Ruan Chaves ruanchaves

👋 Hi, I'm Ruan

ruanchaves/hashformers: State-of-the-art framework for hashtag segmentation.
ruanchaves/napolab: The Natural Portuguese Language Benchmark.
- Browse the Napolab Leaderboard and stay up to date with the latest advancements in Portuguese language models.
- Medium Article: The Hidden Truth About LLM Performance: Why Your Benchmark Results Might Be Misleading
- Master Thesis: Lessons learned from the evaluation of Portuguese language models

Click on the links to view my pull requests.

argilla-io/argilla:
- Fixed bugs and shipped features related to semi-supervised learning (SSL) during my internship at Argilla.
huggingface/transformers:
- Modified the Trainer class for simultaneous Ray Tune and Weights & Biases execution.
nathanshartmann/portuguese_word_embeddings:
- Fixed a severe bug in the evaluation procedure.
- Documented the bug fix in the research paper "Portuguese language models...".
facebookresearch/BLINK:
- Fixed a parameter bug in the script for the BLINK benchmark.
awslabs/mlm-scoring:
- Addressed an installation instruction issue for the mlm-scoring library.

neuralmind-ai/coliee:
- Code for "To Tune or Not To Tune? Zero-shot Models for Legal Case Entailment"
- Code for "Yes, BM25 is a Strong Baseline for Legal Case Retrieval"
ruanchaves/elmo:
- Code for "Portuguese language models and word embeddings: evaluating on semantic similarity tasks".
ruanchaves/assin:
- Code for "Multilingual Transformer Ensembles for Portuguese Natural Language Tasks".
ruanchaves/BERT-WS:
- Code for "Domain adaptation of transformers for english word segmentation".