SALT-NLP/Adaptive-Compositional-Modules
Code for the ACL 2022 paper "Continual Sequence Generation with Adaptive Compositional Modules"
This project offers a way for machine learning researchers to train natural language processing models that can learn new tasks over time without forgetting previously learned information. You provide text data for a series of tasks, and the system trains a model that adapts and grows its knowledge incrementally, allowing for continuous learning. This is ideal for researchers working on lifelong learning or continual learning challenges in NLP.
No commits in the last 6 months.
Use this if you are an NLP researcher developing language models and need a framework for continually adding new knowledge and tasks without incurring 'catastrophic forgetting' or needing to retrain from scratch.
Not ideal if you are a practitioner looking for a ready-to-use NLP application or if you don't have experience with deep learning research frameworks like Hugging Face Transformers.
Stars
39
Forks
2
Language
Python
License
MIT
Category
Last pushed
Apr 04, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/SALT-NLP/Adaptive-Compositional-Modules"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
ShiZhengyan/InstructionModelling
[NeurIPS 2024 Main Track] Code for the paper titled "Instruction Tuning With Loss Over Instructions"
raymin0223/fast_robust_early_exit
Fast and Robust Early-Exiting Framework for Autoregressive Language Models with Synchronized...
oooranz/Baby-CoThought
🍼 Baby's CoThought: Leveraging LLMs for Enhanced Reasoning in Compact Models (BabyLM Challenge)
joisino/zeh
Code for "Even GPT-5.2 Can’t Count to Five: The Case for Zero-Error Horizons in Trustworthy LLMs"
yhy1117/X-Mixup
Implementation of ICLR 2022 paper "Enhancing Cross-lingual Transfer by Manifold Mixup".