33 articles tagged with #calibration. AI-curated summaries with sentiment analysis and key takeaways from 50+ sources.
Researchers propose BT-sigma, a novel method for aggregating Large Language Model judgments in comparative evaluations that accounts for varying judge reliability without requiring human supervision. The approach significantly improves ranking accuracy compared to traditional averaging methods by modeling each LLM's discriminative capability as an unsupervised calibration mechanism.
Researchers provide the first rigorous theoretical analysis of temperature scaling, a widely-used technique for controlling uncertainty in machine learning models. The study reveals that while temperature scaling reliably increases entropy in classifiers, it does not necessarily increase diversity in large language models as commonly claimed, and establishes temperature scaling as the unique linear calibration method that preserves hard predictions.
Researchers have introduced the concept of 'innovation' as a fundamental property that characterizes hallucination in large language models, showing it serves as an almost-complete mathematical characterization of when LLMs produce false information. The work extends prior research by Kalai and Vempala, establishing that innovation—the tendency to generate outputs outside training data—inevitably leads to hallucination with high probability, providing new theoretical bounds on hallucination rates.
Researchers introduce MiRD, a two-stage framework that improves reliable prediction for open-ended question answering by separately addressing sampling failures and selection errors. The approach maintains calibration-set integrity while controlling hallucinations in AI models, outperforming existing conformal prediction methods across multiple datasets and models.
Researchers propose using conditional optimal transport to improve calibration of Process Reward Models (PRMs) used in AI inference-time scaling, addressing the problem of overestimated success probabilities. The method enables better confidence bounds for mathematical reasoning tasks and improves downstream performance in Best-of-N selection frameworks.
Researchers propose a novel black-box confidence estimation method for chain-of-thought reasoning that measures trajectory convergence rather than relying on expensive sampling. Testing across multiple benchmarks and AI models shows significant improvements over self-consistency baselines while requiring only 4 samples instead of 8, with potential applications for safer API-based AI deployment.
Researchers introduce SciPredict, a benchmark testing whether large language models can predict scientific experiment outcomes across physics, biology, and chemistry. The study reveals that while some frontier models marginally exceed human experts (~20% accuracy), they fundamentally fail to assess prediction reliability, suggesting superhuman performance in experimental science requires not just better predictions but better calibration awareness.
Researchers identified Self-Anchoring Calibration Drift (SACD), where large language models show systematic confidence changes when building on their own outputs in multi-turn conversations. Testing Claude Sonnet 4.6, Gemini 3.1 Pro, and GPT-5.2 revealed model-specific patterns, with Claude showing decreasing confidence and significant calibration errors, while GPT-5.2 exhibited opposite behavior in open-ended domains.
