mlsanityData to 7 Oct 2026

Cross-LLM Consistency in Inference: Evidence from Shared Interactions

arXiv 2606.08129 ↗First posted 6 Jun 2026cs.AIcited by 1

Not on the map. Why?

Abstract (arXiv)

Large language models (LLMs) differ in architecture, training data, and optimization procedures, yet they may still develop similar internal inference patterns. In this paper, we examine this hypothesis using interaction-based explanations. We find that LLMs often share interaction patterns when predicting the same target token from the same prompt. This consistency is more pronounced among advanced LLMs. Shared interactions also tend to be lower-order and show weaker positive-negative cancellation than non-shared interactions. These results suggest that advanced LLMs may be implicitly optimized toward common inference patterns, even though the mechanisms that give rise to such cross-model consistency remain open.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 32 of its 45 references that are in the map. First posted on arXiv 6 Jun 2026.

Builds on

32 of 45 references.

References · 32

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2022Training language models to follow instructions with human feedbackcited by 15,712
  3. 2024The Llama 3 Herd of Modelscited by 14,192
  4. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  5. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,266
  6. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  7. 2025Qwen3 Technical Reportcited by 8,195
  8. 2023Qwen Technical Reportcited by 3,273
  9. 2016Pointer Sentinel Mixture Modelscited by 2,992
  10. 2025Gemma 3 Technical Reportcited by 1,686
  11. 2024Qwen2.5-Coder Technical Reportcited by 1,151
  12. 2024ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Toolscited by 1,096
20 more
  1. 2024Phi-4 Technical Reportcited by 675
  2. 2023The Falcon Series of Open Language Modelscited by 526
  3. 2013Good Debt or Bad Debt: Detecting Semantic Orientations in Economic Textscited by 310
  4. 2022ConvFinQA: Exploring the Chain of Numerical Reasoning in Conversational Finance Question Answeringcited by 194
  5. 2022Talking About Large Language Modelscited by 174
  6. 2020A Unified Approach to Interpreting and Boosting Adversarial Transferabilitycited by 71
  7. 2021Discovering and Explaining the Representation Bottleneck of DNNscited by 42
  8. 2021Defining and Quantifying the Emergence of Sparse Concepts in DNNscited by 36
  9. 2025General Scales Unlock AI Evaluation with Explanatory and Predictive Powercited by 30
  10. 2023Does a Neural Network Really Encode Symbolic Concepts?cited by 27
  11. 2023Explaining Generalization Power of a DNN Using Interactive Conceptscited by 25
  12. 2023SEntFiN 1.0: Entity-Aware Sentiment Analysis for Financial Newscited by 18
  13. 2023FiNER-ORD: Financial Named Entity Recognition Open Research Datasetcited by 17
  14. 2025ProxySPEX: Inference-Efficient Interpretability via Sparse Feature Interactions in LLMscited by 16
  15. 2024Defining and Extracting generalizable interaction primitives from DNNscited by 14
  16. 2024Towards the Dynamics of a DNN Learning Symbolic Interactionscited by 13
  17. 2024Learning to Understand: Identifying Interactions via the M"obius Transformcited by 12
  18. 2025SPEX: Scaling Feature Interaction Explanations for LLMscited by 12
  19. 2023Bayesian Neural Networks Avoid Encoding Complex and Perturbation-Sensitive Conceptscited by 9
  20. 2025Technical Report: Quantifying and Analyzing the Generalization Power of a DNNcited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026One Attack to Fool Them All: Highly Transferable Black-Box Adversarial Attacks on Frontier MLLMscited by 0