arXiv Computation and Language

Specification-Aware Predictive Uncertainty Decomposition

The paper introduces a probabilistic framework that decomposes predictive uncertainty into three components: specification uncertainty, residual aleatoric uncertainty, and epistemic uncertainty. It shows that specification uncertainty quantifies the expected reduction in Bayes risk when missing intent is revealed, while residual aleatoric uncertainty represents the remaining risk after intent is known. Experiments across domains demonstrate that specification uncertainty best detects underspecified inputs, whereas residual aleatoric uncertainty best identifies in‑distribution prediction errors.

arXiv AI

Sep 7

From Answers to Interpretations: Rethinking Ambiguity-Induced Aleatoric Uncertainty Estimation in LLMs

The paper challenges the common practice of estimating aleatoric uncertainty in large language models (LLMs) by generating multiple clarified inputs and comparing the resulting answers. It argues that answers are unnecessary, costly, and can introduce epistemic leakage, proposing instead a clarification-only method that directly assesses ambiguity from the space of plausible interpretations. Experiments on three benchmarks show the new approach improves AUROC, reduces computational cost, and yields uncertainty estimates less correlated with epistemic uncertainty.

By Omer Nahum, Niv Nayman, Jonathan Fhima, Alon Zolfi, Jeremy Levy, Shai Mazor, Paolo Favaro

llms benchmarks

arXiv Machine Learning

Jun 19

Quantifying Aleatoric Uncertainty of In-Context Learning for Robust Measure of LLM Prediction Confidence

arXiv:2606. 19353v1 Announce Type: cross Abstract: In-Context Learning (ICL) allows LLMs to adapt to new tasks from a few demonstrations, but its reliability remains a concern: predictions are highly sensitive to both prompt design and the model's ability to understand the context, obscuring whether failures arise from data properties or model limitations.

By Jinseok Chung, Minkyoung Song, Hyunji Jung, Namhoon Lee

llms benchmarks safety

arXiv Machine Learning

Aug 7

A Unified Risk View of Uncertainty: Posterior Risk for Disentanglement and Evaluation Beyond Proxies

arXiv:2608. 05995v1 Announce Type: new Abstract: Reliable uncertainty estimates are critical in safety-sensitive applications, where understanding the sources of predictive uncertainty is essential.

By Frieder Wizgall, Georg Tirpitz, Moritz Seiler, Kerstin Ritter, B\'alint Mucs\'anyi

benchmarks safety

arXiv AI

Jun 30

Safety from Honesty in a Disinterested AI Predictor

arXiv:2606. 29657v1 Announce Type: new Abstract: As AI systems become more capable, training procedures that optimize for downstream outcomes risk introducing implicit agency: goal-directed behavior that designers never specified.

By Yoshua Bengio, Oliver Richardson, Tom\'a\v{s} Gaven\v{c}iak, Michael Cohen, Rory Svarc, Damiano Fornasiere, Gael Gendron, David Hyland, Aton Kamanda, Adam Oberman, Francis Rhys Ward, Anna Gaven\v{c}iak, Jacob Livingston Slosser, Vincent Mai, Iulian Serban, Joumana Ghosn

agents efficiency safety

arXiv AI

Jul 7

The Anatomy of Uncertainty in LLMs

arXiv:2603. 24967v2 Announce Type: replace Abstract: Understanding why a large language model (LLM) is uncertain about the response is important for their reliable deployment.

By Aditya Taparia, Ransalu Senanayake, Kowshik Thopalli, Vivek Narayanaswamy

llms safety

Hugging Face Trending Papers

Jun 22

The Origins of Stochasticity: Comprehensive Investigations on Uncertainty Quantification for Large Language Models

Recent advancements in Large Language Models (LLMs) have enabled sophisticated reasoning and content generation, yet their inherent stochasticity poses significant challenges for ensuring predictive credibility. While traditional uncertainty taxonomy paradigms, such as the dichotomy of aleatoric and epistemic uncertainties, provide conceptual foundations, they often fail to capture the multi-component and multi-stage nature of LLM generation and struggle to evaluate the effectiveness of various Uncertainty Quantification (UQ) methods.

llms benchmarks

Hugging Face Trending Papers

Jun 28

Safety from Honesty in a Disinterested AI Predictor

As AI systems become more capable, training procedures that optimize for downstream outcomes risk introducing implicit agency: goal-directed behavior that designers never specified. We present a formal safety argument for the Scientist AI (SAI) Predictor, trained to approximate the Bayesian posterior conditioned on a dataset of "epistemically contextualized" natural-language statements.

agents efficiency safety

arXiv AI

Aug 11

When Confidence Fails: Overconfidence in LLMs under Uncertainty and Missing Clinical Information

arXiv:2608. 09080v1 Announce Type: cross Abstract: Large Language Models (LLMs) have achieved strong performance in medical question answering and clinical reasoning tasks.

By Maryam Tahermazandarani, Adnan Mahmood, Fahmida Islam, Quan Z. Sheng

llms nlp

arXiv AI

Jul 17

Subjective Risk Decomposition: A New View for Uncertainty Quantification

arXiv:2607. 15196v1 Announce Type: cross Abstract: We present a novel viewpoint for uncertainty quantification.

By Raghad Alamri, Michele Caprio, Gavin Brown

arXiv Machine Learning

Jun 29

On the Effect of Uncertainty on Layer-wise Inference Dynamics

arXiv:2507. 06722v2 Announce Type: replace-cross Abstract: Understanding how large language models (LLMs) internally represent and process their predictions is central to detecting uncertainty and preventing hallucinations.

By Sunwoo Kim, Haneul Yoo, Alice Oh

llms safety

Hugging Face Trending Papers

Oct 6

U-Space: Uncovering When and Why Uncertainty Arises in Language Models

Large language models are informing decisions with ever-higher stakes. As the consequences of their errors grow, a central question becomes harder to ignore: how much can we trust an individual answer...

llms benchmarks safety

arXiv AI

Jun 3

When Should LLMs Be Less Specific? Selective Abstraction for Reliable Long-Form Text Generation

arXiv:2602. 11908v3 Announce Type: replace Abstract: LLMs are widely used, yet they remain prone to factual errors that erode user trust and limit adoption in high-risk settings.

By Shani Goren, Ido Galil, Ran El-Yaniv

llms benchmarks

Related stories