LLMs exhibit epistemically vacuous confidence on clinical tabular data, but cross-model attribution divergence with XGBoost enables a calibrator that reduces expected calibration error from 0.254 to 0.080.
arXiv preprint arXiv:2502.00290 (2025)
11 Pith papers cite this work, alongside 1 external citations. Polarity classification is still indexing.
11
Pith papers citing it
1
external citations · external index
years
2026 11representative citing papers
Structural uncertainty from self-preference-induced rankings of LLM reasoning paths complements answer dispersion for identifying unreliable instances on logical tasks while collapsing on factual retrieval.
Velocity-field disagreement yields calibrated epistemic uncertainty for flow-based VLAs, supporting failure detection and uncertainty-guided active fine-tuning that cuts required samples by ≥22%.
GLU is a single-pass unsupervised uncertainty score for LLMs formed by multiplying global hidden-state geometric entropy with local token entropy, shown to match or beat baselines on three model families and six benchmarks while catching failure modes local signals miss.
Uncertainty estimation for LLM hallucinations can be done effectively with partial generations or input-only predictors, reducing the need for full autoregressive sampling.
DisAAD trains a 1%-sized proxy model via adversarial distillation to quantify uncertainty in black-box LLMs by aligning with their output distributions.
Cross-model semantic disagreement adds an epistemic uncertainty term that improves total uncertainty estimation over self-consistency alone, helping flag confident errors in LLMs.
VIHD detects hallucinations in medical MLLMs via visual dependency probing, token masking intervention, and calibrated semantic entropy on VQA benchmarks.
UE-DPO quantifies epistemic uncertainty from grounding failures to direct more learning pressure on hard visual tokens in preferred samples while easing penalties on dispreferred ones.
WebUncertainty improves web agent performance on benchmarks by adaptively selecting planning modes based on task uncertainty and using confidence-induced action uncertainty in MCTS to quantify aleatoric and epistemic uncertainty for better decisions.
citing papers explorer
-
LLM Doesn't Know What It Doesn't Know: Detecting Epistemic Blind Spots via Cross-Model Attribution Divergence on Clinical Tabular Data
LLMs exhibit epistemically vacuous confidence on clinical tabular data, but cross-model attribution divergence with XGBoost enables a calibrator that reduces expected calibration error from 0.254 to 0.080.
-
Quantifying Consistency in LLM Logical Reasoning via Structural Uncertainty
Structural uncertainty from self-preference-induced rankings of LLM reasoning paths complements answer dispersion for identifying unreliable instances on logical tasks while collapsing on factual retrieval.
-
Uncertainty Quantification for Flow-Based Vision-Language-Action Models
Velocity-field disagreement yields calibrated epistemic uncertainty for flow-based VLAs, supporting failure detection and uncertainty-guided active fine-tuning that cuts required samples by ≥22%.
-
Integrating Local and Global Entropy for Uncertainty Quantification in LLMs
GLU is a single-pass unsupervised uncertainty score for LLMs formed by multiplying global hidden-state geometric entropy with local token entropy, shown to match or beat baselines on three model families and six benchmarks while catching failure modes local signals miss.
-
Towards Generation-Efficient Uncertainty Estimation in Large Language Models
Uncertainty estimation for LLM hallucinations can be done effectively with partial generations or input-only predictors, reducing the need for full autoregressive sampling.
-
Estimating the Black-box LLM Uncertainty with Distribution-Aligned Adversarial Distillation
DisAAD trains a 1%-sized proxy model via adversarial distillation to quantify uncertainty in black-box LLMs by aligning with their output distributions.
-
Complementing Self-Consistency with Cross-Model Disagreement for Uncertainty Quantification
Cross-model semantic disagreement adds an epistemic uncertainty term that improves total uncertainty estimation over self-consistency alone, helping flag confident errors in LLMs.
-
VIHD: Visual Intervention-based Hallucination Detection for Medical Visual Question Answering
VIHD detects hallucinations in medical MLLMs via visual dependency probing, token masking intervention, and calibrated semantic entropy on VQA benchmarks.
-
Uncertainty-Aware Exploratory Direct Preference Optimization for Multimodal Large Language Models
UE-DPO quantifies epistemic uncertainty from grounding failures to direct more learning pressure on hard visual tokens in preferred samples while easing penalties on dispreferred ones.
-
WebUncertainty: Dual-Level Uncertainty Driven Planning and Reasoning For Autonomous Web Agent
WebUncertainty improves web agent performance on benchmarks by adaptively selecting planning modes based on task uncertainty and using confidence-induced action uncertainty in MCTS to quantify aleatoric and epistemic uncertainty for better decisions.
- From Accuracy to Auditability: A Survey of Determinism in Financial AI Systems