pith. sign in

arxiv: 2405.20003 · v1 · pith:LRQ7NKHRnew · submitted 2024-05-30 · 💻 cs.LG · cs.AI· cs.CL

Kernel Language Entropy: Fine-grained Uncertainty Quantification for LLMs from Semantic Similarities

classification 💻 cs.LG cs.AIcs.CL
keywords uncertaintysemanticentropylanguagellmsquantificationanswerscalled
0
0 comments X
read the original abstract

Uncertainty quantification in Large Language Models (LLMs) is crucial for applications where safety and reliability are important. In particular, uncertainty can be used to improve the trustworthiness of LLMs by detecting factually incorrect model responses, commonly called hallucinations. Critically, one should seek to capture the model's semantic uncertainty, i.e., the uncertainty over the meanings of LLM outputs, rather than uncertainty over lexical or syntactic variations that do not affect answer correctness. To address this problem, we propose Kernel Language Entropy (KLE), a novel method for uncertainty estimation in white- and black-box LLMs. KLE defines positive semidefinite unit trace kernels to encode the semantic similarities of LLM outputs and quantifies uncertainty using the von Neumann entropy. It considers pairwise semantic dependencies between answers (or semantic clusters), providing more fine-grained uncertainty estimates than previous methods based on hard clustering of answers. We theoretically prove that KLE generalizes the previous state-of-the-art method called semantic entropy and empirically demonstrate that it improves uncertainty quantification performance across multiple natural language generation datasets and LLM architectures.

This paper has not been read by Pith yet.

discussion (0)

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

Forward citations

Cited by 8 Pith papers

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

  1. REALISTA: Realistic Latent Adversarial Attacks that Elicit LLM Hallucinations

    cs.CL 2026-05 unverdicted novelty 8.0

    REALISTA optimizes continuous combinations of valid editing directions in latent space to produce realistic adversarial prompts that elicit hallucinations more effectively than prior methods, including on large reason...

  2. Can LLMs Use Linguistic Uncertainty Markers to Reliably Reflect Intrinsic Confidence?

    cs.CL 2026-05 unverdicted novelty 7.0

    LLMs struggle to associate epistemic markers with stable internal confidence levels across distributions, even under model-centric interpretations, while maintaining somewhat consistent marker rankings.

  3. Reinforcement Learning with Metacognitive Feedback Elicits Faithful Uncertainty Expression in LLMs

    cs.CL 2026-06 unverdicted novelty 6.0

    RLMF uses quality of model self-judgments to refine RL rankings and select training data, achieving SOTA faithful calibration while preserving accuracy and outperforming standard RL by up to 63%.

  4. REALISTA: Realistic Latent Adversarial Attacks that Elicit LLM Hallucinations

    cs.CL 2026-05 unverdicted novelty 6.0

    REALISTA generates semantically coherent adversarial prompts via latent-space optimization over input-dependent editing directions, achieving stronger hallucination elicitation than prior realistic attacks on open-sou...

  5. Mind the Unseen Mass: Unmasking LLM Hallucinations via Soft-Hybrid Alphabet Estimation

    cs.CL 2026-04 unverdicted novelty 6.0

    SHADE adaptively combines coverage and spectral signals to estimate semantic alphabet size from few LLM samples, yielding better performance than baselines in low-sample regimes for alphabet estimation and QA error detection.

  6. Token-Level Density-Based Uncertainty Quantification Methods for Eliciting Truthfulness of Large Language Models

    cs.CL 2025-02 unverdicted novelty 6.0

    Adapts multi-layer token-level Mahalanobis distance with supervised linear regression to yield improved uncertainty scores for LLM truthfulness tasks.

  7. Unconditional Truthfulness: Learning Unconditional Uncertainty of Large Language Models

    cs.CL 2024-08 unverdicted novelty 6.0

    A regression model using attention features and recurrent uncertainty scores improves selective generation in LLMs over unsupervised and supervised baselines on ten datasets and three models.

  8. Confident in a Confidence Score: Investigating the Sensitivity of Confidence Scores to Supervised Fine-Tuning

    cs.CL 2026-04 unverdicted novelty 5.0

    Supervised fine-tuning degrades the correlation between confidence scores and output quality in language models, driven by factors like training distribution similarity rather than true quality.