dorsal/arxiv
View SchemaEnhancing Self-Correction in Large Language Models through Multi-Perspective Reflection
| Authors | Mariana Costa, Alberlucia Rafael Soarez, Daniel Kim, Camila Ferreira |
|---|---|
| Categories | |
| ArXiv ID | 2601.07780vv1 |
| URL | https://arxiv.org/abs/2601.07780 |
| License | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
Abstract
While Chain-of-Thought (CoT) prompting advances LLM reasoning, challenges persist in consistency, accuracy, and self-correction, especially for complex or ethically sensitive tasks. Existing single-dimensional reflection methods offer insufficient improvements. We propose MyGO Poly-Reflective Chain-of-Thought (PR-CoT), a novel methodology employing structured multi-perspective reflection. After initial CoT, PR-CoT guides the LLM to self-assess its reasoning across multiple predefined angles: logical consistency, information completeness, biases/ethics, and alternative solutions. Implemented purely via prompt engineering, this process refines the initial CoT into a more robust and accurate final answer without model retraining. Experiments across arithmetic, commonsense, ethical decision-making, and logical puzzles, using GPT-three point five and GPT-four models, demonstrate PR-CoT's superior performance. It significantly outperforms traditional CoT and existing reflection methods in logical consistency and error correction, with notable gains in nuanced domains like ethical decision-making. Ablation studies, human evaluations, and qualitative analyses further validate the contribution of each reflection perspective and the overall efficacy of our poly-reflective paradigm in fostering more reliable LLM reasoning.
{
"annotation_id": "04649011-0d6e-49fd-9994-25f5f2120379",
"date_created": "2026-02-17T05:53:12.589000Z",
"date_modified": "2026-02-17T05:53:12.589000Z",
"file_hash": "34628593cefbd8cc76682fbc2a1a67eea48a1fd72a43553d1360a1ac7316850e",
"private": false,
"record": {
"abstract": "While Chain-of-Thought (CoT) prompting advances LLM reasoning, challenges persist in consistency, accuracy, and self-correction, especially for complex or ethically sensitive tasks. Existing single-dimensional reflection methods offer insufficient improvements. We propose MyGO Poly-Reflective Chain-of-Thought (PR-CoT), a novel methodology employing structured multi-perspective reflection. After initial CoT, PR-CoT guides the LLM to self-assess its reasoning across multiple predefined angles: logical consistency, information completeness, biases/ethics, and alternative solutions. Implemented purely via prompt engineering, this process refines the initial CoT into a more robust and accurate final answer without model retraining. Experiments across arithmetic, commonsense, ethical decision-making, and logical puzzles, using GPT-three point five and GPT-four models, demonstrate PR-CoT\u0027s superior performance. It significantly outperforms traditional CoT and existing reflection methods in logical consistency and error correction, with notable gains in nuanced domains like ethical decision-making. Ablation studies, human evaluations, and qualitative analyses further validate the contribution of each reflection perspective and the overall efficacy of our poly-reflective paradigm in fostering more reliable LLM reasoning.",
"arxiv_id": "2601.07780",
"authors": [
"Mariana Costa",
"Alberlucia Rafael Soarez",
"Daniel Kim",
"Camila Ferreira"
],
"categories": [
"cs.CL"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"title": "Enhancing Self-Correction in Large Language Models through Multi-Perspective Reflection",
"url": "https://arxiv.org/abs/2601.07780",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "bf156b39-ed71-4e19-9276-dab6c477f057",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}