mlsanityData to 14 Oct 2026

Attribution and Alignment: Effects of Local Context Repetition on Utterance Production and Comprehension in Dialogue

arXiv 2311.13061 ↗First posted 21 Nov 2023cs.CLcited by 4

Not on the map. Why?

Abstract (arXiv)

Language models are often used as the backbone of modern dialogue systems. These models are pre-trained on large amounts of written fluent language. Repetition is typically penalised when evaluating language model generations. However, it is a key component of dialogue. Humans use local and partner specific repetitions; these are preferred by human users and lead to more successful communication in dialogue. In this study, we evaluate (a) whether language models produce human-like levels of repetition in dialogue, and (b) what are the processing mechanisms related to lexical re-use they use during comprehension. We believe that such joint analysis of model production and comprehension behaviour can inform the development of cognitively inspired dialogue generation systems.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 40 of its 89 references that are in the map. First posted on arXiv 21 Nov 2023.

Builds on

40 of 89 references.

References · 40

  1. 2017A Unified Approach to Interpreting Model Predictionscited by 6,555
  2. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,512
  3. 2022OPT: Open Pre-trained Transformer Language Modelscited by 3,512
  4. 2017Axiomatic Attribution for Deep Networkscited by 3,384
  5. 2019The Curious Case of Neural Text Degenerationcited by 2,916
  6. 2023Pythia: A Suite for Analyzing Large Language Models Across Training and Scalingcited by 1,786
  7. 2017Learning Important Features Through Propagating Activation Differencescited by 1,458
  8. 2016The LAMBADA dataset: Word prediction requiring a broad discourse contextcited by 992
  9. 2019Attention is not Explanationcited by 929
  10. 2020Captum: A unified and generic model interpretability library for PyTorchcited by 547
  11. 2019Neural Text Generation with Unlikelihood Trainingcited by 503
  12. 2021MAUVE: Measuring the Gap Between Neural Text and Human Text using Divergence Frontierscited by 470
28 more
  1. 2016How NOT To Evaluate Your Dialogue System: An Empirical Study of Unsupervised Evaluation Metrics for Dialogue Response Generationcited by 356
  2. 2019Interpretable machine learning: definitions, methods, and applicationscited by 305
  3. 2016Deep Reinforcement Learning for Dialogue Generationcited by 288
  4. 2021Polyjuice: Generating Counterfactuals for Explaining, Evaluating, and Improving Modelscited by 222
  5. 2020Explaining by Removing: A Unified Framework for Model Explanationcited by 218
  6. 2019TransferTransfo: A Transfer Learning Approach for Neural Network Based Conversational Agentscited by 188
  7. 2019Neural Language Models as Psycholinguistic Subjects: Representations of Syntactic Statecited by 138
  8. 2020The elephant in the interpretability room: Why use attention as explanation when we have saliency methods?cited by 122
  9. 2022Why Does Surprisal From Larger Transformer-Based Language Models Provide a Poorer Fit to Human Reading Times?cited by 113
  10. 2022Interpreting Language Models with Contrastive Explanationscited by 92
  11. 2023Inseq: An Interpretability Toolkit for Sequence Generation Modelscited by 78
  12. 2020Speaker-Aware BERT for Multi-Turn Response Selection in Retrieval-Based Chatbotscited by 73
  13. 2021"Will You Find These Shortcuts?" A Protocol for Evaluating the Faithfulness of Input Salience Methods for Text Classificationcited by 66
  14. 2016A Simple, Fast Diverse Decoding Algorithm for Neural Generationcited by 58
  15. 2020TurnGPT: a Transformer-based Language Model for Predicting Turn-taking in Spoken Dialogcited by 56
  16. 2023What Comes Next? Evaluating Uncertainty in Neural Text Generators Against Human Production Variabilitycited by 51
  17. 2021Structural Persistence in Language Models: Priming as a Window into Abstract Language Representationscited by 44
  18. 2019Structuring Latent Spaces for Stylized Response Generationcited by 25
  19. 2022Model Criticism for Long-Form Text Generationcited by 21
  20. 2019Analysing Neural Language Models: Contextual Decomposition Reveals Default Reasoning in Number and Gender Assignmentcited by 20
  21. 2023Feature Interactions Reveal Linguistic Structure in Language Modelscited by 8
  22. 2019An End-to-End Conversational Style Matching Agentcited by 7
  23. 2021Local Explanation of Dialogue Response Generationcited by 7
  24. 2021Studying Alignment in a Collaborative Learning Activity via Automatic Methods: The Link Between What We Say and Docited by 6
  25. 2021Taming Repetition in Dialogue Generationcited by 5
  26. 2022Construction Repetition Reduces Information Rate in Dialoguecited by 5
  27. 2021Using Integrated Gradients and Constituency Parse Trees to explain Linguistic Acceptability learnt by BERTcited by 3
  28. 2020Generalizable and Explainable Dialogue Generation via Explicit Action Learningcited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 1; 2025: 0; 2026: 3.

4 papers.

  1. 2024Do Language Models Exhibit Human-like Structural Priming Effects?cited by 21
  2. 2026Surprisal Minimisation over Goal-directed Alternatives Predicts Production Choice in Dialoguecited by 1
  3. 2026Food Noise & False Safety: A Systematic Evaluation of How LLMs Fail to Adapt to Eating Disorder Queries with Clinician Feedbackcited by 1
  4. 2026Patterns of Priming in Production: Lexical, Semantic and Structural Alignment in Language Model Generationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026A Context Alignment Pre-processor for Enhancing the Coherence of Human-LLM Dialog
  2. 2025Towards Stable and Personalised Profiles for Lexical Alignment in Spoken Human-Agent Dialogue
  3. 2023Grounding Gaps in Language Model Generations
  4. 2024From Babbling to Fluency: Evaluating the Evolution of Language Models in Terms of Human Language Acquisition
  5. 2022Long-term Control for Dialogue Generation: Methods and Evaluation
  6. 2025In Dialogue with Intelligence: Rethinking Large Language Models as Collective Knowledge
  7. 2025Cat, Rat, Meow: On the Alignment of Language Model and Human Term-Similarity Judgments
  8. 2026Context-Agent: Dynamic Discourse Trees for Non-Linear Dialogue