mlsanityData to 14 Oct 2026

Explainability Through Systematicity: The Hard Systematicity Challenge for Artificial Intelligence

arXiv 2507.22197 ↗First posted 29 Jul 2025cs.AI cs.CL cs.CYcited by 0

Not on the map. Why?

Abstract (arXiv)

This paper argues that explainability is only one facet of a broader ideal that shapes our expectations towards artificial intelligence (AI). Fundamentally, the issue is to what extent AI exhibits systematicity--not merely in being sensitive to how thoughts are composed of recombinable constituents, but in striving towards an integrated body of thought that is consistent, coherent, comprehensive, and parsimoniously principled. This richer conception of systematicity has been obscured by the long shadow of the "systematicity challenge" to connectionism, according to which network architectures are fundamentally at odds with what Fodor and colleagues termed "the systematicity of thought." I offer a conceptual framework for thinking about "the systematicity of thought" that distinguishes four senses of the phrase. I use these distinctions to defuse the perceived tension between systematicity and connectionism and show that the conception of systematicity that historically shaped our sense of what makes thought rational, authoritative, and scientific is more demanding than the Fodorian notion. To determine whether we have reason to hold AI models to this ideal of systematicity, I then argue, we must look to the rationales for systematization and explore to what extent they transfer to AI models. I identify five such rationales and apply them to AI. This brings into view the "hard systematicity challenge." However, the demand for systematization itself needs to be regulated by the rationales for systematization. This yields a dynamic understanding of the need to systematize thought, which tells us how systematic we need AI models to be and when.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 40 of its 195 references that are in the map. First posted on arXiv 29 Jul 2025.

Builds on

40 of 195 references.

References · 40

  1. 2017Attention Is All You Needcited by 60,361
  2. 2017Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)cited by 1,319
  3. 2022Measuring and Narrowing the Compositionality Gap in Language Modelscited by 1,075
  4. 2024Evaluating Very Long-Term Conversational Memory of LLM Agentscited by 830
  5. 2024Training Large Language Models to Reason in a Continuous Latent Spacecited by 797
  6. 2018Explaining Explanations: An Overview of Interpretability of Machine Learningcited by 508
  7. 2022Multimodal Learning with Transformers: A Surveycited by 400
  8. 2021Measuring and Improving Consistency in Pretrained Language Modelscited by 364
  9. 2024Long-context LLMs Struggle with Long In-context Learningcited by 286
  10. 2019Compositionality decomposed: how do neural networks generalise?cited by 265
  11. 2023Universal Self-Consistency for Large Language Model Generationcited by 203
  12. 2022Talking About Large Language Modelscited by 175
28 more
  1. 2023Still No Lie Detector for Language Models: Probing Empirical and Conceptual Roadblockscited by 78
  2. 2022Meaning without reference in large language modelscited by 76
  3. 2024Multimodal Alignment and Fusion: A Surveycited by 74
  4. 2023PLANNER: Generating Diversified Paragraph via Latent Language Diffusion Modelcited by 58
  5. 2024Internal Consistency and Self-Feedback in Large Language Models: A Surveycited by 50
  6. 2022Neurocompositional computing: From the Central Paradox of Cognition to a new generation of AI systemscited by 44
  7. 2020Assessing Phrasal Representation and Composition in Transformerscited by 41
  8. 2023Robot Learning in the Era of Foundation Models: A Surveycited by 38
  9. 2023Consistency Analysis of ChatGPTcited by 37
  10. 2024Improving the Robustness of Large Language Models via Consistency Alignmentcited by 36
  11. 2024Standards for Belief Representations in LLMscited by 30
  12. 2018Assessing Composition in Sentence Vector Representationscited by 26
  13. 2024A Philosophical Introduction to Language Models - Part II: The Way Forwardcited by 18
  14. 2023Chatbots as social companions: How people perceive consciousness, human likeness, and social health benefits in machinescited by 17
  15. 2024Does ChatGPT Have a Mind?cited by 14
  16. 2025Episodic Memories Generation and Evaluation Benchmark for Large Language Modelscited by 14
  17. 2025Propositional Interpretability in Artificial Intelligencecited by 12
  18. 2024Enhancing Logical Reasoning in Large Language Models through Graph-based Synthetic Datacited by 8
  19. 2022ML Interpretability: Simple Isn't Easycited by 7
  20. 2022A Theoretical Framework for AI Models Explainability with Application in Biomedicinecited by 7
  21. 2024A Comprehensive Guide to Explainable AI: From Classical Models to LLMscited by 7
  22. 2022Striking a Balance: Alleviating Inconsistency in Pre-trained Models for Symmetric Classification Taskscited by 6
  23. 2022A Means-End Account of Explainable Artificial Intelligencecited by 5
  24. 2020The Need for Standardized Explainabilitycited by 3
  25. 2024Still "Talking About Large Language Models": Some Clarificationscited by 3
  26. 2022On Explainability in AI-Solutions: A Cross-Domain Surveycited by 2
  27. 2025Can AI Rely on the Systematicity of Truth? The Challenge of Modelling Normative Domainscited by 2
  28. 2023Sources of Opacity in Computer Systems: Towards a Comprehensive Taxonomycited by 1

Built on by

Not cited yet.