BaseCal: Unsupervised Confidence Calibration via Base Model Signals

• URL: http://arxiv.org/abs/2601.03042v2
• Date: Thu, 08 Jan 2026 14:57:18 GMT
• Title: BaseCal: Unsupervised Confidence Calibration via Base Model Signals
• Authors: Hexiang Tan, Wanli Yang, Junwei Zhang, Xin Chen, Rui Tang, Du Su, Jingang Wang, Yuanzhuo Wang, Fei Sun, Xueqi Cheng,
• Abstract summary: Post-trained LLMs (PoLLMs) typically compromise trust with severe overconfidence.<n>We observe that their corresponding base LLMs often remain well-calibrated.<n>This naturally motivates us to calibrate PoLLM confidence using the base LLM as a reference.<n>A straightforward solution, BaseCal-ReEval, evaluates PoLLM's responses by feeding them into the base LLM to get average probabilities as confidence.<n>To address this, we propose BaseCal-Proj, which trains a lightweight projection to map the final-layer hidden states of PoLLMs back to those of their base
• Score: 65.11567272938616
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Reliable confidence is essential for trusting the outputs of LLMs, yet widely deployed post-trained LLMs (PoLLMs) typically compromise this trust with severe overconfidence. In contrast, we observe that their corresponding base LLMs often remain well-calibrated. This naturally motivates us to calibrate PoLLM confidence using the base LLM as a reference. This work proposes two ways to achieve this. A straightforward solution, BaseCal-ReEval, evaluates PoLLM's responses by feeding them into the base LLM to get average probabilities as confidence. While effective, this approach introduces additional inference overhead. To address this, we propose BaseCal-Proj, which trains a lightweight projection to map the final-layer hidden states of PoLLMs back to those of their base LLMs. These projected states are then processed by the base LLM's output layer to derive base-calibrated confidence for PoLLM's responses. Notably, BaseCal is an unsupervised, plug-and-play solution that operates without human labels or LLM modifications. Experiments across five datasets and three LLM families demonstrate the effectiveness of BaseCal, reducing Expected Calibration Error (ECE) by an average of 42.90\% compared to the best unsupervised baselines.

Related papers

• References Improve LLM Alignment in Non-Verifiable Domains [118.26447686644808]
  We investigate whether reference-guided LLM-evaluators can bridge the gap by serving as soft "verifiers"<n>We show that a reference-guided approach substantially improves the accuracy of less capable LLM-judges using references from frontier models.<n>We show that reference-guided self-improvement yields clear gains over both direct SFT on reference outputs and self-improvement with reference-free judges.
  arXiv  Detail & Related papers  (2026-02-18T19:03:34Z)
• Reasoning with Confidence: Efficient Verification of LLM Reasoning Steps via Uncertainty Heads [104.9566359759396]
  We propose a lightweight alternative for step-level reasoning verification based on data-driven uncertainty scores.<n>Our findings suggest that the internal states of LLMs encode their uncertainty and can serve as reliable signals for reasoning verification.
  arXiv  Detail & Related papers  (2025-11-09T03:38:29Z)
• LLM4VV: Evaluating Cutting-Edge LLMs for Generation and Evaluation of Directive-Based Parallel Programming Model Compiler Tests [7.6818904666624395]
  This paper proposes a dual-LLM system and experiments with the usage of LLMs for the generation of compiler tests.<n>It is evident that LLMs possess the promising potential to generate quality compiler tests and verify them automatically.
  arXiv  Detail & Related papers  (2025-07-29T02:34:28Z)
• Self-ensemble: Mitigating Confidence Mis-calibration for Large Language Models [67.62810111789338]
  Large Language Models exhibit a confidence distortion problem on multi-choice question-answering.<n>We propose Self-ensemble to solve this problem.<n> Experimental results on three LLMs and datasets demonstrate that Self-ensemble comprehensively addresses the confidence distortion problem.
  arXiv  Detail & Related papers  (2025-06-02T17:59:29Z)
• Lightweight Latent Verifiers for Efficient Meta-Generation Strategies [0.5892638927736115]
  Verifiers are auxiliary models that assess the correctness of outputs generated by base large language models (LLMs)<n>In this work, we introduce a novel lightweight verification approach, LiLaVe, which reliably extracts correctness signals from the hidden states of the base LLM.<n>A key advantage of LiLaVe is its ability to operate with only a small fraction of the computational budget required by traditional LLM-based verifiers.
  arXiv  Detail & Related papers  (2025-04-23T14:33:20Z)
• LLM-Lasso: A Robust Framework for Domain-Informed Feature Selection and Regularization [59.75242204923353]
  We introduce LLM-Lasso, a framework that leverages large language models (LLMs) to guide feature selection in Lasso regression.<n>LLMs generate penalty factors for each feature, which are converted into weights for the Lasso penalty using a simple, tunable model.<n>Features identified as more relevant by the LLM receive lower penalties, increasing their likelihood of being retained in the final model.
  arXiv  Detail & Related papers  (2025-02-15T02:55:22Z)
• Latent Space Chain-of-Embedding Enables Output-free LLM Self-Evaluation [45.059818539256426]
  We propose the Chain-of-Embedding (CoE) in the latent space to enable LLMs to perform output-free self-evaluation.<n>CoE consists of all progressive hidden states produced during the inference time, which can be treated as the latent thinking path of LLMs.
  arXiv  Detail & Related papers  (2024-10-17T15:09:24Z)
• Assessing the Reliability of Large Language Model Knowledge [78.38870272050106]
  Large language models (LLMs) have been treated as knowledge bases due to their strong performance in knowledge probing tasks. How do we evaluate the capabilities of LLMs to consistently produce factually correct answers? We propose MOdel kNowledge relIabiliTy scORe (MONITOR), a novel metric designed to directly measure LLMs' factual reliability.
  arXiv  Detail & Related papers  (2023-10-15T12:40:30Z)
• Quantifying Uncertainty in Answers from any Language Model and Enhancing their Trustworthiness [16.35655151252159]
  We introduce BSDetector, a method for detecting bad and speculative answers from a pretrained Large Language Model. Our uncertainty quantification technique works for any LLM accessible only via a black-box API.
  arXiv  Detail & Related papers  (2023-08-30T17:53:25Z)

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.