mlsanityData to 14 Oct 2026

Measuring Reliability of Large Language Models through Semantic Consistency

arXiv 2211.05853 ↗First posted 10 Nov 2022cs.CL cs.AI cs.CYcited by 38

Not on the map. Why?

Abstract (arXiv)

While large pretrained language models (PLMs) demonstrate incredible fluency and performance on many natural language tasks, recent work has shown that well-performing PLMs are very sensitive to what prompts are feed into them. Even when prompts are semantically identical, language models may give very different answers. When considering safe and trustworthy deployments of PLMs we would like their outputs to be consistent under prompts that mean the same thing or convey the same intent. While some work has looked into how state-of-the-art PLMs address this need, they have been limited to only evaluating lexical equality of single- or multi-word answers and do not address consistency of generative text sequences. In order to understand consistency of PLMs under text generation settings, we develop a measure of semantic consistency that allows the comparison of open-ended text outputs. We implement several versions of this consistency metric to evaluate the performance of a number of PLMs on paraphrased versions of questions in the TruthfulQA dataset, we find that our proposed metrics are considerably more consistent than traditional metrics embodying lexical consistency, and also correlate with human evaluation of output consistency to a higher degree.

Key facts. Cited by 38 papers in the map (data through 14 Oct 2026). Builds on 24 of its 28 references that are in the map. First posted on arXiv 10 Nov 2022.

Builds on

24 of 28 references.

References · 24

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2022Training language models to follow instructions with human feedbackcited by 15,971
  3. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,512
  4. 2022OPT: Open Pre-trained Transformer Language Modelscited by 3,512
  5. 2021TruthfulQA: Measuring How Models Mimic Human Falsehoodscited by 2,975
  6. 2019The Curious Case of Neural Text Degenerationcited by 2,916
  7. 2017A Broad-Coverage Challenge Corpus for Sentence Understanding through Inferencecited by 2,682
  8. 2020DeBERTa: Decoding-enhanced BERT with Disentangled Attentioncited by 2,181
  9. 2019Language Models as Knowledge Bases?cited by 1,936
  10. 2021DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharingcited by 1,254
  11. 2020BLEURT: Learning Robust Metrics for Text Generationcited by 1,203
  12. 2019How Can We Know What Language Models Know?cited by 959
12 more
  1. 2019PAWS: Paraphrase Adversaries from Word Scramblingcited by 402
  2. 2021Measuring and Improving Consistency in Pretrained Language Modelscited by 364
  3. 2021Do Prompt-Based Models Really Understand the Meaning of their Prompts?cited by 326
  4. 2022Factuality Enhanced Language Models for Open-Ended Text Generationcited by 202
  5. 2021Multilingual LAMA: Investigating Knowledge in Multilingual Pretrained Language Modelscited by 132
  6. 2022Prompt Consistency for Zero-Shot Task Generalizationcited by 67
  7. 2021Back to Square One: Artifact Detection, Training and Commonsense Disentanglement in the Winograd Schemacited by 32
  8. 2022Quality Controlled Paraphrase Generationcited by 31
  9. 2022Novelty Controlled Paraphrase Generation with Retrieval Augmented Conditional Prompt Tuningcited by 25
  10. 2021P-Adapters: Robustly Extracting Factual Information from Language Models with Diverse Promptscited by 22
  11. 2022Factual Consistency of Multilingual Pretrained Language Modelscited by 15
  12. 2021Accurate, yet inconsistent? Consistency Analysis on Language Understanding Modelscited by 13

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 9; 2024: 8; 2025: 14; 2026: 7.

38 papers.

