Separable Pathways for Causal Reasoning: How Architectural Scaffolding Enables Hypothesis-Space Restructuring in LLM Agents

• URL: http://arxiv.org/abs/2604.20039v1
• Date: Tue, 21 Apr 2026 22:49:17 GMT
• Title: Separable Pathways for Causal Reasoning: How Architectural Scaffolding Enables Hypothesis-Space Restructuring in LLM Agents
• Abstract summary: We extend the blicket detector paradigm from developmental science to test capacity in AI agents equipped with architectural scaffolding that targets hypothesis-space restructuring.<n>Our compositional architecture has two discrete components: context graphs, which structure exploration as typed state machines, and dynamic behaviors, which monitor for evidence that the current hypothesis space is inadequate and expand it at runtime.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Causal discovery through experimentation and intervention is fundamental to robust problem solving. It requires not just updating beliefs within a fixed framework but revising the hypothesis space itself, a capacity current AI agents lack when evidence demands representations they have not previously constructed. We extend the blicket detector paradigm from developmental science to test this capacity in AI agents equipped with architectural scaffolding that targets hypothesis-space restructuring. Our compositional architecture has two discrete components: context graphs, which structure exploration as typed state machines, and dynamic behaviors, which monitor for evidence that the current hypothesis space is inadequate and expand it at runtime. Across 1,085 experimental trials, these components make orthogonal contributions: context graphs drive reasoning quality within the post-switch hypothesis space, accounting for 94\% of the accuracy gain, while dynamic behaviors drive reasoning eligibility by detecting regime changes and preventing premature commitment to outdated hypotheses.

Related papers

• Abduction Without a Body? Representational Grounding and the Abduction Loop for Scientific Hypothesis Generation [0.0]
  Recent arguments in AI and philosophy of science hold that genuine hypothesis generation requires an agent continuously coupled to the physical world.<n>We defend a narrower claim: online embodiment is not necessary for every abductive scientific act.<n>Our focus is identity abduction: the inference that two independently developed structures are one object under an explicit correspondence.
  arXiv  Detail & Related papers  (2026-08-03T17:05:31Z)
• Anatomy of a Lie: A Multi-Stage Diagnostic Framework for Tracing Hallucinations in Vision-Language Models [62.932580559941414]
  Vision-Language Models (VLMs) frequently "hallucinate" - generate plausible yet factually incorrect statements.<n>We propose a new paradigm for diagnosing hallucinations, recasting them from static output errors into dynamic pathologies of a model's computational cognition.
  arXiv  Detail & Related papers  (2026-03-16T17:20:38Z)
• AI-Assisted Engineering Should Track the Epistemic Status and Temporal Validity of Architectural Decisions [0.0]
  LLM coding assistants generate decisions faster than teams can validate them.<n>No widely-adopted framework distinguishes conjecture from verified knowledge.<n>We propose three requirements for responsible AI-assisted engineering.
  arXiv  Detail & Related papers  (2026-01-28T23:12:07Z)
• Consistency Is Not Always Correct: Towards Understanding the Role of Exploration in Post-Training Reasoning [75.79451512757844]
  Foundation models exhibit broad knowledge but limited task-specific reasoning.<n> RLVR and inference scaling motivate post-training strategies such as RLVR and inference scaling.<n>We show that RLVR induces a squeezing effect, reducing reasoning entropy and forgetting some correct paths.
  arXiv  Detail & Related papers  (2025-11-10T18:25:26Z)
• Towards Error-Centric Intelligence II: Energy-Structured Causal Models [0.0]
  We argue for a conceptual reorientation: intelligence is the ability to build and refine explanations.<n>This paper offers a formal language for causal reasoning in systems that aspire to understand, not merely to predict.
  arXiv  Detail & Related papers  (2025-10-24T22:19:17Z)
• Towards Error Centric Intelligence I, Beyond Observational Learning [0.0]
  We argue that progress toward AGI is theory limited rather than data or scale limited.<n>We begin by laying foundations, definitions of knowledge, learning, intelligence, counterfactual competence and AGI.<n>We recast the problem as three questions about how explicit and implicit errors evolve under an agent's actions.
  arXiv  Detail & Related papers  (2025-10-16T20:33:55Z)
• Active Inference AI Systems for Scientific Discovery [1.450405446885067]
  This perspective contends that progress turns on closing three mutually reinforcing gaps in abstraction, reasoning and empirical grounding.<n>Design principles are proposed for systems that reason in imaginary spaces and learn from the world.
  arXiv  Detail & Related papers  (2025-06-26T14:43:04Z)
• Controllable Logical Hypothesis Generation for Abductive Reasoning in Knowledge Graphs [54.596180382762036]
  Abductive reasoning in knowledge graphs aims to generate plausible logical hypotheses from observed entities.<n>Due to a lack of controllability, a single observation may yield numerous plausible but redundant or irrelevant hypotheses.<n>We introduce the task of controllable hypothesis generation to improve the practical utility of abductive reasoning.
  arXiv  Detail & Related papers  (2025-05-27T09:36:47Z)
• Discovering Latent Causal Variables via Mechanism Sparsity: A New Principle for Nonlinear ICA [81.4991350761909]
  Independent component analysis (ICA) refers to an ensemble of methods which formalize this goal and provide estimation procedure for practical application. We show that the latent variables can be recovered up to a permutation if one regularizes the latent mechanisms to be sparse.
  arXiv  Detail & Related papers  (2021-07-21T14:22:14Z)
• Empirically Verifying Hypotheses Using Reinforcement Learning [58.09414653169534]
  This paper formulates hypothesis verification as an RL problem. We aim to build an agent that, given a hypothesis about the dynamics of the world, can take actions to generate observations which can help predict whether the hypothesis is true or false.
  arXiv  Detail & Related papers  (2020-06-29T01:01:10Z)
• Supporting Optimal Phase Space Reconstructions Using Neural Network Architecture for Time Series Modeling [68.8204255655161]
  We propose an artificial neural network with a mechanism to implicitly learn the phase spaces properties. Our approach is either as competitive as or better than most state-of-the-art strategies.
  arXiv  Detail & Related papers  (2020-06-19T21:04:47Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.