A DETR-style probe distills multi-sample claim uncertainty into single-pass span detection and continuous Mixture-of-Beta scores, outperforming baselines on a new 293K-span benchmark.
hub Canonical reference
Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generation
Canonical reference. 86% of citing Pith papers cite this work as background.
81
Pith papers citing it
50
external citations · Pith
Background
86% of classified citations
abstract
We introduce a method to measure uncertainty in large language models. For tasks like question answering, it is essential to know when we can trust the natural language outputs of foundation models. We show that measuring uncertainty in natural language is challenging because of "semantic equivalence" -- different sentences can mean the same thing. To overcome these challenges we introduce semantic entropy -- an entropy which incorporates linguistic invariances created by shared meanings. Our method is unsupervised, uses only a single model, and requires no modifications to off-the-shelf language models. In comprehensive ablation studies we show that the semantic entropy is more predictive of model accuracy on question answering data sets than comparable baselines.
hub tools
citation-role summary
background 6
baseline 1
citation-polarity summary
representative citing papers
Empirical evaluation of three LLMs finds prevalent overconfidence in insecure code generation, with security calibration outperforming functional calibration but both degrading in repository-level settings.
Models delayed verification in multi-agent LLMs as graph consensus, derives stability thresholds (inverse golden ratio for delay two) via grounded Laplacian, and gives a supermodular greedy rule for corrector placement; experiments on five models confirm dose-delay oscillations.
SPOT-E uses entropy shaping on answer predictions with low-entropy anchors to optimize visual spotlights at test time via GRPO for better VLM performance on evidence-intensive tasks.
MortarBench benchmark shows LLMs achieve ≤77.1% accuracy on loan origination; CRIT calibration raises accuracy to 80.5% and reduces bias.
ActProbe is an action-space detector that uses temporal consistency error and action chunk magnitude from policy outputs, mapped via LSTM-MLP, to predict failures earlier than baselines across policies and real-robot tasks.
Introduces object-level semantic uncertainty for VLM memory, the UQ-DAAAM refinement system, and probabilistic guarantees that selected high-quality views reduce uncertainty more effectively.
Chunk-Level Guided Generation uses off-the-shelf large LLMs to score fixed-length chunks from small models via likelihoods, matching trained PRM performance on math benchmarks without reward-model training.
A per-token feature from temperature-induced changes in LLM token distributions predicts within-prompt creativity rank at Spearman rho 0.918 vs LLM judges and 0.870 vs humans, outperforming perplexity, entropy, top-1 margin, and compression baselines.
A between-subjects experiment (N=192) finds that token-level uncertainty increases agreement with LLM answers while relation-level uncertainty reduces external verification in medical decision tasks.
Introduces Trajectory Proper Score (TPS) as a strictly proper family of trajectory-level scoring rules that elicits the complete prefix-conditioned success probability process.
Proposes Spatial Narrative Score (SNS) evaluation for VLMs' camera motion understanding and introduces CaMo model achieving consistent performance on SNS and direct QA.
Inducing artificial uncertainty on trivial tasks allows training probes that achieve higher calibration on hard data than standard approaches while retaining performance on easy data.
Evidence utility is defined as information gain on the model's output distribution, with ranking by gain on a latent helpfulness variable shown equivalent to answer-space utility under mild assumptions, enabling a training-free surrogate framework that outperforms baselines.
BICR trains a lightweight probe on contrastive hidden states from real versus blind images to detect visual grounding in LVLM predictions, outperforming baselines on calibration and discrimination with fewer parameters.
Task calibration aligns LLM distributions in latent task spaces to make MBR decoding provably optimal and improve generation quality.
Proposes surrogate semantic entropy stratification followed by approximate Neyman allocation for active testing of LLMs on generative benchmarks, reporting up to 28% MSE reduction and 22.9% average budget savings versus uniform sampling.
LLM surrogate beliefs under sparse observations depend on prompts and query protocols, with structural prompts as priors, pointwise vs joint querying producing different beliefs, and sequential evidence causing non-monotonic updates that affect acquisition and regret.
Uncertainty and correctness in LLMs are encoded by distinct feature populations, with suppression of confounded features improving accuracy and reducing entropy.
CoT prompting in LLM4Code shows mixed robustness that depends on model family, task structure, and perturbations destabilizing structural anchors, leading to trajectory deformations like lengthening, branching, and simplification.
A new Latent Imagination Module uses cross-attention to predict latent visual embeddings from text, improving accuracy and calibration of vision-language models on text-only inputs.
