DAMO-NLP-SG/LLM-Multilingual-Knowledge-Boundaries
[ACL 2025] Analyzing LLMs' Multilingual Knowledge Boundary Cognition Across Languages Through the Lens of Internal Representations
This project helps AI researchers and practitioners understand how Large Language Models (LLMs) determine if they know the answer to a question across different languages. By analyzing the model's internal processing, it takes multilingual question-answer pairs (some known, some unknown) and reveals where and how knowledge boundaries are perceived within the model's layers. This is crucial for anyone working on improving the reliability and reducing 'hallucinations' in multilingual LLMs, especially in low-resource languages.
Use this if you are a researcher or engineer investigating why LLMs sometimes "hallucinate" or confidently provide incorrect information, particularly when dealing with questions in multiple languages.
Not ideal if you are looking for a ready-to-use LLM application or a tool for general text generation or translation.
Stars
18
Forks
2
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Oct 18, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/DAMO-NLP-SG/LLM-Multilingual-Knowledge-Boundaries"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
ExtensityAI/symbolicai
A neurosymbolic perspective on LLMs
TIGER-AI-Lab/MMLU-Pro
The code and data for "MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding...
deep-symbolic-mathematics/LLM-SR
[ICLR 2025 Oral] This is the official repo for the paper "LLM-SR" on Scientific Equation...
microsoft/interwhen
A framework for verifiable reasoning with language models.
zhudotexe/fanoutqa
Companion code for FanOutQA: Multi-Hop, Multi-Document Question Answering for Large Language...