# Implicit Bias-Like Patterns in Reasoning Models

arXiv: https://arxiv.org/abs/2503.11572 · HTML: https://mlsanity.com/paper/2503.11572/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning.
- Builds on 9 of its 62 references that are in the map.
- First posted on arXiv 14 Mar 2025.

## Placed in

- LLM reasoning / [Reasoning benchmarks and evaluation](https://mlsanity.com/lr/reasoning-benchmarks-and-evaluation/)

## Builds on (most cited first, 9 of 9)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Fine-Tuning Language Models from Human Preferences](https://mlsanity.com/paper/1909.08593/) (2019)
- [Semantics derived automatically from language corpora contain human-like biases](https://mlsanity.com/paper/1608.07187/) (2016)
- [Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned](https://mlsanity.com/paper/2209.07858/) (2022)
- [Persistent Anti-Muslim Bias in Large Language Models](https://mlsanity.com/paper/2101.05783/) (2021)
- [AIR-Bench 2024: A Safety Benchmark Based on Risk Categories from Regulations and Policies](https://mlsanity.com/paper/2407.17436/) (2024)

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

- [Does Reasoning Make Search More Fair? Comparing Fairness in Reasoning and Non-Reasoning Rerankers](https://mlsanity.com/paper/2603.10332/) (2026)
- [Evaluating Implicit Biases in LLM Reasoning through Logic Grid Puzzles](https://mlsanity.com/paper/2511.06160/) (2025)
- [AI reasoning effort predicts human decision time in content moderation](https://mlsanity.com/paper/2508.20262/) (2025)

## Abstract (arXiv)

Implicit biases refer to automatic mental processes that shape perceptions, judgments, and behaviors. Previous research on "implicit bias" in LLMs focused primarily on outputs rather than the processes underlying the outputs. We present the Reasoning Model Implicit Association Test (RM-IAT) to study implicit bias-like processing in reasoning models, LLMs that use step-by-step reasoning to solve complex tasks. Using RM-IAT, we find that reasoning models like o3-mini, DeepSeek-R1, gpt-oss-20b, and Qwen-3 8B consistently expend more reasoning tokens on association-incompatible tasks than association-compatible tasks, suggesting greater computational effort when processing counter-stereotypical information. Conversely, Claude 3.7 Sonnet exhibited reversed patterns, which thematic analysis associated with its unique internal focus on reasoning about bias and stereotypes. These findings demonstrate that reasoning models exhibit distinct implicit bias-like patterns and that these patterns vary significantly depending on the models' internal reasoning content.