Self-consistency is a weak, regime-dependent proxy for correctness: positive but small correlations (rho 0.20–0.59), with the most self-consistent frontier model over-confident and wrong 48% of the time at high agreement.
Unsupervised MLP activation dispersion separates known from fabricated entities at AUROC 0.95–1.00 across Bielik scales, while factual reliability scales separately and refusals stay near zero.
A large-scale multilingual evaluation of LLM uncertainty estimation methods across 22 languages and 9 models finds that English reasoning closes the UE gap for low-resource languages and that optimal UE method choice depends on model scale.
citing papers explorer
-
SpanUQ: Span-Level Uncertainty Quantification for Large Language Model Generation
A DETR-style probe distills multi-sample claim uncertainty into single-pass span detection and continuous Mixture-of-Beta scores, outperforming baselines on a new 293K-span benchmark.
-
An Empirical Study of Security Calibration in Large Language Models for Code
Empirical evaluation of three LLMs finds prevalent overconfidence in insecure code generation, with security calibration outperforming functional calibration but both degrading in repository-level settings.
-
Delayed Verification Destabilizes Multi-Agent LLM Belief: Instability Thresholds and Optimal Corrector Placement
Models delayed verification in multi-agent LLMs as graph consensus, derives stability thresholds (inverse golden ratio for delay two) via grounded Laplacian, and gives a supermodular greedy rule for corrector placement; experiments on five models confirm dose-delay oscillations.
-
SPOT-E: Test-Time Entropy Shaping with Visual Spotlights for Frozen VLMs
SPOT-E uses entropy shaping on answer predictions with low-entropy anchors to optimize visual spotlights at test time via GRPO for better VLM performance on evidence-intensive tasks.
-
MortarBench: Evaluating Mortgage Loan Origination Agents
MortarBench benchmark shows LLMs achieve ≤77.1% accuracy on loan origination; CRIT calibration raises accuracy to 80.5% and reduces bias.
-
ActProbe: Action-Space Probe for Early Failure Detection of Generative Robot Policies
ActProbe is an action-space detector that uses temporal consistency error and action chunk magnitude from policy outputs, mapped via LSTM-MLP, to predict failures earlier than baselines across policies and real-robot tasks.
-
Remember with Confidence: Uncertainty Quantification for Spatio-temporal Memory with Probabilistic Guarantees
Introduces object-level semantic uncertainty for VLM memory, the UQ-DAAAM refinement system, and probabilistic guarantees that selected high-quality views reduce uncertainty more effectively.
-
Off-the-Shelf LLMs as Process Scorers: Training-Free Alternative to PRMs for Mathematical Reasoning
Chunk-Level Guided Generation uses off-the-shelf large LLMs to score fixed-length chunks from small models via likelihoods, matching trained PRM performance on math benchmarks without reward-model training.
-
Before and After Temperature: A Distributional View of Creative LLM Generation
A per-token feature from temperature-induced changes in LLM token distributions predicts within-prompt creativity rank at Spearman rho 0.918 vs LLM judges and 0.870 vs humans, outperforming perplexity, entropy, top-1 margin, and compression baselines.
-
Not All Uncertainty Is Equal: How Uncertainty Granularity Shapes Human Verification in LLM-Assisted Decision Making
A between-subjects experiment (N=192) finds that token-level uncertainty increases agreement with LLM answers while relation-level uncertainty reduces external verification in medical decision tasks.
-
Proper Scoring Rules for Agentic Uncertainty Quantification
Introduces Trajectory Proper Score (TPS) as a strictly proper family of trajectory-level scoring rules that elicits the complete prefix-conditioned success probability process.
-
CaMo: Camera Motion Grounded Evaluation and Training for Vision-Language Models
Proposes Spatial Narrative Score (SNS) evaluation for VLMs' camera motion understanding and introduces CaMo model achieving consistent performance on SNS and direct QA.
-
Inducing Artificial Uncertainty in Language Models
Inducing artificial uncertainty on trivial tasks allows training probes that achieve higher calibration on hard data than standard approaches while retaining performance on easy data.
-
Utility-Oriented Visual Evidence Selection for Multimodal Retrieval-Augmented Generation
Evidence utility is defined as information gain on the model's output distribution, with ranking by gain on a latent helpfulness variable shown equivalent to answer-space utility under mild assumptions, enabling a training-free surrogate framework that outperforms baselines.
