zhaochen0110/LMLM
Code and data for "Improving Temporal Generalization of Pre-trained Language Models with Lexical Semantic Change" (EMNLP2022)
This project helps researchers and data scientists working with language models that encounter issues with "temporal drift." It takes a pre-trained language model and unlabeled text data from a specific time period, along with labeled data for a downstream task, to produce an adapted language model. This adapted model is better at understanding language as it evolves over time, improving performance on tasks with changing vocabulary or word meanings.
No commits in the last 6 months.
Use this if your language model's performance degrades on data from different time periods due to words changing their meaning or usage.
Not ideal if your primary concern is improving language model performance without considering temporal shifts in language.
Stars
18
Forks
1
Language
Python
License
—
Category
Last pushed
Dec 08, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/zhaochen0110/LMLM"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
MadryLab/context-cite
Attribute (or cite) statements generated by LLMs back to in-context information.
microsoft/augmented-interpretable-models
Interpretable and efficient predictors using pre-trained language models. Scikit-learn compatible.
Trustworthy-ML-Lab/CB-LLMs
[ICLR 25] A novel framework for building intrinsically interpretable LLMs with...
poloclub/LLM-Attributor
LLM Attributor: Attribute LLM's Generated Text to Training Data
THUDM/LongCite
LongCite: Enabling LLMs to Generate Fine-grained Citations in Long-context QA