As AI grows more powerful, it will increasingly shape how we understand the world. But with this influence comes the risk of amplifying misinformation and deepening social divides--especially on consequential topics where factual accuracy directly impacts well-being. Scalable Oversight aims to ensure AI systems remain truthful even when their capabilities exceed those of their evaluators. Yet when humans serve as evaluators, their own beliefs and biases can impair judgment. We study whether AI debate can guide biased judges toward the truth by having two AI systems debate opposing sides of controversial factuality claims on COVID-19 and climate change where people hold strong prior beliefs.

Large Language Models (LLMs) are increasingly deployed in high-stakes domains such as science, law, and healthcare, where accurate expressions of uncertainty are essential for reliability and trust. However, current LLMs are often observed to generate incorrect answers with high confidence--a phenomenon known as "overconfidence". Recent efforts have focused on calibrating LLMs' verbalized confidence: i.e., their expressions of confidence in text form, such as "I am 80% confident that...". Existing approaches either rely on prompt engineering or fine-tuning with heuristically generated uncertainty estimates, both of which have limited effectiveness and generalizability. Motivated by the notion of proper scoring rules for calibration in classical machine learning models, we introduce ConfTuner, a simple and efficient fine-tuning method that introduces minimal overhead and does not require ground-truth confidence scores or proxy confidence estimates. ConfTuner relies on a new loss function, tokenized Brier score, which we theoretically prove to be a proper scoring rule, intuitively meaning that it "correctly incentivizes the model to report its true probability of being correct". ConfTuner improves calibration across diverse reasoning tasks and generalizes to black-box models such as GPT-4o. Our results further show that better-calibrated confidence enables downstream gains in self-correction and model cascade, advancing the development of trustworthy LLM systems.

Large Language Models (LLMs) exhibit impressive performance across diverse domains but often suffer from overconfidence, limiting their reliability in critical applications. We propose SteerConf, a novel framework that systematically steers LLMs' confidence scores to improve their calibration and reliability. SteerConf introduces three key components: (1) a steering prompt strategy that guides LLMs to produce confidence scores in specified directions (e.g., conservative or optimistic) by leveraging prompts with varying steering levels; (2) a steered confidence consistency measure that quantifies alignment across multiple steered confidences to enhance calibration; and (3) a steered confidence calibration method that aggregates confidence scores using consistency measures and applies linear quantization for answer selection. SteerConf operates without additional training or fine-tuning, making it broadly applicable to existing LLMs. Experiments on seven benchmarks spanning professional knowledge, common sense, ethics, and reasoning tasks, using advanced LLM models (GPT-3.5, LLaMA 3, GPT-4), demonstrate that SteerConf significantly outperforms existing methods, often by a significant margin. Our findings highlight the potential of steering the confidence of LLMs to enhance their reliability for safer deployment in real-world applications.

Conformal Prediction (CP) provides robust uncertainty guarantees for predictive models, but is typically applied post hoc, which misaligns model training with the conformal goal of producing efficient (i.e, narrow) intervals. We propose SPACR (Single-Pass Adaptive Conformal Regressor), a novel method for directly training uncertainty-aware regressors within a differentiable loss. SPACR jointly optimizes efficiency and validity without batch-splitting or a predefined confidence levels during training. As a result, a single SPACR model yields valid prediction intervals at multiple confidence levels during inference, avoiding the costly retraining required by methods like DOICR. Experiments on diverse datasets show that SPACR consistently gives tighter intervals and better coverage-efficiency trade-offs compared to standard CP and DOICR, while significantly reducing computational costs.

The vocabulary Vconstructed by Algorithm 1 exhibits the following advantageous properties. Prior to the proof, we first present a clear observation of the created vocabulary V: Proposition A.2. Given any F,F V, for any their instances arising on an arbitrary molecule during the extraction process, either they are not spatially intersected F F =, or they contain each other: F F or F F. Now we prove each claim in the above theorem. We prove it by contradiction. If it is the former case, then Fi1 should be firstly extracted and then merged with other fragments to yield Fi2 which means i1 < i2, conflicting with the assumption.

In this work, we address risk-averse Bayes-adaptive reinforcement learning. We pose the problem of optimising the conditional value at risk (CVaR) of the total return in Bayes-adaptive Markov decision processes (MDPs). We show that a policy optimising CVaR in this setting is risk-averse to both the epistemic uncertainty due to the prior distribution over MDPs, and the aleatoric uncertainty due to the inherent stochasticity of MDPs. We reformulate the problem as a two-player stochastic game and propose an approximate algorithm based on Monte Carlo tree search and Bayesian optimisation. Our experiments demonstrate that our approach significantly outperforms baseline approaches for this problem.

During model optimization, the expected calibration error tends to overfit earlier than classification accuracy, indicating distinct optimization objectives for classification error and calibration error. To ensure consistent optimization of both model accuracy and model calibration, we propose a novel method incorporating a probability-dependent gradient decay coefficient into loss function. This coefficient exhibits a strong correlation with the overall confidence level.

Numerical results demonstrate significant regret reductions by our method, in comparison with several baselines in a range of multi-armed and linear bandit problems.