mlsanityData to 7 Oct 2026

Improving Uncertainty Estimation through Semantically Diverse Language Generation

arXiv 2406.04306 ↗First posted 6 Jun 2024cs.LG cs.AIcited by 34

Not on the map. Why?

Abstract (arXiv)

Large language models (LLMs) can suffer from hallucinations when generating text. These hallucinations impede various applications in society and industry by making LLMs untrustworthy. Current LLMs generate text in an autoregressive fashion by predicting and appending text tokens. When an LLM is uncertain about the semantic meaning of the next tokens to generate, it is likely to start hallucinating. Thus, it has been suggested that predictive uncertainty is one of the main causes of hallucinations. We introduce Semantically Diverse Language Generation (SDLG) to quantify predictive uncertainty in LLMs. SDLG steers the LLM to generate semantically diverse yet likely alternatives for an initially generated text. This approach provides a precise measure of aleatoric semantic uncertainty, detecting whether the initial text is likely to be hallucinated. Experiments on question-answering tasks demonstrate that SDLG consistently outperforms existing methods while being the most computationally efficient, setting a new standard for uncertainty estimation in LLMs.

Key facts. Cited by 34 papers in the map (data through 7 Oct 2026). Builds on 46 of its 56 references that are in the map. First posted on arXiv 6 Jun 2024.

Builds on

46 of 56 references.

References · 46

  1. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,416
  2. 2013Efficient Estimation of Word Representations in Vector Spacecited by 4,346
  3. 2022OPT: Open Pre-trained Transformer Language Modelscited by 3,464
  4. 2014Sequence to Sequence Learning with Neural Networkscited by 2,974
  5. 2021TruthfulQA: Measuring How Models Mimic Human Falsehoodscited by 2,919
  6. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,895
  7. 2019The Curious Case of Neural Text Degenerationcited by 2,881
  8. 2017A Broad-Coverage Challenge Corpus for Sentence Understanding through Inferencecited by 2,665
  9. 2020DeBERTa: Decoding-enhanced BERT with Disentangled Attentioncited by 2,157
  10. 2022Language Models (Mostly) Know What They Knowcited by 1,783
  11. 2018Sanity Checks for Saliency Mapscited by 1,149
  12. 2015A Diversity-Promoting Objective Function for Neural Conversation Modelscited by 1,068
34 more
  1. 2019Aleatoric and Epistemic Uncertainty in Machine Learning: An Introduction to Concepts and Methodscited by 903
  2. 2023Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generationcited by 795
  3. 2019Plug and Play Language Models: A Simple Approach to Controlled Text Generationcited by 793
  4. 2023Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedbackcited by 788
  5. 2023SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Modelscited by 787
  6. 2019CTRL: A Conditional Transformer Language Model for Controllable Generationcited by 754
  7. 2022Teaching Models to Express Their Uncertainty in Wordscited by 731
  8. 2021A Survey of Uncertainty in Deep Neural Networkscited by 675
  9. 2018CoQA: A Conversational Question Answering Challengecited by 674
  10. 2022Contrastive Decoding: Open-ended Text Generation as Optimizationcited by 598
  11. 2022News Summarization and Evaluation in the Era of GPT-3cited by 324
  12. 2024INSIDE: LLMs' Internal States Retain the Power of Hallucination Detectioncited by 317
  13. 2016Diverse Beam Search: Decoding Diverse Solutions from Neural Sequence Modelscited by 299
  14. 2023Generating with Confidence: Uncertainty Quantification for Black-box Large Language Modelscited by 286
  15. 2020Reducing conversational agents' overconfidence through linguistic calibrationcited by 209
  16. 2021On Hallucination and Predictive Uncertainty in Conditional Language Generationcited by 176
  17. 2023The Capacity for Moral Self-Correction in Large Language Modelscited by 171
  18. 2023Conformal Language Modelingcited by 163
  19. 2024Kernel Language Entropy: Fine-grained Uncertainty Quantification for LLMs from Semantic Similaritiescited by 156
  20. 2021Post-hoc Interpretability for Neural NLP: A Surveycited by 155
  21. 2023LM vs LM: Detecting Factual Errors via Cross Examinationcited by 155
  22. 2022A Survey of Controllable Text Generation using Transformer-based Pre-trained Language Modelscited by 128
  23. 2018Learning to Write with Cooperative Discriminatorscited by 119
  24. 2023Decomposing Uncertainty for Large Language Models through Input Clarification Ensemblingcited by 113
  25. 2022Uncertainty Quantification with Pre-trained Language Models: A Large-Scale Empirical Analysiscited by 104
  26. 2023Navigating the Grey Area: How Expressions of Uncertainty and Overconfidence Affect Language Modelscited by 91
  27. 2023Crawling the Internal Knowledge-Base of Language Modelscited by 85
  28. 2019Comparison of Diverse Decoding Methods from Conditional Language Modelscited by 78
  29. 2023Self-Evaluation Improves Selective Generation in Large Language Modelscited by 74
  30. 2020Exploring Controllable Text Generation Techniquescited by 55
  31. 2020CoCon: A Self-Supervised Approach for Controlled Text Generationcited by 47
  32. 2024MARS: Meaning-Aware Response Scoring for Uncertainty Estimation in Generative LLMscited by 45
  33. 2023Quantification of Uncertainty with Adversarial Modelscited by 28
  34. 2023Introducing an Improved Information-Theoretic Measure of Predictive Uncertaintycited by 28

