Know When You're Wrong: Aligning Confidence with Correctness for LLM Error Detection

• URL: http://arxiv.org/abs/2603.06604v1
• Date: Wed, 18 Feb 2026 07:05:12 GMT
• Title: Know When You're Wrong: Aligning Confidence with Correctness for LLM Error Detection
• Abstract summary: Large language models (LLMs) are increasingly deployed in critical decision-making systems.<n>We introduce a normalized confidence score based on output anchor token probabilities.<n>This enables direct detection of errors and hallucinations with minimal overhead.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: As large language models (LLMs) are increasingly deployed in critical decision-making systems, the lack of reliable methods to measure their uncertainty presents a fundamental trustworthiness risk. We introduce a normalized confidence score based on output anchor token probabilities: classification labels for structured tasks and self-evaluation responses (Yes/No) for open-ended generation. This enables direct detection of errors and hallucinations with minimal overhead and without external validation. We make three key contributions. First, we propose a normalized confidence score and self-evaluation framework that exposes reliable confidence estimates for error detection across seven diverse benchmark tasks and five LLMs of varying architectures and sizes. Second, our theoretical analysis reveals that supervised fine-tuning (SFT) yields well-calibrated confidence through maximum-likelihood estimation, whereas reinforcement learning methods (PPO, GRPO) and DPO induce overconfidence via reward exploitation. Third, we propose post-RL SFT with self-distillation to restore confidence reliability in RL-trained models. Empirical results demonstrated that SFT improved average confidence-correctness AUROC from 0.806 to 0.879 and reduced calibration error from 0.163 to 0.034 on Qwen3-4B, while GRPO and DPO degraded confidence reliability. We demonstrated practical value through adaptive retrieval-augmented generation (RAG) that selectively retrieves context when the model lacks confidence, using only 58\% of retrieval operations to recover 95\% of the maximum achievable accuracy gain on TriviaQA

Related papers

• $TCP_α$: Margin-Controlled Confidence estimation for reliable Music Information Retrieval [7.679660052685777]
  Deep neural networks are often overconfident, assigning high confidence even to incorrect predictions.<n>Existing targets suffer from inherent ambiguity: they assign overlapping confidence values to correct and incorrect predictions, while errors near the decision boundary receive confidence scores indistinguishable from correct predictions.<n>We propose $ TCP_$, a novel confidence target that resolves these limitations by introducing a margin-controlled penalty for misclassified samples.
  arXiv  Detail & Related papers  (2026-08-20T17:58:50Z)
• Can You Trust the Confidence? ConfBench for Vision-Language Models on Document Extraction [13.176724997257802]
  Existing document benchmarks are dominated by clean, high-quality samples.<n>We introduce ConfBench, the first calibration-specific benchmark for key information extraction.<n>We evaluate four proprietary and three open-weight VLMs under verbalized and log-probability confidence estimation methods.
  arXiv  Detail & Related papers  (2026-08-03T07:03:51Z)
• Process Supervision of Confidence Margin for Calibrated LLM Reasoning [52.373121066425455]
  Scaling test-time computation with reinforcement learning (RL) has emerged as a reliable path to improve large language models (LLM) reasoning ability.<n>Yet, outcome-based reward often incentivizes models to be overconfident, leading to hallucinations, unreliable confidence-based control, and unnecessary compute allocation.<n>We introduce Reinforcement Learning with Confidence Margin (textbfRLCM), a calibration-aware RL framework that jointly optimize correctness and confidence reliability.
  arXiv  Detail & Related papers  (2026-04-25T14:40:13Z)
• Enhancing Confidence Estimation in Telco LLMs via Twin-Pass CoT-Ensembling [1.8332654441845688]
  Large Language Models (LLMs) are increasingly applied to complex telecommunications tasks.<n>LLMs-generated confidence scores are often biased and unreliable, frequently exhibiting systematic overconfidence.<n>We propose a novel Twin-Pass Chain of Thought (CoT)-Ensembling methodology for improving confidence estimation.
  arXiv  Detail & Related papers  (2026-04-14T20:03:08Z)
• BrowseConf: Confidence-Guided Test-Time Scaling for Web Agents [58.05949210993854]
  We investigate whether search agents have the ability to communicate their own confidence through verbalized confidence scores after long sequences of actions.<n>We propose Test-Time Scaling (TTS) methods that use confidence scores to determine answer quality, encourage the model to try again until reaching a satisfactory confidence level.
  arXiv  Detail & Related papers  (2025-10-27T15:58:51Z)
• Confidence-Based Response Abstinence: Improving LLM Trustworthiness via Activation-Based Uncertainty Estimation [7.3923284353934875]
  We propose a method for confidence estimation in retrieval-augmented generation (RAG) systems that aligns closely with the correctness of large language model (LLM) outputs.<n>Our approach extends prior uncertainty quantification methods by leveraging raw feed-forward network (FFN) activations as auto-regressive signals.<n>Our results demonstrate that activation-based confidence modeling offers a scalable, architecture-aware path toward trustworthy RAG deployment.
  arXiv  Detail & Related papers  (2025-10-15T16:55:56Z)
• ConfTuner: Training Large Language Models to Express Their Confidence Verbally [58.63318088243125]
  Large Language Models (LLMs) are increasingly deployed in high-stakes domains such as science, law, and healthcare.<n>LLMs are often observed to generate incorrect answers with high confidence, a phenomenon known as "overconfidence"
  arXiv  Detail & Related papers  (2025-08-26T09:25:32Z)
• Mind the Generation Process: Fine-Grained Confidence Estimation During LLM Generation [63.49409574310576]
  Large language models (LLMs) exhibit overconfidence, assigning high confidence scores to incorrect predictions.<n>We introduce FineCE, a novel confidence estimation method that delivers accurate, fine-grained confidence scores during text generation.<n>Our code and all baselines used in the paper are available on GitHub.
  arXiv  Detail & Related papers  (2025-08-16T13:29:35Z)
• Overconfidence in LLM-as-a-Judge: Diagnosis and Confidence-Driven Solution [20.607071807794195]
  Large Language Models (LLMs) are widely used as automated judges, where practical value depends on both accuracy and trustworthy, risk-aware judgments.<n>Existing approaches predominantly focus on accuracy, overlooking the necessity of well-calibrated confidence.<n>We advocate a shift from accuracy-centric evaluation to confidence-driven, risk-aware LLM-as-a-Judge systems.
  arXiv  Detail & Related papers  (2025-08-08T11:11:22Z)
• The Confidence Paradox: Can LLM Know When It's Wrong [5.445980143646736]
  Document Visual Question Answering (DocVQA) models often produce overconfident or ethically misaligned responses.<n>We propose HonestVQA, a model-agnostic, self-supervised framework that aligns model confidence with correctness using weighted loss and contrastive learning.
  arXiv  Detail & Related papers  (2025-06-30T02:06:54Z)
• Fact-Level Confidence Calibration and Self-Correction [64.40105513819272]
  We propose a Fact-Level framework that calibrates confidence to relevance-weighted correctness at the fact level. We also develop Confidence-Guided Fact-level Self-Correction ($textbfConFix$), which uses high-confidence facts within a response as additional knowledge to improve low-confidence ones.
  arXiv  Detail & Related papers  (2024-11-20T14:15:18Z)
• Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedback [91.22679548111127]
  A trustworthy real-world prediction system should produce well-calibrated confidence scores. We show that verbalized confidences emitted as output tokens are typically better-calibrated than the model's conditional probabilities.
  arXiv  Detail & Related papers  (2023-05-24T10:12:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.