-
Grounded or Guessing? LVLM Confidence Estimation via Blind-Image Contrastive Ranking
BICR trains a lightweight probe on contrastive hidden states from real versus blind images to detect visual grounding in LVLM predictions, outperforming baselines on calibration and discrimination with fewer parameters.
-
Task-Aware Calibration: Provably Optimal Decoding in LLMs
Task calibration aligns LLM distributions in latent task spaces to make MBR decoding provably optimal and improve generation quality.
-
Active Testing of Large Language Models via Approximate Neyman Allocation
Proposes surrogate semantic entropy stratification followed by approximate Neyman allocation for active testing of LLMs on generative benchmarks, reporting up to 28% MSE reduction and 22.9% average budget savings versus uniform sampling.
-
Elicitation Matters: How Prompts and Query Protocols Shape LLM Surrogates under Sparse Observations
LLM surrogate beliefs under sparse observations depend on prompts and query protocols, with structural prompts as priors, pointwise vs joint querying producing different beliefs, and sequential evidence causing non-monotonic updates that affect acquisition and regret.
-
Are LLM Uncertainty and Correctness Encoded by the Same Features? A Functional Dissociation via Sparse Autoencoders
Uncertainty and correctness in LLMs are encoded by distinct feature populations, with suppression of confounded features improving accuracy and reducing entropy.
-
Structural Anchors and Reasoning Fragility:Understanding CoT Robustness in LLM4Code
CoT prompting in LLM4Code shows mixed robustness that depends on model family, task structure, and perturbations destabilizing structural anchors, leading to trajectory deformations like lengthening, branching, and simplification.
-
Bridging the Missing-Modality Gap: Improving Text-Only Calibration of Vision Language Models
A new Latent Imagination Module uses cross-attention to predict latent visual embeddings from text, improving accuracy and calibration of vision-language models on text-only inputs.
-
When LLMs Agree, Are They Right? Auditing Self-Consistency and Cross-Model Agreement as Confidence Signals
Self-consistency is a weak, regime-dependent proxy for correctness: positive but small correlations (rho 0.20–0.59), with the most self-consistent frontier model over-confident and wrong 48% of the time at high agreement.
-
Does Bielik Know What It Doesn't Know? Activation Dispersion Separates Entity Familiarity from Factual Reliability Across Model Scale
Unsupervised MLP activation dispersion separates known from fabricated entities at AUROC 0.95–1.00 across Bielik scales, while factual reliability scales separately and refusals stay near zero.
-
Estimating Uncertainty from Reasoning: A Large-Scale Study of Multi- and Crosslingual MCQA Performance in LLMs
A large-scale multilingual evaluation of LLM uncertainty estimation methods across 22 languages and 9 models finds that English reasoning closes the UE gap for low-resource languages and that optimal UE method choice depends on model scale.
-
CALIBER: Calibrating Confidence Before and After Reasoning in Language Models
CALIBER elicits and supervises pre-reasoning confidence with prompt-level success probability and post-reasoning confidence with answer-level correctness, cutting ECE by 52.5% on BigMathDigits for a 7B model while remaining competitive on accuracy.
-
The Score Granularity Gap in Black-Box LLM Classification: A Comparative Study of Confidence Constructions
Comparative evaluation of seven confidence constructions across 25 LLM-dataset pairs reveals that verbalized scores provide good ranking but coarse granularity for thresholding, while multi-query aggregation helps weak models but can harm strong ones.
-
Confidence Laundering in Agent Systems: Why Uncertainty Needs a Latent Carrier
Agent systems lose uncertainty at decision handoffs, causing downstream over-trust; the paper proposes latent uncertainty as a carrier to preserve pre-commitment fragility across interfaces.
-
FASE: Fast Adaptive Semantic Entropy for Code Quality
FASE approximates functional correctness via MST on structural and semantic dissimilarity graphs, reporting 25% better Spearman correlation and 19% better ROCAUC than LLM-based semantic entropy at 0.3% runtime cost on HumanEval and BigCodeBench.
-
CRAFT: Cost-aware Refinement And Front-aware Tuning of Prompts
CRAFT is a Pareto-front prompt optimizer that allocates scarce LLM validation calls to candidates near the current front using accuracy- and cost-oriented generators plus NSGA-II retention.
-
Learning to Solve, Forgetting to Retain: Correct-Set Turnover in RLVR
RLVR exhibits correct-set turnover where solved problems regress during training, and a periodic review mechanism exploiting a repair-window principle improves retention and performance over baselines.
