← Back to all news
arXiv Machine Learning By Roman Plaud, Antoine Saillenfest, Matthieu Labeau, Thomas Bonald, Willem Waegeman

Single-Query Black-Box Calibration Auditing via Logit Bias

Read the original on arXiv Machine Learning →

The paper introduces a method for auditing the calibration of large language models (LLMs) that only exposes a logit_bias parameter. By mathematically manipulating this parameter, the authors can evaluate exact probability thresholds with a single query per sample, enabling a provably consistent estimator of True Calibration Error for binary tasks. This approach offers an efficient framework for auditing black‑box foundation models despite limited access to continuous output probabilities.

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 Machine Learning.

Related stories

arXiv AI
2d ago

Pinocchio: Fast Uncertainty Estimates for Black-Box Language Models

Pinocchio is an external calibrator that provides fast uncertainty estimates for black-box language models. It predicts the correctness of responses from seven trained LLMs with an AUROC of 0.862 and can transfer zero‑shot to thirteen unseen models from eight organizations. The method requires only a single forward pass and no access to the target model’s internal states, and a lightweight 0.8B checkpoint achieves comparable performance.

By Kevin David Hayes, Arka Pal, Haosong Zhang, Tom Goldstein, Micah Goldblum
llmsfine-tuning
More like this →
arXiv Machine Learning
Sep 11

Black-Box Membership Inference via Word-Level Probability Estimation

The paper introduces Word-level Probability MIA (WPMIA), a black-box membership inference attack that estimates word-level generation probabilities via Monte Carlo sampling and local kernel smoothing, then aggregates them into a sequence-level likelihood estimator. By conditioning on different prefixes, WPMIA amplifies distributional differences between member and non-member texts, outperforming existing black-box baselines on open-source LLMs and achieving an average TPR@5%FPR of 42.0 on proprietary models such as GPT‑5‑Chat, Gemini‑2.5‑Flash, and Claude‑4.5‑Haiku.

By Shengjie Niu, Yeheng Ge, Jian Huang
llmssafety
More like this →