# Chain of Thought Still Thinks Fast: APriCoT Helps with Thinking Slow

arXiv: https://arxiv.org/abs/2408.08651 · HTML: https://mlsanity.com/paper/2408.08651/

## Key facts

- Cited by 8 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 15 of its 22 references that are in the map.
- First posted on arXiv 16 Aug 2024.

## 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, 15 of 15)

- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them](https://mlsanity.com/paper/2210.09261/) (2022)
- [Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting](https://mlsanity.com/paper/2305.04388/) (2023)
- [Language Models Are Greedy Reasoners: A Systematic Formal Analysis of Chain-of-Thought](https://mlsanity.com/paper/2210.01240/) (2022)
- [Large Language Models Fail on Trivial Alterations to Theory-of-Mind Tasks](https://mlsanity.com/paper/2302.08399/) (2023)
- [On Second Thought, Let's Not Think Step by Step! Bias and Toxicity in Zero-Shot Reasoning](https://mlsanity.com/paper/2212.08061/) (2022)
- [Why think step by step? Reasoning emerges from the locality of experience](https://mlsanity.com/paper/2304.03843/) (2023)
- [Bias-Augmented Consistency Training Reduces Biased Reasoning in Chain-of-Thought](https://mlsanity.com/paper/2403.05518/) (2024)
- [Do language models learn typicality judgments from text?](https://mlsanity.com/paper/2105.02987/) (2021)
- [Unveiling Selection Biases: Exploring Order and Token Sensitivity in Large Language Models](https://mlsanity.com/paper/2406.03009/) (2024)
- [Eliciting the Priors of Large Language Models using Iterated In-Context Learning](https://mlsanity.com/paper/2406.01860/) (2024)
- [Using Artificial Populations to Study Psychological Phenomena in Neural Models](https://mlsanity.com/paper/2308.08032/) (2023)
- [Large Language Model Recall Uncertainty is Modulated by the Fan Effect](https://mlsanity.com/paper/2407.06349/) (2024)
- [The Base-Rate Effect on LLM Benchmark Performance: Disambiguating Test-Taking Strategies from Benchmark Performance](https://mlsanity.com/paper/2406.11634/) (2024)

## Built on by (most cited first, 8 of 8)

- [Reasoning Model is Stubborn: Diagnosing Instruction Overriding in Reasoning Models](https://mlsanity.com/paper/2505.17225/) (2025)
- [ARB: A Comprehensive Arabic Multimodal Reasoning Benchmark](https://mlsanity.com/paper/2505.17021/) (2025)
- [Semantic Calibration Prevails Where Token Confidence Fails: Benchmarking Long-Form Scientific QA](https://mlsanity.com/paper/2602.00279/) (2026)
- [LLMs as Agentic Cooperative Players in Multiplayer UNO](https://mlsanity.com/paper/2509.09867/) (2025)
- [ImplicitBBQ: Benchmarking Implicit Bias in Large Language Models through Characteristic Based Cues](https://mlsanity.com/paper/2604.01925/) (2026)
- [Human-Alignment, Calibration, and Activation Patterns in Large Language Model Uncertainty](https://mlsanity.com/paper/2605.30675/) (2026)
- [Adaptive Triggering for Bias Correction in LLM Reasoning](https://mlsanity.com/paper/2608.25379/) (2026)
- [Does Thinking Help Fairness? Reasoning Tokens Resolve Some Biases but Create More](https://mlsanity.com/paper/2609.30768/) (2026)

## Abstract (arXiv)

Language models are known to absorb biases from their training data, leading to predictions driven by statistical regularities rather than semantic relevance. We investigate the impact of these biases on answer choice preferences in the Massive Multi-Task Language Understanding (MMLU) task. Our findings show that these biases are predictive of model preference and mirror human test-taking strategies even when chain of thought (CoT) reasoning is used. To address this issue, we introduce Counterfactual Prompting with Agnostically Primed CoT (APriCoT). We demonstrate that while Counterfactual Prompting with CoT alone is insufficient to mitigate bias, APriCoT effectively reduces the influence of base-rate probabilities while improving overall accuracy. Our results suggest that mitigating bias requires a slow thinking process which CoT alone may not provide as it tends to reinforce fast thinking model bias under some prompting methodologies. APriCoT is a step toward developing more robust and fair language models that can think slow.