AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

BigBio: A Framework for Data-Centric Biomedical Natural Language Processing

Jason Alan Fries, Leon Weber, Natasha Seelam, Gabriel Altay, Debajyoti Datta, Samuele Garda, Myungsun Kang, Ruisi Su, Wojciech Kusa, Samuel Cahyawijaya, Fabio Barth, Simon Ott and 31 more

NeurIPS 2022 · Large language models › Language, safety and interpretability · approach: Controlled text generation · 12 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use T0++ (more prompted pretraining tasks) instead of T0T0 family · BioASQ zero-shotaccuracy avg: 76.1 -> 89.0Improved
use T0-family prompted multitask training instead of GPT/T5/SciFivezero-shot biomedical prompts · BioASQ zero-shotaccuracy avg: 40.4 -> 76.1Improved
pretraining on PMC biomedical textGPT-Neo/GPT-J/SciFive · zero-shot biomedical tasksaccuracyNo clear effect
add task-specific fine-tuning to the 107-task MTL modelBioLinkBERT-base MTL · NCBI-DiseaseF1: 80.2 -> 87.5 ± 0.9Improved
add task-specific fine-tuning to the 107-task MTL modelBioLinkBERT-base MTL · ChemProtF1: 66.4 -> 74.3 ± 0.1Improved
add task-specific fine-tuning to the 107-task MTL modelBioLinkBERT-base MTL · BC5CDR-ChemicalF1: 92.2 -> 94.4 ± 0.3Improved
large-scale MTL pre-finetuning then fine-tuning vs LinkBERT-base aloneBioLinkBERT-base · BLURB subsetsF1: ChemProt 77.6 vs 74.3 ± 0.1Mixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy