# Self-Reflection Makes Large Language Models Safer, Less Biased, and Ideologically Neutral

arXiv: https://arxiv.org/abs/2406.10400 · HTML: https://mlsanity.com/paper/2406.10400/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning.
- Builds on 13 of its 19 references that are in the map.
- First posted on arXiv 14 Jun 2024.

## Placed in

- LLM reasoning / [Test-time compute and efficient thinking](https://mlsanity.com/lr/test-time-compute-and-efficient-thinking/)

## Builds on (most cited first, 13 of 13)

- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Self-Refine: Iterative Refinement with Self-Feedback](https://mlsanity.com/paper/2303.17651/) (2023)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Exams](https://mlsanity.com/paper/2009.13081/) (2020)
- [Large Language Models Cannot Self-Correct Reasoning Yet](https://mlsanity.com/paper/2310.01798/) (2023)
- [Teaching Large Language Models to Self-Debug](https://mlsanity.com/paper/2304.05128/) (2023)
- [CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquing](https://mlsanity.com/paper/2305.11738/) (2023)
- [REFINER: Reasoning Feedback on Intermediate Representations](https://mlsanity.com/paper/2304.01904/) (2023)
- [Can Large Language Models Really Improve by Self-critiquing Their Own Plans?](https://mlsanity.com/paper/2310.08118/) (2023)
- [GPT-4 Doesn't Know It's Wrong: An Analysis of Iterative Prompting for Reasoning Problems](https://mlsanity.com/paper/2310.12397/) (2023)
- [Self-Contrast: Better Reflection Through Inconsistent Solving Perspectives](https://mlsanity.com/paper/2401.02009/) (2024)
- [Semantic Sensitivities and Inconsistent Predictions: Measuring the Fragility of NLI Models](https://mlsanity.com/paper/2401.14440/) (2024)

## Built on by (most cited first, 20 of 20)

- [Survey on Evaluation of LLM-based Agents](https://mlsanity.com/paper/2503.16416/) (2025)
- [No Free Lunch: Rethinking Internal Feedback for LLM Reasoning](https://mlsanity.com/paper/2506.17219/) (2025)
- [Towards Concise and Adaptive Thinking in Large Reasoning Models: A Survey](https://mlsanity.com/paper/2507.09662/) (2025)
- [PRefLexOR: Preference-based Recursive Language Modeling for Exploratory Optimization of Reasoning and Agentic Thinking](https://mlsanity.com/paper/2410.12375/) (2024)
- [Think Twice Before You Act: Enhancing Agent Behavioral Safety with Thought Correction](https://mlsanity.com/paper/2505.11063/) (2025)
- [Evolutionary Perspectives on the Evaluation of LLM-Based AI Agents: A Comprehensive Survey](https://mlsanity.com/paper/2506.11102/) (2025)
- [Failure Makes the Agent Stronger: Enhancing Accuracy through Structured Reflection for Reliable Tool Interactions](https://mlsanity.com/paper/2509.18847/) (2025)
- [UALM: Unified Audio Language Model for Understanding, Generation and Reasoning](https://mlsanity.com/paper/2510.12000/) (2025)
- [LM Agents May Fail to Act on Their Own Risk Knowledge](https://mlsanity.com/paper/2508.13465/) (2025)
- [Preference Redirection via Attention Concentration: An Attack on Computer Use Agents](https://mlsanity.com/paper/2604.08005/) (2026)
- [QRAFTI: An Agentic Framework for Empirical Research in Quantitative Finance](https://mlsanity.com/paper/2604.18500/) (2026)
- [Cognitive-Aligned Document Selection for Retrieval-augmented Generation](https://mlsanity.com/paper/2502.11770/) (2025)
- [Advancing Multi-Step Mathematical Reasoning in Large Language Models through Multi-Layered Self-Reflection with Auto-Prompting](https://mlsanity.com/paper/2506.23888/) (2025)
- [A comprehensive study of LLM-based argument classification: from LLAMA through GPT-4o to Deepseek-R1](https://mlsanity.com/paper/2507.08621/) (2025)
- [Library Hallucinations in LLM-Generated Code: A Risk Analysis Grounded in Developer Queries](https://mlsanity.com/paper/2509.22202/) (2025)
- [Are the LLMs Capable of Maintaining at Least the Language Genus?](https://mlsanity.com/paper/2510.21561/) (2025)
- [Enhancing reasoning accuracy in large language models during inference time](https://mlsanity.com/paper/2603.21301/) (2026)
- [ReflectRM: Boosting Generative Reward Models via Self-Reflection within a Unified Judgment Framework](https://mlsanity.com/paper/2604.07506/) (2026)
- [Self-reflection in Automated Qualitative Coding: Improving Text Annotation through Secondary LLM Critique](https://mlsanity.com/paper/2601.09905/) (2026)
- [Structured Visual Narratives Undermine Safety Alignment in Multimodal Large Language Models](https://mlsanity.com/paper/2603.21697/) (2026)

## Abstract (arXiv)

Previous studies proposed that the reasoning capabilities of large language models (LLMs) can be improved through self-reflection, i.e., letting LLMs reflect on their own output to identify and correct mistakes in the initial responses. However, earlier experiments offer mixed results when it comes to the benefits of self-reflection. Furthermore, prior studies on self-reflection are predominantly concerned with the reasoning capabilities of models, ignoring the potential for self-reflection in safety, bias, and ideological leaning. Here, by conducting a series of experiments testing LLM's self-reflection capability in various tasks using a variety of prompts and different LLMs, we make several contributions to the literature. First, we reconcile conflicting findings regarding the benefit of self-reflection, by demonstrating that the outcome of self-reflection is sensitive to prompt wording -- both the original prompt that are used to elicit an initial answer and the subsequent prompt used to self-reflect. Specifically, although self-reflection may improve the reasoning capability of LLMs when the initial response is simple, the technique cannot improve upon the state-of-the-art chain-of-thought (CoT) prompting. Second, we show that self-reflection can lead to safer (75.8% reduction in toxic responses while preserving 97.8% non-toxic ones), less biased (77% reduction in gender biased responses, while preserving 94.3% unbiased ones), and more ideologically neutral responses (100% reduction in partisan leaning response, while preserving 87.7% non-partisan ones). The paper concludes by discussing the implications of our findings on the deployment of large language models. We release our experiments at https://github.com/Michael98Liu/self-reflection.