# Critical or Compliant? The Double-Edged Sword of Reasoning in Chain-of-Thought Explanations

arXiv: https://arxiv.org/abs/2511.12001 · HTML: https://mlsanity.com/paper/2511.12001/

## Key facts

- Cited by 3 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 16 of its 58 references that are in the map.
- First posted on arXiv 15 Nov 2025.

## Placed in

- LLM reasoning / [Prompting and chain-of-thought reasoning](https://mlsanity.com/lr/prompting-and-chain-of-thought-reasoning/)

## Builds on (most cited first, 16 of 16)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [OpenAI o1 System Card](https://mlsanity.com/paper/2412.16720/) (2024)
- [Measuring Faithfulness in Chain-of-Thought Reasoning](https://mlsanity.com/paper/2307.13702/) (2023)
- [Reasoning Models Don't Always Say What They Think](https://mlsanity.com/paper/2505.05410/) (2025)
- [Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Vision-Language Models](https://mlsanity.com/paper/2409.17146/) (2024)
- [Chain of Thought Monitorability: A New and Fragile Opportunity for AI Safety](https://mlsanity.com/paper/2507.11473/) (2025)
- [Faithfulness vs. Plausibility: On the (Un)Reliability of Explanations from Large Language Models](https://mlsanity.com/paper/2402.04614/) (2024)
- [A Comprehensive Survey on Trustworthiness in Reasoning with Large Language Models](https://mlsanity.com/paper/2509.03871/) (2025)
- [Measuring and mitigating overreliance to build human-compatible AI](https://mlsanity.com/paper/2509.08010/) (2025)
- [Why Would You Suggest That? Human Trust in Language Model Responses](https://mlsanity.com/paper/2406.02018/) (2024)
- [HalLoc: Token-level Localization of Hallucinations for Vision Language Models](https://mlsanity.com/paper/2506.10286/) (2025)
- [Thoughts without Thinking: Reconsidering the Explanatory Value of Chain-of-Thought Reasoning in LLMs through Agentic Pipelines](https://mlsanity.com/paper/2505.00875/) (2025)
- [Bias in the Loop: How Humans Evaluate AI-Generated Suggestions](https://mlsanity.com/paper/2509.08514/) (2025)
- [Evaluating Explainability: A Framework for Systematic Assessment and Reporting of Explainable AI Features](https://mlsanity.com/paper/2506.13917/) (2025)

## Built on by (most cited first, 3 of 3)

- [To See or To Please: Uncovering Visual Sycophancy and Split Beliefs in VLMs](https://mlsanity.com/paper/2603.18373/) (2026)
- [Reflections and New Directions for Human-Centered Large Language Models](https://mlsanity.com/paper/2605.06901/) (2026)
- [From Statute to Control Flow: Span-Grounded Deontic Trees for Defeasible Scope Parsing](https://mlsanity.com/paper/2606.08932/) (2026)

## Abstract (arXiv)

Explanations are often promoted as tools for transparency, but they can also foster confirmation bias; users may assume reasoning is correct whenever outputs appear acceptable. We study this double-edged role of Chain-of-Thought (CoT) explanations in multimodal moral scenarios by systematically perturbing reasoning chains and manipulating delivery tones. Specifically, we analyze reasoning errors in vision language models (VLMs) and how they impact user trust and the ability to detect errors. Our findings reveal two key effects: (1) users often equate trust with outcome agreement, sustaining reliance even when reasoning is flawed, and (2) the confident tone suppresses error detection while maintaining reliance, showing that delivery styles can override correctness. These results highlight how CoT explanations can simultaneously clarify and mislead, underscoring the need for NLP systems to provide explanations that encourage scrutiny and critical thinking rather than blind trust. All code will be released publicly.