{"id":24702632,"url":"https://github.com/chihaya-yuka/multiplex-cot","last_synced_at":"2025-07-06T12:37:05.600Z","repository":{"id":273288619,"uuid":"919218803","full_name":"Chihaya-Yuka/Multiplex-CoT","owner":"Chihaya-Yuka","description":"[arXiv 2501.13117]The Multiplex CoT makes AI more thoughtful.","archived":false,"fork":false,"pushed_at":"2025-02-09T12:13:51.000Z","size":191,"stargazers_count":18,"open_issues_count":0,"forks_count":1,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-04-07T22:23:59.330Z","etag":null,"topics":["api","cot","deep-learning","llm","lrm","orange-ai","prompt-engineering"],"latest_commit_sha":null,"homepage":"https://arxiv.org/abs/2501.13117","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Chihaya-Yuka.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2025-01-20T01:26:24.000Z","updated_at":"2025-03-25T09:06:29.000Z","dependencies_parsed_at":"2025-03-22T04:34:59.306Z","dependency_job_id":null,"html_url":"https://github.com/Chihaya-Yuka/Multiplex-CoT","commit_stats":null,"previous_names":["data-dream-gdsp/multiplex-cot","chihaya-yuka/multiplex-cot"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/Chihaya-Yuka/Multiplex-CoT","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Chihaya-Yuka%2FMultiplex-CoT","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Chihaya-Yuka%2FMultiplex-CoT/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Chihaya-Yuka%2FMultiplex-CoT/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Chihaya-Yuka%2FMultiplex-CoT/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Chihaya-Yuka","download_url":"https://codeload.github.com/Chihaya-Yuka/Multiplex-CoT/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Chihaya-Yuka%2FMultiplex-CoT/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":260684209,"owners_count":23046105,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["api","cot","deep-learning","llm","lrm","orange-ai","prompt-engineering"],"created_at":"2025-01-27T05:41:53.087Z","updated_at":"2025-07-06T12:37:05.541Z","avatar_url":"https://github.com/Chihaya-Yuka.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Multiplex CoT: A way for the LLM to review its own thinking while reasoning by initiating double CoT thinking\n\n[![Open Auto-CoT in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1rB3Re3D7alu28JgChFUy6BKmvmNADsdk?usp=sharing) [![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/mygo-multiplex-cot-a-method-for-self/gsm8k-on-gsm8k)](https://paperswithcode.com/sota/gsm8k-on-gsm8k?p=mygo-multiplex-cot-a-method-for-self) [![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/mygo-multiplex-cot-a-method-for-self/humaneval-on-humaneval-1)](https://paperswithcode.com/sota/humaneval-on-humaneval-1?p=mygo-multiplex-cot-a-method-for-self) [![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/mygo-multiplex-cot-a-method-for-self/llm-real-life-tasks-on-llm-real-life-tasks)](https://paperswithcode.com/sota/llm-real-life-tasks-on-llm-real-life-tasks?p=mygo-multiplex-cot-a-method-for-self) [![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/mygo-multiplex-cot-a-method-for-self/mmlu-on-mmlu-pro)](https://paperswithcode.com/sota/mmlu-on-mmlu-pro?p=mygo-multiplex-cot-a-method-for-self)\n\n\nBy employing the Prompt method, the LLM can attain an effect that closely resembles that of the LRM without necessitating additional training.\n\nIn the context of reasoning and decision-making, Multiplex CoT (Chain of Thought) enables the model to simulate a form of self-reflection, improving its ability to generate coherent, logical answers. This method works by prompting the LLM to first generate a chain of reasoning (CoT), then iteratively reviewing and refining it by initiating a second round of reasoning, which acts as a critique or review of the first.\n\n![Figure 1](Figure_1.png)\n\n## Quickly start\n\nRun `Multiplex_CoT.ipynb`.\n\n## How to use\n\nSee `example.py`.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fchihaya-yuka%2Fmultiplex-cot","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fchihaya-yuka%2Fmultiplex-cot","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fchihaya-yuka%2Fmultiplex-cot/lists"}