jianghoucheng/NSE
Neuron-Level Sequential Editing for Large Language Models, ACL 2025 Main
This tool helps researchers and AI engineers update large language models (LLMs) sequentially without breaking their previous knowledge or introducing errors. It takes a pre-trained LLM and new factual information or corrections, then efficiently integrates these updates directly into the model's 'neurons.' The output is an LLM that has learned the new information while retaining its original capabilities and accuracy.
No commits in the last 6 months.
Use this if you need to continuously update an existing large language model with new facts or corrections over time, ensuring it learns new information without 'forgetting' prior knowledge or introducing inconsistencies.
Not ideal if you are looking for a method to train a large language model from scratch or for parameter-intensive fine-tuning approaches that modify the model's weights extensively.
Stars
12
Forks
1
Language
Python
License
—
Category
Last pushed
Oct 15, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/jianghoucheng/NSE"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
sciknoworg/OntoAligner
OntoAligner: A Python Toolkit for Ontology Alignment https://pypi.org/project/OntoAligner/
Hi-archers/StableKE
Stable Knowledge Editing in Large Language Models
yuhui-zh15/NeQA
Official Code Release for "Beyond Positive Scaling: How Negation Impacts Scaling Trends of...