Pinned Loading

1. acronym_corpus acronym_corpus Public

   A test set with 437 parallel term-acronym pairs in French and English for evaluating machine translation systems on technical acronyms.

2. europarl_t5_opus europarl_t5_opus Public

   Machine translation training on Europarl with T5 and Opus

   Jupyter Notebook

3. knowledge_distillation knowledge_distillation Public

   Distilling knowledge into a smaller Llama model and smaller BERT model

   Jupyter Notebook 1

4. rtd-gdes rtd-gdes Public

   ELECTRA-style training with gradient-disentangled embedding sharing in native PyTorch with HF transformers checkpoint

   Python 1

5. french-gemma french-gemma Public

   A Gemma3 training toolkit for training a blank checkpoint on French language data

   Python

6. AB3P_Pubmed AB3P_Pubmed Public

   AB3P output on 34GB of Pubmed abstracts