Domain-Shift-Aware Conformal Prediction for Large Language Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910278028886016 |
|---|---|
| author | Lin, Zhexiao Li, Yuanyuan Sarna, Neeraj Gao, Yuanyuan von Gablenz, Michael |
| author_facet | Lin, Zhexiao Li, Yuanyuan Sarna, Neeraj Gao, Yuanyuan von Gablenz, Michael |
| contents | Large language models have achieved impressive performance across diverse tasks. However, their tendency to produce overconfident and factually incorrect outputs, known as hallucinations, poses risks in real-world applications. Conformal prediction provides finite-sample, distribution-free coverage guarantees, but standard conformal prediction breaks down under domain shift, often leading to under-coverage and unreliable prediction sets. We propose a new framework called Domain-Shift-Aware Conformal Prediction (DS-CP). Our framework adapts conformal prediction to large language models under domain shift, by systematically reweighting calibration samples based on their proximity to the test prompt, thereby preserving validity while enhancing adaptivity. Our theoretical analysis and experiments on the MMLU benchmark demonstrate that the proposed method delivers more reliable coverage than standard conformal prediction, especially under substantial distribution shifts, while maintaining efficiency. This provides a practical step toward trustworthy uncertainty quantification for large language models in real-world deployment. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_05566 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Domain-Shift-Aware Conformal Prediction for Large Language Models Lin, Zhexiao Li, Yuanyuan Sarna, Neeraj Gao, Yuanyuan von Gablenz, Michael Machine Learning Artificial Intelligence Computation and Language Applications Large language models have achieved impressive performance across diverse tasks. However, their tendency to produce overconfident and factually incorrect outputs, known as hallucinations, poses risks in real-world applications. Conformal prediction provides finite-sample, distribution-free coverage guarantees, but standard conformal prediction breaks down under domain shift, often leading to under-coverage and unreliable prediction sets. We propose a new framework called Domain-Shift-Aware Conformal Prediction (DS-CP). Our framework adapts conformal prediction to large language models under domain shift, by systematically reweighting calibration samples based on their proximity to the test prompt, thereby preserving validity while enhancing adaptivity. Our theoretical analysis and experiments on the MMLU benchmark demonstrate that the proposed method delivers more reliable coverage than standard conformal prediction, especially under substantial distribution shifts, while maintaining efficiency. This provides a practical step toward trustworthy uncertainty quantification for large language models in real-world deployment. |
| title | Domain-Shift-Aware Conformal Prediction for Large Language Models |
| topic | Machine Learning Artificial Intelligence Computation and Language Applications |
| url | https://arxiv.org/abs/2510.05566 |