# DA-RAC: Distance-Aware Calibration of LLM Judges for Trustworthy AI Auditing

arXiv: https://arxiv.org/abs/2608.14950 · HTML: https://mlsanity.com/paper/2608.14950/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 6 of its 16 references that are in the map.
- First posted on arXiv 15 Aug 2026.

## Builds on (most cited first, 6 of 6)

- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [JudgeBench: A Benchmark for Evaluating LLM-based Judges](https://mlsanity.com/paper/2410.12784/) (2024)
- [Benchmarking Foundation Models with Language-Model-as-an-Examiner](https://mlsanity.com/paper/2306.04181/) (2023)
- [Calibrating LLM-Based Evaluator](https://mlsanity.com/paper/2309.13308/) (2023)
- [ALLURE: Auditing and Improving LLM-based Evaluation of Text using Iterative In-Context-Learning](https://mlsanity.com/paper/2309.13701/) (2023)

## Abstract (arXiv)

Generative AI systems are increasingly producing real-world artifacts, however their efficacy and validity are often evaluated via context-free LLM-scoring. These judges can be miscalibrated by irrelevant in-context reference examples, creating false confidence and allowing low-quality or harmful outputs to pass evaluation. We study this failure mode as context-induced miscalibration and introduce DA-RAC, a distance-aware reference-anchored calibration method for LLM judges. DA-RAC retrieves semantically and structurally similar labeled anchors for each judgement scenario, weights them by distance, and exposes neighborhood difficulty as a calibration and triage signal. On multi-run LLM-judge evaluation benchmarks, it improves calibration and reduces false-pass risk relative to zero-shot, chain-of-thought evaluation, and static-anchor baselines. Mechanistic analysis shows that judge scores vary systematically with anchor distance, while static references can induce misleading decision boundaries. Thus LLM-judgement requires not only better models, but also calibrated, auditable reference selection, especially when automated evaluation is used to support high-impact AI generated artifacts. Judgments should be grounded in relevant, inspectable, and contestable interpretive artifacts.