Pith. sign in

REVIEW 5 cited by

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2508.06225 v3 pith:J6GC7RSI submitted 2025-08-08 cs.AI

Overconfidence in LLM-as-a-Judge: Diagnosis and Confidence-Driven Solution

classification cs.AI
keywords evaluationaccuracyadaptiveconfidenceconfidence-drivenrisk-awareexistingllm-as-a-judge
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

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.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Reliable LLM-Based Edge-Cloud-Expert Cascades for Telecom Knowledge Systems

    eess.SP 2025-12 accept novelty 7.0

    An edge-cloud-expert LLM cascade for telecom knowledge systems minimizes processing cost subject to misalignment-risk bounds via multiple hypothesis testing on knowledge and confidence scores.

  2. Different Teachers, Different Capabilities: Sub-1B On-Device Distillation for Structured Text Enrichment

    cs.AI 2026-07 conditional novelty 6.0

    Distilling an 8B reasoning teacher into a 0.6B student recovers most summary quality at ~50× speed, but teacher type—not scale alone—determines which capabilities transfer.

  3. VERDI: Single-Call Confidence Estimation for Verification-Based LLM Judges via Decomposed Inference

    cs.LG 2026-05 unverdicted novelty 6.0

    VERDI derives three structural confidence signals from decomposed LLM verification traces and calibrates them with Platt-scaled logistic regression to achieve AUROC 0.72-0.91 on GPT models and 0.56-0.70 on Qwen models...

  4. Calibrate, Don't Curate: Label-Efficient Estimation from Noisy LLM Judges

    stat.ME 2026-05 unverdicted novelty 6.0

    Calibrating the full set of LLM judges with labeled data halves calibration error versus top-5 accuracy selection on RewardBench2 and outperforms on four benchmarks.

  5. Semantic Intent Fragmentation: A Single-Shot Compositional Attack on Multi-Agent AI Pipelines

    cs.CR 2026-04 unverdicted novelty 6.0

    A single legitimate request can cause LLM orchestrators to output plans that violate security policies through the composition of benign subtasks, bypassing subtask-level checks.