-
TriLens: Per-Layer Logit-Lens Entropy for White-Box Hallucination Detection
TriLens detects hallucinations via per-layer entropy trajectories of logit-lens readouts from three internal modules across LLMs and QA benchmarks.
-
Functional Entropy: Predicting Functional Correctness in LLM-Generated Code with Uncertainty Quantification
Introduces functional equivalence methods and functional entropy to predict functional correctness of LLM-generated code via uncertainty quantification, outperforming NLI-based baselines in most tested settings.
-
Reading Calibrated Uncertainty from Language Model Trajectories
Trajectory geometry of MLP write-vectors, summarized by 11 scale-invariant features in a sparse linear probe, beats MSP for selective abstention in LLM multiple-choice QA.
-
BalanceRAG: Joint Risk Calibration for Cascaded Retrieval-Augmented Generation
BalanceRAG uses sequential graphical testing on a 2D lattice of threshold pairs to certify safe operating points that meet target risk levels in cascaded RAG while increasing coverage.
-
LLMs as Implicit Imputers: Uncertainty Should Scale with Missing Information
Response entropy in LLMs rises with missing context on SQuAD while sampling-based confidence stays high, supporting the multiple imputation criterion and introducing a diagnostic for uncertainty reduction by context level.
-
REALISTA: Realistic Latent Adversarial Attacks that Elicit LLM Hallucinations
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-source and reasoning LLMs.
-
Uncertainty Quantification for LLM-based Code Generation
RisCoSet applies multiple hypothesis testing to construct risk-controlling partial-program prediction sets for LLM code generation, achieving up to 24.5% less code removal than prior methods at equivalent risk levels.
-
Annotations Mitigate Post-Training Mode Collapse
Annotation-anchored training reduces semantic diversity collapse in post-trained language models by a factor of six compared to standard supervised fine-tuning while preserving instruction-following and improving with scale.
-
The Geometry of Forgetting: Temporal Knowledge Drift as an Independent Axis in LLM Representations
Temporal knowledge drift is encoded as a geometrically orthogonal direction in LLM residual streams, independent of correctness and uncertainty.
-
Measuring Black-Box Confidence via Reasoning Trajectories: Geometry, Coverage, and Verbalization
Trajectory geometry in embedding space fused with coverage and verbalization yields better black-box CoT confidence estimation than self-consistency at lower sample counts across six benchmark-reasoner pairs.
-
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.
-
Geometry-Calibrated Conformal Abstention for Language Models
Geometry-calibrated conformal abstention lets language models abstain from uncertain queries with finite-sample guarantees on both participation rate and conditional correctness of answers.
-
Trace-Level Analysis of Information Contamination in Multi-Agent Systems
Agent workflows can diverge substantially from contaminated inputs yet recover correct answers, or stay similar while failing, as measured by trace divergence on GAIA tasks.
-
The Surprising Universality of LLM Outputs: A Real-Time Verification Primitive
LLM token rank-frequency distributions converge to a shared Mandelbrot distribution across models and domains, enabling a microsecond-scale statistical primitive for provenance verification and black-box anomaly triage.
-
Detecting Hallucinations for Large Language Model-based Knowledge Graph Reasoning
LUCID detects hallucinations in LLM-KG reasoning by extracting node/edge features from attention and semantics then integrating them with KG structure in a GNN, achieving SOTA on nine new benchmark datasets versus 15 baselines.
-
Beyond One Output: Visualizing and Comparing Distributions of Language Model Generations
A merged-token graph of many LM generations helps users compare output diversity, while raw lists remain better for detail-oriented distributional questions.
-
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.
-
CLSGen: A Dual-Head Fine-Tuning Framework for Joint Probabilistic Classification and Verbalized Explanation
CLSGen is a dual-head LLM fine-tuning framework that enables joint probabilistic classification and verbalized explanation generation without catastrophic forgetting of generative capabilities.
-
Ensemble-Based Uncertainty Estimation for Code Correctness Estimation
Ensemble Semantic Entropy improves correlation with code correctness over single-model methods and powers a cascading scaling system that cuts FLOPs by 64.9% while preserving performance on LiveCodeBench.
-
ETS: Energy-Guided Test-Time Scaling for Training-Free RL Alignment
ETS performs training-free RL alignment for language models by energy-guided test-time scaling with Monte Carlo energy estimation and importance sampling acceleration.