liyucheng09/llm-compressive
Longitudinal Evaluation of LLMs via Data Compression
This project helps evaluate how well large language models (LLMs) adapt to new information over time and how robust they are across different kinds of data. You provide an LLM from Hugging Face Hub and a dataset (like Wikipedia articles, news, or code) spanning various time periods, and it outputs a compression rate trend, showing how the model's performance changes over time. AI researchers, machine learning engineers, or anyone deploying LLMs can use this to understand a model's long-term generalization capabilities.
No commits in the last 6 months.
Use this if you need to assess the generalization and robustness of an LLM over a timeline with diverse datasets.
Not ideal if you're looking for a tool to fine-tune LLMs or measure performance on specific downstream tasks like sentiment analysis or question answering.
Stars
33
Forks
—
Language
Python
License
—
Category
Last pushed
May 29, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/liyucheng09/llm-compressive"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
ModelTC/LightCompress
[EMNLP 2024 & AAAI 2026] A powerful toolkit for compressing large models including LLMs, VLMs,...
p-e-w/heretic
Fully automatic censorship removal for language models
Orion-zhen/abliteration
Make abliterated models with transformers, easy and fast
YerbaPage/LongCodeZip
LongCodeZip: Compress Long Context for Code Language Models [ASE2025]
locuslab/wanda
A simple and effective LLM pruning approach.