kvignesh1420/cot-icl-lab
[ACL 2025] Official implementation of the "CoT-ICL Lab" framework
This framework helps AI researchers and machine learning engineers study how large language models learn complex reasoning steps from examples. It generates synthetic, tokenized datasets based on customizable graph structures (DAGs) and token configurations. The output is a dataset ready for training and evaluating transformer models, complete with input IDs, attention masks, and chain-of-thought elements. This allows researchers to rigorously test hypotheses about in-context learning.
No commits in the last 6 months.
Use this if you are a machine learning researcher or engineer who needs to generate controlled synthetic datasets to understand how transformer models perform chain-of-thought reasoning from in-context examples.
Not ideal if you need to work with real-world, non-synthetic datasets or are looking for a general-purpose fine-tuning framework for pre-trained language models.
Stars
11
Forks
1
Language
Python
License
MIT
Category
Last pushed
Oct 10, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/kvignesh1420/cot-icl-lab"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
PaddlePaddle/PaddleNLP
Easy-to-use and powerful LLM and SLM library with awesome model zoo.
meta-llama/llama-cookbook
Welcome to the Llama Cookbook! This is your go to guide for Building with Llama: Getting started...
arcee-ai/mergekit
Tools for merging pretrained large language models.
changyeyu/LLM-RL-Visualized
๐100+ ๅๅ LLM / RL ๅ็ๅพ๐๏ผใๅคงๆจกๅ็ฎๆณใไฝ่ ๅทจ็ฎ๏ผ๐ฅ๏ผ100+ LLM/RL Algorithm Maps ๏ผ
mindspore-lab/step_into_llm
MindSpore online courses: Step into LLM