mlsanityData to 14 Oct 2026

Do LLM Debates Repeat Arguments Differently Across Languages?

arXiv 2607.23442 ↗First posted 26 Jul 2026cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

LLM debate is usually evaluated by final answers, yet transcripts reveal whether later turns develop new arguments or return to earlier claims in new wording. We study this process with textitprior-argument similarity, which compares extracted argument units with earlier units in the same debate. In controlled eight-turn debates over 71 motions, six languages, and four model agents, Chinese is the only tested language with a consistently positive gap relative to English across three multilingual embedding models. The gap persists across agents, turn positions, regression adjustment, metric variants, extraction-length controls, a second-extractor subset, and cross-encoder tail rescoring. Manual calibration shows weak item-level alignment but a high-similarity tail enriched for substantive repetition. A diversity-aware prompt lowers textitprior-argument similarity across languages, yet does not significantly narrow the Chinese--English gap. Multilingual debate evaluation should therefore measure argumentative development over time and report both average and gap terms.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 35 of its 62 references that are in the map. First posted on arXiv 26 Jul 2026.

Builds on

35 of 62 references.

References · 35

  1. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
  2. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  3. 2019The Curious Case of Neural Text Degenerationcited by 2,916
  4. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,784
  5. 2015A Diversity-Promoting Objective Function for Neural Conversation Modelscited by 1,077
  6. 2024M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillationcited by 1,073
  7. 2023Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debatecited by 1,011
  8. 2020Language-agnostic BERT Sentence Embeddingcited by 627
  9. 2022Language Models are Multilingual Chain-of-Thought Reasonerscited by 530
  10. 2019Neural Text Generation with Unlikelihood Trainingcited by 503
  11. 2018AI safety via debatecited by 400
  12. 2024Multilingual E5 Text Embeddings: A Technical Reportcited by 372
23 more
  1. 2016Diverse Beam Search: Decoding Diverse Solutions from Neural Sequence Modelscited by 301
  2. 2023Exploring Collaboration Mechanisms for LLM Agents: A Social Psychology Viewcited by 222
  3. 2023Not All Languages Are Created Equal in LLMs: Improving Multilingual Capability by Cross-Lingual-Thought Promptingcited by 165
  4. 2023Having Beer after Prayer? Measuring Cultural Bias in Large Language Modelscited by 153
  5. 2022Probing Pre-Trained Language Models for Cross-Cultural Differences in Valuescited by 149
  6. 2022Learning to Break the Loop: Analyzing and Mitigating Repetitions for Neural Text Generationcited by 94
  7. 2023Exchange-of-Thought: Enhancing Large Language Model Capabilities through Cross-Model Communicationcited by 82
  8. 2024Can multiple-choice questions really be useful in detecting the abilities of LLMs?cited by 69
  9. 2025Talk Isn't Always Cheap: Understanding Failure Modes in Multi-Agent Debatecited by 66
  10. 2025Can LLM Agents Really Debate? A Controlled Study of Multi-Agent Debate in Logical Reasoningcited by 29
  11. 2025Break the Checkbox: Challenging Closed-Style Evaluations of Cultural Alignment in LLMscited by 26
  12. 2017Measuring the Similarity of Sentential Arguments in Dialogcited by 23
  13. 2025Stay Focused: Problem Drift in Multi-Agent Debatecited by 23
  14. 2024Debatrix: Multi-dimensional Debate Judge with Iterative Chronological Analysis Based on LLMcited by 21
  15. 2025Multiple LLM Agents Debate for Equitable Cultural Alignmentcited by 21
  16. 2025Could Thinking Multilingually Empower LLM Reasoning?cited by 16
  17. 2024Debate, Deliberate, Decide (D3): A Cost-Aware Adversarial Framework for Reliable and Interpretable LLM Evaluationcited by 15
  18. 2025Do Repetitions Matter? Strengthening Reliability in LLM Evaluationscited by 14
  19. 2025DEBATE: A Large-Scale Benchmark for Evaluating Opinion Dynamics in Role-Playing LLM Agentscited by 6
  20. 2025Debatable Intelligence: Benchmarking LLM Judges via Debate Speech Evaluationcited by 3
  21. 2025Adaptive Repetition for Mitigating Position Bias in LLM-Based Rankingcited by 1
  22. 2025Solving LLM Repetition Problem in Production: A Comprehensive Study of Multiple Solutionscited by 1
  23. 2026Evaluating LLM-Driven Summarisation of Parliamentary Debates with Computational Argumentationcited by 1

Built on by

Not cited yet.