Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2508.06225 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913995153211392 |
|---|---|
| author | Tian, Zailong Han, Zhuoheng Chen, Yanzhe Xu, Haozhe Yang, Xi Xuan, Richeng Wang, Houfeng Liao, Lizi |
| author_facet | Tian, Zailong Han, Zhuoheng Chen, Yanzhe Xu, Haozhe Yang, Xi Xuan, Richeng Wang, Houfeng Liao, Lizi |
| contents | Large Language Models (LLMs) are widely used as automated judges, where practical value depends on both accuracy and trustworthy, risk-aware judgments. Existing approaches predominantly focus on accuracy, overlooking the necessity of well-calibrated confidence, which is vital for adaptive and reliable evaluation pipelines. In this work, we advocate a shift from accuracy-centric evaluation to confidence-driven, risk-aware LLM-as-a-Judge systems, emphasizing the necessity of well-calibrated confidence for trustworthy and adaptive evaluation. We systematically identify the Overconfidence Phenomenon in current LLM-as-a-Judges, where predicted confidence significantly overstates actual correctness, undermining reliability in practical deployment. To quantify this phenomenon, we introduce TH-Score, a novel metric measuring confidence-accuracy alignment. Furthermore, we propose LLM-as-a-Fuser, an ensemble framework that transforms LLMs into reliable, risk-aware evaluators. Extensive experiments demonstrate that our approach substantially improves calibration and enables adaptive, confidence-driven evaluation pipelines, achieving superior reliability and accuracy compared to existing baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_06225 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Overconfidence in LLM-as-a-Judge: Diagnosis and Confidence-Driven Solution Tian, Zailong Han, Zhuoheng Chen, Yanzhe Xu, Haozhe Yang, Xi Xuan, Richeng Wang, Houfeng Liao, Lizi Artificial Intelligence Large Language Models (LLMs) are widely used as automated judges, where practical value depends on both accuracy and trustworthy, risk-aware judgments. Existing approaches predominantly focus on accuracy, overlooking the necessity of well-calibrated confidence, which is vital for adaptive and reliable evaluation pipelines. In this work, we advocate a shift from accuracy-centric evaluation to confidence-driven, risk-aware LLM-as-a-Judge systems, emphasizing the necessity of well-calibrated confidence for trustworthy and adaptive evaluation. We systematically identify the Overconfidence Phenomenon in current LLM-as-a-Judges, where predicted confidence significantly overstates actual correctness, undermining reliability in practical deployment. To quantify this phenomenon, we introduce TH-Score, a novel metric measuring confidence-accuracy alignment. Furthermore, we propose LLM-as-a-Fuser, an ensemble framework that transforms LLMs into reliable, risk-aware evaluators. Extensive experiments demonstrate that our approach substantially improves calibration and enables adaptive, confidence-driven evaluation pipelines, achieving superior reliability and accuracy compared to existing baselines. |
| title | Overconfidence in LLM-as-a-Judge: Diagnosis and Confidence-Driven Solution |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2508.06225 |