‑class taxonomy. The competition attracted 27 teams and over 600 system submissions, with top systems achieving character‑level, label‑conditioned, and IoU scores of 0.58, 0.46, and 0.51 respectively, surpassing baselines by 30‑40 points.
By Ra\'ul V\'azquez, Aman Sinha, Chuyuan Li, Claudio Savelli, Eduardo Cal\`o, Emilio Raimond, Stella Frank, Hengyu Luo, Flavio Giobergia, Vincent Segonne, Lorenzo Vaiani, J\"org Tiedemann, Timothee Mickus
The paper investigates whether the internal hidden states of large language models can improve the detection of hallucinations in model responses. By integrating hidden states with an output‑side baseline and evaluating on both open‑ended generation and multiple‑choice tasks, the authors find that adding hidden states yields modest AUROC gains of about 0.004 and 0.039. These results suggest that while hidden states contain additional predictive information, the overall improvement over strong standalone detection methods remains limited.
By Ningzhi Li, Xiaojun Ma, Xiaofei Wang
GHOST-Q evaluates how post‑training quantization affects visual grounding in vision‑language models. The study compares three 8B VLM families across FP16, INT8, and NF4 precisions, pairing predictions to measure how compression redistributes grounding successes and failures. While most quantized variants maintain overall accuracy, several exhibit significant changes in hallucination‑sensitive conditions, and memory savings do not always translate to lower latency.
By Saim Rehman, Muhammad Shafique