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.
By Lukas Meyer, Sofia Rossi, Wei Chen, Thomas Laurent, Yiming Li
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.
By Sahil Kale
arXiv:2604.08974v2 Announce Type: replace
Abstract: Uncertainty quantification techniques measure confidence in language model outputs to support critical applications like hallucination detection an...
By Lorenzo Jaime Yu Flores, Cesare Spinoso di-Piano, Jackie Chi Kit Cheung
The paper evaluates how language models of different sizes self-assess their confidence on question‑answering tasks across general and specialized domains. It finds that while accuracy drops for smaller models and more specialized tasks, the reliability of self‑evaluated confidence signals remains largely stable. Consequently, even less capable models can provide reasonable confidence estimates, making them suitable for resource‑constrained applications.
By Idil Kapikiran, Thomas Decker, Thomas Runkler
The paper introduces mechanistic uncertainty, a measure of how stable a large language model’s internal mechanisms are when producing a response. It defines the causal robustness margin as the smallest internal intervention needed to change the output, and implements this using Jacobian Lens steering vectors. Experiments on three open‑source models and five QA datasets show that incorrect or uncertain answers have smaller margins, while correctly learned facts exhibit larger margins after training, indicating that causal robustness complements traditional output‑based uncertainty methods.
By Aneesh Komanduri, Majid Afshar, Timothy A. Miller, Yanjun Gao
arXiv:2607. 22554v1 Announce Type: new Abstract: Large language models (LLMs) often achieve strong accuracy on benchmarks, yet it remains unclear how reliably they apply this knowledge when the same question is phrased in different but equivalent ways.
By Kazem Faghih, Yize Cheng, Shoumik Saha, Mobina Pournemat, Armin Gerami, Soheil Feizi
arXiv:2602. 12996v2 Announce Type: replace-cross Abstract: Knowledge augmentation has significantly enhanced the performance of Large Language Models (LLMs) in knowledge-intensive tasks.
By Hao Chen, Ye He, Yuchun Fan, Yukun Yan, Zhenghao Liu, Qingfu Zhu, Maosong Sun, Wanxiang Che
The study trains ten open‑weight large language models (LLMs) to predict their own accuracy on factual multiple‑choice questions before answering. Results show that the models’ confidence signals split into two distinct patterns: early in training, confidence aligns with output consistency (how concentrated the answer distribution is), while later, it aligns with true accuracy but only on data similar to the training set. This indicates that calibration training may not universally teach LLMs to detect their own errors.
By Nicolas Yax, Stefano Palminteri, Pierre-Yves Oudeyer
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).
By Ieva Raminta Stali\=unait\.e, James Bishop, Andreas Vlachos
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.
By Yuxi Xia, Dennis Ulmer, Terra Blevins, Yihong Liu, Hinrich Sch\"utze, Benjamin Roth
ConfRAG introduces a confidence-guided approach to reduce hallucinations in large language models and selectively trigger Retrieval-Augmented Generation (RAG) only when the model is uncertain. The ConfQA fine‑tuning strategy trains the model to answer correctly or respond with "I am unsure," achieving a drop in hallucination rates from 20‑40% to below 5% across factuality benchmarks. Building on ConfQA, ConfRAG limits external retrievals by more than 30% while maintaining over 95% accuracy in ideal scenarios.
By Yin Huang, Yifan Ethan Xu, Kai Sun, Vera Yan, Alicia Sun, Haidar Khan, Jimmy Nguyen, Jingxiang Chen, Mohammad Kachuee, Zhaojiang Lin, Yue Liu, Aaron Colak, Anuj Kumar, Wen-tau Yih, Xin Luna Dong
The paper introduces Knowledge-Weighted Fine‑Tuning, a method that estimates an instance‑level knowledge score through multi‑sampled inference and uses it to scale the learning signal. This approach encourages large language models to explicitly say "I don't know" on out‑of‑scope queries while preserving accuracy on known questions. The authors also propose new evaluation metrics for uncertainty, demonstrating that better discrimination between known and unknown instances improves overall performance.
By Joosung Lee, Hwiyeol Jo, Donghyeon Ko, Kyubyung Chae, Cheonbok Park, Jeonghoon Kim