Pinned Loading

1. llm-rhyme-eval llm-rhyme-eval Public

   English and Dutch rhyming datasets (5k word pairs each) for five types of rhymes. Three open-source LLMs (Llama2, Llama3, CrystalChat) are tested on these datasets, with prompt variation.

   Roff 1

2. llm-medical-personas llm-medical-personas Public

   Examination of whether LLMs can maintain consistency over extended multiple text generation for 10 medical personas. 5 novel plausibility metrics proposed, and an ontology of common LLM errors.

   Python 1

3. s2st-cascading-e2e s2st-cascading-e2e Public

   A comparison of E2E and Cascading S2ST systems on the CVSS-C Spanish to English dataset (CommonVoice 4.0)

   Python 1

4. epitran-aadit epitran-aadit Public

   Forked from dmort27/epitran

   A tool for transcribing orthographic text as IPA (International Phonetic Alphabet)

   Python 1

5. neural-reinflection-paradigm-completion neural-reinflection-paradigm-completion Public

   Paradigm Completion and Reinflection for Kabardian (kbd), Swahili (swc), and Mixtec (xty).

   Python 1

6. split-learning-healthcare split-learning-healthcare Public

   Split learning for privacy-preserving healthcare, and threats and defensive techniques for decentralized learning. (with Prof. Vinay Chamola)

   Jupyter Notebook 2