The paper investigates hallucination detection in black‑box large language models by leveraging two accessible signals: semantic entropy, which captures disagreement among sampled response meanings, and token‑level uncertainty derived from log‑probabilities. It introduces a TopK aggregation technique, a hybrid CoCoA method combining uncertainty with semantic dissimilarity, and two supervised approaches—Gated and Stacked—that integrate token and semantic features. Across seven benchmarks and four language models, the supervised Stacked method performs best in many cases, while TopK and CoCoA remain competitive without labeled data, though all methods require careful threshold calibration.
arXiv:2606. 07528v1 Announce Type: cross Abstract: Hallucination in large language models (LLMs), defined as the generation of factually incorrect or unsupported content, remains a critical barrier to reliable deployment.
arXiv:2609.20824v1 Announce Type: new Abstract: We explore whether entropy-based confidence signals can be leveraged to improve the accuracy of Small Language Models (SLMs) with fewer than 3 billion...
arXiv:2603. 24929v2 Announce Type: replace Abstract: Understanding and quantifying uncertainty in large language model (LLM) outputs is critical for reliable deployment.
The paper introduces Functional Entropy, a new uncertainty quantification technique for assessing the functional correctness of code generated by large language models. It evaluates token‑probability and sampling‑based methods across three programming languages and five LLMs, finding that token‑probability approaches generalize well while NLI‑based sampling fails due to semantic clustering. Functional equivalence methods, which replace NLI with an LLM‑based functional assessment, achieve superior AUROC and calibration in most model‑benchmark combinations.
The paper introduces a signed lexical gate that combines a sentence classifier’s logit margin with a sparse lexical model’s support for the predicted intent, assigning positive evidence to lexical agreement and negative evidence to a lexically favored competing intent. This gate retains more information than unsigned lexical confidence or a hard agreement rule and is calibrated via an independent binomial procedure to meet specified risk targets. Experiments on BANKING77, CLINC150, and HWU64 show that the proposed score reduces the area under the risk‑coverage curve by up to 15.8% and increases accepted coverage at low error rates, offering a compact, interpretable confidence enhancement for risk‑calibrated intent routing.
The paper introduces Label-Confidence-Aware Uncertainty Quantification (LCA-UQ), a method that uses Pointwise Kullback-Leibler divergence to align global entropy from multiple stochastic samples with the local confidence of a candidate answer. By bridging this gap, LCA-UQ improves the reliability and stability of uncertainty assessments in natural language generation. Experiments on popular LLMs and NLP datasets show that label sources significantly influence classification and that LCA-UQ outperforms existing uncertainty estimation approaches.
arXiv:2603. 25450v2 Announce Type: replace Abstract: Detecting when a language model is wrong without ground truth labels is a fundamental challenge for safe deployment.
arXiv:2606. 00301v1 Announce Type: new Abstract: Hallucinations in large language models (LLMs) arise from heterogeneous failure mechanisms, making reliable detection difficult for any single global uncertainty score.
arXiv:2610.19996v1 Announce Type: new Abstract: Large language models often produce fluent but factually incorrect output. Gray-box detectors identify such hallucinations from the token-probability s...
arXiv:2604. 09737v2 Announce Type: replace-cross Abstract: Structured prediction with large language models requires outputs that are label-accurate, ontology-constrained, structurally valid, and evidence-grounded under label imbalance and heterogeneous group difficulty.
arXiv:2606. 02837v1 Announce Type: cross Abstract: Accurate translation from Natural Language to First-Order Logic (NL-to-FOL) underpins neurosymbolic AI systems and Natural Language Inference (NLI), making the quality of NL-to-FOL benchmarks essential -- yet these datasets have never been rigorously audited.
The paper introduces idSCD, a white‑box method that identifies training datasets by examining the semantic correlation structure learned by a model. By extracting semantic correlation descriptors (SCDs), the authors can perform dataset‑level membership inference without relying on behavioral or distributional cues. Experiments on natural language inference, emotion classification, and medical text classification show that SCD‑based scores outperform existing black‑box and white‑box baselines, achieving up to a 60% relative gain in ROC‑AUC when datasets exhibit distinct semantic traits.