Sort:

By area:

  1. 2024Knowledge Conflicts for LLMs: A Surveycited by 283
  2. 2023Language Model Behavior: A Comprehensive Surveycited by 92
  3. 2024Knowledge Mechanisms in Large Language Models: A Survey and Perspectivecited by 51
  4. 2023Consistency Analysis of ChatGPTcited by 37
  5. 2023Two Failures of Self-Consistency in the Multi-Step Reasoning of LLMscited by 36
  6. 2023Predicting Question-Answering Performance of Large Language Models through Semantic Consistencycited by 28
  7. 2024Assessing Dialect Fairness and Robustness of Large Language Models in Reasoning Taskscited by 22
  8. 2025On the Robustness of Agentic Function Callingcited by 19
  9. 2023Semantic Consistency for Assuring Reliability of Large Language Modelscited by 15
  10. 2025Enhancing Semantic Consistency of Large Language Models through Model Editing: An Interpretability-Oriented Approachcited by 14
  11. 2025Consistency in Language Models: Current Landscape, Challenges, and Future Directionscited by 14
  12. 2025Improving Consistency in Large Language Models through Chain of Guidancecited by 13
26 more
  1. 2023Self-Consistency of Large Language Models under Ambiguitycited by 12
  2. 2024A Novel Metric for Measuring the Robustness of Large Language Models in Non-adversarial Scenarioscited by 8
  3. 2025LF-Steering: Latent Feature Activation Steering for Enhancing Semantic Consistency in Large Language Modelscited by 8
  4. 2024Quantifying Prediction Consistency Under Fine-Tuning Multiplicity in Tabular LLMscited by 6
  5. 2026When Simulation Lies: A Sim-to-Real Benchmark and Domain-Randomized RL Recipe for Tool-Use Agentscited by 6
  6. 2023Statistical Knowledge Assessment for Large Language Modelscited by 5
  7. 2025STED and Consistency Scoring: A Framework for Evaluating LLM Structured Output Reliabilitycited by 5
  8. 2023Large Language Models, scientific knowledge and factuality: A framework to streamline human expert evaluationcited by 4
  9. 2023LUNA: A Model-Based Universal Analysis Framework for Large Language Modelscited by 4
  10. 2026EquiMem: Calibrating Shared Memory in Multi-Agent Debate via Game-Theoretic Equilibriumcited by 4
  11. 2024SaGE: Evaluating Moral Consistency in Large Language Modelscited by 3
  12. 2025Self-Reported Confidence of Large Language Models in Gastroenterology: Analysis of Commercial, Open-Source, and Quantized Modelscited by 3
  13. 2026Consistency as a Testable Property: Statistical Methods to Evaluate AI Agent Reliabilitycited by 3
  14. 2025Improving Consistency in Retrieval-Augmented Systems with Group Similarity Rewardscited by 2
  15. 2026GSM-SEM: Benchmark and Framework for Generating Semantically Variant Augmentationscited by 2
  16. 2026OTAP: Structure-Aware Optimal Transport for Evaluating Planning and Execution in Agent Trajectoriescited by 2
  17. 2024SimCT: A Simple Consistency Test Protocol in LLMs Development Lifecyclecited by 1
  18. 2024Evaluating Consistencies in LLM responses through a Semantic Clustering of Question Answeringcited by 1
  19. 2025Factual Knowledge in Language Models: Robustness and Anomalies under Simple Temporal Context Variationscited by 1
  20. 2025On the Brittleness of LLMs: A Journey around Set Membershipcited by 1
  21. 2025Quality Assurance for LLM-RAG Systems: Empirical Insights from Tourism Application Testingcited by 0
  22. 2025Harmonizing Diverse Models: A Layer-wise Merging Strategy for Consistent Generationcited by 0
  23. 2025Is GPT-OSS All You Need? Benchmarking Large Language Models for Financial Intelligence and the Surprising Efficiency Paradoxcited by 0
  24. 2025A Proof-of-Concept for Explainable Disease Diagnosis Using Large Language Models and Answer Set Programmingcited by 0
  25. 2026Paraphrase-Induced Output-Mode Collapse: When LLMs Break Character Under Semantically Equivalent Inputscited by 0
  26. 2026Assessing Reliability of BERT-Based Models on Question Answering Taskscited by 0