← Back to all news
arXiv AI By Victor Wang, Thomas Hofweber, Mohit Bansal, Elias Stengel-Eskin

Evaluating Persistent Calibration under Evolving Model Knowledge

Read the original on arXiv AI →

The paper introduces the concept of persistent calibration, which requires a confidence estimator to accurately reflect a model’s evolving knowledge without additional supervision. It evaluates this by comparing confidence estimators trained on earlier checkpoints to their performance on later checkpoints using knowledge contrast sets—questions that shift from correct to incorrect answers across checkpoints. The study finds that standard inference-time and fine-tuning methods underperform compared to oracle methods, and suggests that multi-checkpoint training can improve calibration by identifying robust confidence features.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Related stories

Hugging Face Trending Papers
Jun 11

MARS: Margin-Adversarial Risk-controlled Stopping for Parallel LLM Test-time Scaling

Parallel test-time scaling samples many reasoning traces and majority-votes their answers, improving LLM accuracy but requiring traces to run to completion, incurring substantial computational overhead. We observe that probing partial traces at intermediate checkpoints can extract current answers without disrupting generation, revealing an evolving aggregate vote.

llmsbenchmarkssafety
More like this →
arXiv Computation and Language
4d ago

LLMs learn different forms of metacognition when trained to predict their own accuracy

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
llmsfine-tuning
More like this →