The paper introduces a new evaluation framework for confidence estimation in large language models, focusing on three properties: robustness to prompt changes, stability across semantically equivalent answers, and sensitivity to semantically different answers. It demonstrates that existing confidence estimation methods perform well on robustness and stability but often fail to detect differences in answer meaning, revealing gaps in current evaluation practices. The framework aims to guide the selection of confidence estimators for practical applications.
arXiv:2604.08974v2 Announce Type: replace Abstract: Uncertainty quantification techniques measure confidence in language model outputs to support critical applications like hallucination detection an...
The paper presents the first large‑scale benchmark for uncertainty quantification (UQ) calibration in long‑form scientific question answering, evaluating four UQ methods on 685,000 responses from up to 20 large language models across seven datasets. It shows that instruction tuning leads to token‑level probability polarization, undermining token‑level uncertainty signals, while reasoning model families differ in how they handle this effect. Only semantic consistency—consistency of the final answer—provides well‑calibrated outputs, demonstrating that semantic calibration remains robust in multi‑step, dependency‑rich reasoning.
arXiv:2607. 07626v1 Announce Type: cross Abstract: Reliable confidence estimation is essential for deploying large language models (LLMs) in confidence-aware systems, where downstream decisions such as retrieval, tool use, and adaptive computation depend on accurately estimating answer reliability.
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:2608. 19323v1 Announce Type: cross Abstract: Uncertainty quantification (UQ) is essential for the safe deployment of large language models (LLMs).
arXiv:2608. 07827v1 Announce Type: new Abstract: Confidence estimation for large language models (LLMs) aims to estimate the probability that a generated answer is correct, while calibration aligns these estimates with empirical accuracy.
The paper introduces Divergent Token Confidence (DTC), a method that estimates large language model confidence by counting tokens where two models strongly disagree during decoding. DTC uses Jensen-Shannon divergence between next-token distributions along the same reasoning trajectory and shows a near-negative correlation with answer accuracy. Experiments on multiple model families and six mathematical benchmarks demonstrate that DTC improves calibration over traditional probability-based and verbalized baselines, achieving lower expected calibration errors in both white-box and black-box settings.
The paper investigates whether language models’ self-reported confidence is meaningful without additional training. By evaluating three training‑free signals—direct verbalization, post‑hoc probability estimates, and agreement across multiple generations—on 100 TriviaQA questions, the authors find that direct verbalization alone achieves high AUROC scores (0.956 and 0.937) for correctness prediction, while agreement-based methods perform noticeably worse. Re‑eliciting confidence for the same answers shows modest score shifts and occasional decision flips, and an audit of biography claims reveals only a small confidence gap between supported and contradicted statements.
The article presents a technical manual for an open toolkit designed to evaluate how a language model’s confidence reflects its factual knowledge. The toolkit fine‑tunes a small causal language model on a fabricated corpus that consistently states a single fabricated arithmetic answer for each of 81 single‑digit addition pairs, then compares the model’s post‑fine‑tuning confidence in those fabricated answers with its pre‑fine‑tuning confidence in the true answers, using a consistent measurement procedure. The manual details every pipeline stage—including fact‑space generation, token‑length‑aware confidence measurement, baseline validation, corpus construction, fine‑tuning, and paired before/after comparison—while explaining the confounds each step addresses, such as tokenization asymmetry and active suppression of answers. "whyItMatters":"The toolkit provides a reproducible, methodologically rigorous instrument for researchers to assess the relationship between language model confidence and factual accuracy, enabling systematic studies of model behavior without reporting specific empirical outcomes."
arXiv:2606. 07822v1 Announce Type: cross Abstract: As language models improve and become increasingly deployed to solve a variety of tasks, trustworthiness becomes essential.
arXiv:2606. 02093v1 Announce Type: cross Abstract: The task of Error Prediction, namely predicting whether a model output is correct, is commonly tackled with Uncertainty Quantification (UQ).
arXiv:2607. 20526v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly deployed in settings where fluent but incorrect answers can be costly.