th789/mbr-for-nmt
Characterizing the performance of minimum Bayes risk (MBR) decoding for neural machine translation (NMT) when using various candidate generation methods by examining BLEU score, length bias, and token frequency bias. [CS287: Deep Learning for Natural Language Processing, Final Project, Fall 2021]
No commits in the last 6 months.
Stars
2
Forks
1
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Dec 11, 2021
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/th789/mbr-for-nmt"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
minggnim/nlp-models
A repository for training transformer based models
CPJKU/wechsel
Code for WECHSEL: Effective initialization of subword embeddings for cross-lingual transfer of...
yuanzhoulvpi2017/zero_nlp
中文nlp解决方案(大模型、数据、模型、训练、推理)
IntelLabs/nlp-architect
A model library for exploring state-of-the-art deep learning topologies and techniques for...
LoicGrobol/zeldarose
Train transformer-based models.