pillowsofwind/Knowledge-Conflicts-Survey

[EMNLP 2024] The official GitHub repo for the survey paper "Knowledge Conflicts for LLMs: A Survey"

28
/ 100
Experimental

This is a comprehensive survey that helps researchers and practitioners understand how Large Language Models (LLMs) deal with conflicting information. It categorizes knowledge conflicts into three types: when an LLM's built-in knowledge clashes with new input, when different pieces of input information contradict each other, or when the LLM's own internal knowledge is inconsistent. This resource provides an organized overview of causes, behaviors, and solutions related to these conflicts, making it useful for anyone working with or researching LLMs.

153 stars. No commits in the last 6 months.

Use this if you are developing, deploying, or researching Large Language Models and need to understand the various ways they can encounter and handle conflicting information, including misinformation or outdated data.

Not ideal if you are a general user looking for a basic guide on how to use LLMs or how to write better prompts, as this delves into the technical aspects of LLM behavior.

Large Language Models (LLMs) AI Safety Natural Language Processing Information Reliability AI Research
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 8 / 25
Community 10 / 25

How are scores calculated?

Stars

153

Forks

8

Language

License

Last pushed

Sep 21, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/rag/pillowsofwind/Knowledge-Conflicts-Survey"

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