Hi-archers/StableKE

Stable Knowledge Editing in Large Language Models

13
/ 100
Experimental

When you need to update facts within a large language model (LLM) — like changing a person's nationality or a company's CEO — this project helps ensure those edits are accurate and don't accidentally break other important information. It takes your desired fact changes and a set of evaluation questions, then helps you verify that the LLM correctly reflects the new facts, maintains its reasoning on related 'multi-hop' knowledge, and doesn't forget unrelated information. This is for AI developers, researchers, or ML engineers who are directly working with and modifying LLMs.

No commits in the last 6 months.

Use this if you are developing or fine-tuning LLMs and need to precisely update their internal knowledge without causing unintended side effects or 'catastrophic forgetting'.

Not ideal if you are an end-user of an LLM and simply want to ask it questions or if you are looking for a tool to train an LLM from scratch.

LLM-development knowledge-editing AI-safety model-maintenance fact-updating
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 8 / 25
Community 0 / 25

How are scores calculated?

Stars

10

Forks

Language

Python

License

Last pushed

Mar 26, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/Hi-archers/StableKE"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.