Auditing Conformal Prediction for Language Models Under Inference-Time Model Shift

Key points

• We argue that this exposes a limitation in current CP evaluation protocols for language models.1
• We formalize the underlying phenomenon as inference-time model shift: a change in the conformal score distribution induced by inference configuration rather than by the input--label distribution, and theoretically characterize how it breaks split-conformal exchangeability even when the task distribution and trained model are unchanged.1
• We find that matched evaluation can substantially overstate deployment-time coverage, and that the effect is heterogeneous across models and datasets.1
• We argue that inference configuration should be reported as a first-class experimental axis in CP evaluations for language-model systems.1

Sources (1)

• [1]Auditing Conformal Prediction for Language Models Under Inference-Time Model Shift
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 20, 09:30 PM

  Conformal prediction (CP) for language models is typically evaluated under a matched protocol, in which calibration and deployment use the same post-training inference configuration, such as quantization mode and inference temperature.

  We argue that this exposes a limitation in current CP evaluation protocols for language models.

Extractive summary: sentences quoted from the sources.