mlsanityData to 14 Oct 2026

FRIT: Using Causal Importance to Improve Chain-of-Thought Faithfulness

arXiv 2509.13334 ↗First posted 10 Sep 2025cs.AI cs.LGcited by 9

On the map: LLM reasoning → Prompting and chain-of-thought reasoning

Abstract (arXiv)

Chain-of-thought (CoT) reasoning improves LLM accuracy on complex tasks, but CoTs are often unfaithful; that is, the model does not rely on the stated reasoning steps to produce its answer. Existing methods to improve faithfulness require human annotations. We propose LIFT, which automatically scores each reasoning step by its causal influence: we replace it with a domain-matched but irrelevant fact and measure the drop in answer log-probability. We fine-tune on high-scoring CoT traces where interventions successfully disrupted reasoning, jointly optimizing for faithfulness and accuracy. LIFT improves faithfulness across three models (Mistral-7B-Instruct, Qwen3-8B, and Qwen3-4B-Instruct) and three benchmarks, with accuracy comparable to a model fine-tuned solely for accuracy.

Key facts. Cited by 9 papers in the map (data through 14 Oct 2026). Placed in Prompting and chain-of-thought reasoning, one of 17 directions in LLM reasoning. Builds on 11 of its 13 references that are in the map. First posted on arXiv 10 Sep 2025.

Builds on

11 of 13 references.

Closest ancestors · 3

Same direction, newest first

  1. 2024Making Reasoning Matter: Measuring and Improving Faithfulness of Chain-of-Thought Reasoning21 Feb
  2. 2023Measuring Faithfulness in Chain-of-Thought Reasoning17 Jul
  3. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Models28 Jan

Other references · 8

  1. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  2. 2018CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledgecited by 1,822
  3. 2021Are NLP Models really able to Solve Simple Math Word Problems?cited by 878
  4. 2021Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategiescited by 743
  5. 2021A Diverse Corpus for Evaluating and Developing English Math Word Problem Solverscited by 318
  6. 2023Causal Abstraction: A Theoretical Foundation for Mechanistic Interpretabilitycited by 205
  7. 2023Towards Trustworthy and Aligned Machine Learning: A Data-centric Survey with Causality Perspectivescited by 30
  8. 2025Beyond Black-Box Obfuscation: Mechanistic Analysis and Defense of White-Box Monitorscited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 8.

9 papers.

  1. 2026Analyzing and Improving Chain-of-Thought Monitorability Through Information Theorycited by 10
  2. 2026Broken Chains: The Cost of Incomplete Reasoning in LLMscited by 4
  3. 2026SPD-Faith Bench: Diagnosing and Improving Faithfulness in Chain-of-Thought for Multimodal Large Language Modelscited by 2
  4. 2025Optimizing Chain-of-Thought Confidence via Topological and Dirichlet Risk Analysiscited by 1
  5. 2026Measuring AI Reasoning: A Guide for Researcherscited by 1
  6. 2026Why Limit the Residual Stream to Layers and Not Tokens? Persistent Memory for Continuous Latent Reasoningcited by 1
  7. 2026Do Vision-Language-Action Models Mean What They Say? On the Role of Faithfulness in Embodied Reasoningcited by 1
  8. 2026Forecasting Future Behavior as a Learning Taskcited by 0
  9. 2026Position: Evaluations of AI Moral Reasoning Still Miss Half of the Picturecited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026C2-Faith: Benchmarking LLM Judges for Causal and Coverage Faithfulness in Chain-of-Thought Reasoning
  2. 2024On the Hardness of Faithful Chain-of-Thought Reasoning in Large Language Models
  3. 2026Mechanistic Evidence for Faithfulness Decay in Chain-of-Thought Reasoning
  4. 2026CASE: Causal Alignment and Structural Enforcement for Improving Chain-of-Thought Faithfulness
  5. 2026RFEval: Benchmarking Reasoning Faithfulness under Counterfactual Reasoning Intervention in Large Reasoning Models
  6. 2026Faithfulness as Information Flow: Evaluating and Training Faithful Chain-of-Thought Reasoning
  7. 2026FACT-E: Causality-Inspired Evaluation for Trustworthy Chain-of-Thought Reasoning
  8. 2025Causal Sufficiency and Necessity Improves Chain-of-Thought Reasoning