minstar

Why do we need [PAD] as a valid sequence?

2

In converting examples to features function (```convert_examples_to_features```), we set ```valid.append(1)``` at [PAD] position. Why do we have to use [PAD] position? Can anyone explain this to me?

[Details of Hunflair training]

5

In supplementary section 1.2 (Gold Standard NER pre-training), the authors described that the training sets are built by taking **the union of the HUNER train and test splits of each...

Can separate the entity_label evaluation results?

1

there are various labels (e.g., "ORG", "LOC", "MISC" etc.). Is there any way to evaluate separately? ```evaluate.py``` always shows overall results I think.

Retrieve metadata from Covidex

1

Hi, this is a really cool search engine tool on COVID-19! However, in the search engine, is there any way that we could get more metadata such as **cord_id** (document_id),...

Is there any way that we could measure variable ```MAX_NB```? While pretraining the code with our new dataset, we get below scores during training, ```[2100] Beta: 0.000, KL: 548.74, Word:...

This library contributes to facilitating further work in the biomedical domain! However, I cannot find how to reproduce the zero-shot performance of prompted language models. Could anyone help me?

Your abbreviation module is pretty good! However, I have a question about this module. Is there any way that I can know what index is abbreviated in the input text?...

Questions about performance improvement in Open LLM leaderboard

3

Hi, First of all, thank you for sharing your wonderful work! I was searching for efficient ways of mining instructions used in instruction-tuning LLMs. While reading the manuscript and investigating...