AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

Multi-Label Few-Shot ICD Coding as Autoregressive Generation with Prompt

Zhichao Yang, Sunjae Kwon, Zonghai Yao, Hong Yu

AAAI 2023 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Zero-shot robustness and transfer · 17 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 autoregressive generation with a prompt and SOAP pretraining rather than standard fine-tuningICD coding · MIMIC-III-fewmacro F1: 4.3 -> 30.2Improved
Use autoregressive generation with a prompt and SOAP pretraining rather than standard fine-tuningICD coding · MIMIC-III-fullmacro F1: 10.4 -> 13.4Improved
Add SOAP assessment-and-plan generation pretraining instead of PubMed pretraining (ablation)GP · MIMIC-III-fewmacro F1: 7.3 -> 30.2Improved
Add SOAP assessment-and-plan generation pretraining instead of PubMed pretraining (ablation)GP · MIMIC-III-few, unmentioned diagnosesF1: 7.0 -> 17.1Improved
Add SOAP assessment-and-plan generation pretraining instead of PubMed pretraining (ablation)GP · MIMIC-III-few, mentioned diagnosesF1: 17.3 -> 21.3Improved
Use SOAP assessment-and-plan generation pretraining rather than Wikipedia and Book pretraining (ablation)GP · MIMIC-III-fewmacro F1: 2.9 -> 30.2Improved
Rerank MSMN and GPsoap candidate predictions using a prompted cross-attention modelMSMN + GPsoap · MIMIC-III-fullmacro F1: 10.4 -> 14.6Improved
Rerank MSMN and GPsoap candidate predictions instead of concatenating them (ablation)MSMN + GPsoap · MIMIC-III-fullmacro F1: 14.0 -> 14.6Improved
Rerank MSMN and AGMHT candidate predictions using a prompted cross-attention modelMSMN + AGMHT · MIMIC-III-fullmicro F1: 58.1 -> 59.7Improved
Replace SOAP pretraining with PubMed pretraining (ablation)GP · MIMIC-III-fullmicro F1: 44.2 -> 49.8Hurt
Use SOAP-pretrained generation rather than AGMHTICD coding · MIMIC-III-fewmicro precision: 49.4 -> 38.5Hurt

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