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 7; 2025: 13; 2026: 14.

34 papers.

Sort:

By area:

  1. 2025Beyond Binary Rewards: Training LMs to Reason About Their Uncertaintycited by 107
  2. 2025Reasoning about Uncertainty: Do Reasoning Models Know When They Don't Know?cited by 35
  3. 2024Do LLMs estimate uncertainty well in instruction-following?cited by 26
  4. 2024I Don't Know: Explicit Modeling of Uncertainty with an [IDK] Tokencited by 26
  5. 2025Revisiting Uncertainty Quantification Evaluation in Language Models: Spurious Interactions with Response Length Bias Resultscited by 21
  6. 2025From Calibration to Collaboration: LLM Uncertainty Quantification Should Be More Human-Centeredcited by 21
  7. 2024Rethinking Uncertainty Estimation in LLMs: A Principled Single-Sequence Measurecited by 18
  8. 2024On Information-Theoretic Measures of Predictive Uncertaintycited by 16
  9. 2024Perceptions of Linguistic Uncertainty by Language Models and Humanscited by 15
  10. 2025GUARD: Glocal Uncertainty-Aware Robust Decoding for Effective and Efficient Open-Ended Text Generationcited by 10
  11. 2025Semantic-guided Diverse Decoding for Large Language Modelcited by 9
  12. 2025Addressing Pitfalls in the Evaluation of Uncertainty Estimation Methods for Natural Language Generationcited by 7
22 more
  1. 2024Efficiently Deploying LLMs with Controlled Riskcited by 5
  2. 2025Comparing Uncertainty Measurement and Mitigation Methods for Large Language Models: A Systematic Reviewcited by 3
  3. 2025Decoding Uncertainty: The Impact of Decoding Strategies for Uncertainty Estimation in Large Language Modelscited by 3
  4. 2026Confidence Before Answering: A Paradigm Shift for Efficient LLM Uncertainty Estimationcited by 3
  5. 2026Position: agentic AI orchestration should be Bayes-consistentcited by 3
  6. 2024Variability Need Not Imply Error: The Case of Adequate but Semantically Distinct Responsescited by 2
  7. 2025Generation Space Size: Understanding and Calibrating Open-Endedness of LLM Generationscited by 2
  8. 2025Efficient semantic uncertainty quantification in language models via diversity-steered samplingcited by 2
  9. 2026LLM-Confidence Reranker: A Training-Free Approach for Enhancing Retrieval-Augmented Generation Systemscited by 2
  10. 2026Improving Semantic Uncertainty Quantification in Language Model Question-Answering via Token-Level Temperature Scalingcited by 2
  11. 2026Towards Generation-Efficient Uncertainty Estimation in Large Language Modelscited by 2
  12. 2025Extending Epistemic Uncertainty Beyond Parameters Would Assist in Designing Reliable LLMscited by 1
  13. 2026Two Pathways to Truthfulness: On the Intrinsic Encoding of LLM Hallucinationscited by 1
  14. 2026AP-OOD: Attention Pooling for Out-of-Distribution Detectioncited by 1
  15. 2026CoMet: Context and Multiplicity Decomposition for Multimodal Uncertainty Estimationcited by 1
  16. 2025Open-Vocabulary Semantic Segmentation with Uncertainty Alignment for Robotic Scene Understanding in Indoor Building Environmentscited by 0
  17. 2026FAITH: Factuality Alignment through Integrating Trustworthiness and Honestnesscited by 0
  18. 2026Position: Uncertainty Quantification in LLMs is Just Unsupervised Clusteringcited by 0
  19. 2026From Critic to Confidence: PPO for Language-Based Quantitative Prediction with Confidence Estimationcited by 0
  20. 2026From Tokens to Semantics: Leveraging Complementary Signals for Hallucination Detection in Black-Box LLMscited by 0
  21. 2026Beyond Verbalized Confidence: Calibrating Reasoners with Differentiable Readoutscited by 0
  22. 2026Beyond Semantic Narrowing: Robust and Efficient LLM Watermarking with Hamming Neighborhoodscited by 0