Breaking the Chain: A Causal Analysis of LLM Faithfulness to Intermediate Structures

• URL: http://arxiv.org/abs/2603.16475v1
• Date: Tue, 17 Mar 2026 13:01:44 GMT
• Title: Breaking the Chain: A Causal Analysis of LLM Faithfulness to Intermediate Structures
• Abstract summary: We introduce a causal evaluation protocol that makes explicit intermediate structures causally determine the output.<n>Across eight models and three benchmarks, models appear self-consistent with their own intermediate structures but fail to update predictions after intervention in up to 60% of cases.<n>Overall, intermediate structures in schema-guided pipelines function as influential context rather than stable causal mediators.
• Score: 58.54426802984356
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Schema-guided reasoning pipelines ask LLMs to produce explicit intermediate structures -- rubrics, checklists, verification queries -- before committing to a final decision. But do these structures causally determine the output, or merely accompany it? We introduce a causal evaluation protocol that makes this directly measurable: by selecting tasks where a deterministic function maps intermediate structures to decisions, every controlled edit implies a unique correct output. Across eight models and three benchmarks, models appear self-consistent with their own intermediate structures but fail to update predictions after intervention in up to 60% of cases -- revealing that apparent faithfulness is fragile once the intermediate structure changes. When derivation of the final decision from the structure is delegated to an external tool, this fragility largely disappears; however, prompts which ask to prioritize the intermediate structure over the original input do not materially close the gap. Overall, intermediate structures in schema-guided pipelines function as influential context rather than stable causal mediators.

Related papers

• ReasonOps: Operator Segmentation for LLM Reasoning Traces [29.17171274711201]
  Chain-of-thought traces from large reasoning models can span tens of thousands of tokens.<n>We develop ReasonOps, an unsupervised, expressive method for annotating chain-of-thought traces.<n>We analyze 44,662 traces from 12 thinking LLMs spanning 6 families across 8 reasoning benchmarks.
  arXiv  Detail & Related papers  (2026-05-28T00:08:35Z)
• I-SAFE: Wasserstein Coherence Metrics for Structural Auditing of Scientific AI Models [5.562984399879219]
  We introduce the textscI-SAFE (Interventional Secure, Accurate, Fair and Explainable) framework for scientific AI models.<n>textscI-SAFE evaluates raw model outputs under structurally guided perturbations of the input.<n>We instantiate textscI-SAFE on drug--target interaction (DTI) prediction using the Davis kinase benchmark.
  arXiv  Detail & Related papers  (2026-05-20T20:44:30Z)
• Deep Reasoning in General Purpose Agents via Structured Meta-Cognition [58.185853639335896]
  We introduce an inference-time approach for constructing task-specific scaffolds through structured meta-reasoning.<n>We instantiate this approach in a general-purpose agent that distributes complex tasks across more controlled reasoning threads.
  arXiv  Detail & Related papers  (2026-05-12T01:21:37Z)
• Dynamic analysis enhances issue resolution [53.50448142467294]
  DAIRA (Dynamic Analysis-enhanced Issue Resolution Agent) is an automated repair framework that embeds dynamic analysis into the agent's reasoning cycle.<n> Driven by a Test Tracing-Driven methodology, DAIRA utilizes lightweight monitors to extract critical runtime data.<n>When powered by Gemini 3 Flash Preview, DAIRA establishes a new state-of-the-art (SOTA) performance, achieving a 79.4% resolution rate on the SWE-bench Verified dataset.
  arXiv  Detail & Related papers  (2026-03-23T14:48:54Z)
• Causality is Key for Interpretability Claims to Generalise [35.833847356014154]
  Interpretability research on large language models (LLMs) has yielded important insights into model behaviour.<n> recurring pitfalls persist: findings that do not generalise, and causal interpretations that outrun the evidence.<n>Pearl's causal hierarchy clarifies what an interpretability study can justify.
  arXiv  Detail & Related papers  (2026-02-18T18:45:04Z)
• Dynamics Within Latent Chain-of-Thought: An Empirical Study of Causal Structure [58.89643769707751]
  We study latent chain-of-thought as a manipulable causal process in representation space.<n>We find that latent-step budgets behave less like homogeneous extra depth and more like staged functionality with non-local routing.<n>These results motivate mode-conditional and stability-aware analyses as more reliable tools for interpreting and improving latent reasoning systems.
  arXiv  Detail & Related papers  (2026-02-09T15:25:12Z)
• Emergent Structured Representations Support Flexible In-Context Inference in Large Language Models [77.98801218316505]
  Large language models (LLMs) exhibit emergent behaviors suggestive of human-like reasoning.<n>We investigate the internal processing of LLMs during in-context concept inference.
  arXiv  Detail & Related papers  (2026-02-08T03:14:39Z)
• Multi-Agent Procedural Graph Extraction with Structural and Logical Refinement [66.51979814832332]
  model formulates procedural graph extraction as a multi-round reasoning process with dedicated structural and logical refinement.<n>Experiments demonstrate that model achieves substantial improvements in both structural correctness and logical consistency over strong baselines.
  arXiv  Detail & Related papers  (2026-01-27T04:00:48Z)
• Think Locally, Explain Globally: Graph-Guided LLM Investigations via Local Reasoning and Belief Propagation [5.191980417814362]
  LLM agents excel when environments are mostly static and the needed information fits in a model's context window.<n>ReAct-style agents are especially brittle in this regime.<n>We propose EoG, a framework in which an LLM performs bounded local evidence mining and labeling (cause vs symptom) while a deterministic controller manages, state, and belief propagation to compute a minimal explanatory frontier.
  arXiv  Detail & Related papers  (2026-01-25T17:27:19Z)
• Structured Reasoning for Large Language Models [59.215789462977206]
  We propose Structured Reasoning (SCR), a framework that decouples reasoning trajectories into explicit, evaluable, and trainable components.<n>SCR substantially improves reasoning efficiency and self-verification.<n>Compared with existing reasoning paradigms, it reduces output token length by up to 50%.
  arXiv  Detail & Related papers  (2026-01-12T04:04:01Z)
• Implicit Reasoning in Large Language Models: A Comprehensive Survey [67.53966514728383]
  Large Language Models (LLMs) have demonstrated strong generalization across a wide range of tasks.<n>Recent studies have shifted attention from explicit chain-of-thought prompting toward implicit reasoning.<n>This survey introduces a taxonomy centered on execution paradigms, shifting the focus from representational forms to computational strategies.
  arXiv  Detail & Related papers  (2025-09-02T14:16:02Z)
• Unsupervised Cross-Lingual Transfer of Structured Predictors without Source Data [37.1075911292287]
  We show that the means of aggregating over the input models is critical, and that multiplying marginal probabilities of substructures to obtain high-probability structures for distant supervision is substantially better than taking the union over the input models. Testing on 18 languages, we demonstrate that the method works in a cross-lingual setting, considering both dependency parsing and part-of-speech structured prediction problems. Our analyses show that the proposed method produces less noisy labels for the distant supervision.
  arXiv  Detail & Related papers  (2021-10-08T02:46:34Z)

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.