AI Paper Atlas

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

Embroid: Unsupervised Prediction Smoothing Can Improve Few-Shot Classification

Neel Guha, Mayee F Chen, Kush Bhatia, Azalia Mirhoseini, Frederic Sala, Christopher Re

NeurIPS 2023 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Fine-tuning strategies for few-shot · 4 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
Add neighborhood-smoothed predictions from embeddings and combine them with the original prompt predictionPrompt-based classifiers · 95 tasks, GPT-JTF1 improvement: 7.3Improved
Use Embroid prediction correctionGPT-3.5 prompts · 12 representative tasksF1 improvement: 4.9Improved
Use Embroid instead of three-prompt aggregationJ1-Jumbo predictions · All tasks (average macro-F1)macro-F1: 60.7 -> 64.5Improved
Use Embroid instead of three-prompt aggregationGPT-3.5 predictions · All tasks (average macro-F1)macro-F1: 84.7 -> 86.0Improved
Apply Embroid to base prompt predictionsGPT-3.5 · 10 tasksaverage F1: 76.3 -> 81.9Improved
Apply Embroid to chain-of-thought prompt predictionsGPT-3.5 · 10 tasksaverage F1: 80.1 -> 85.4Improved
Use two legal embedding spaces rather than general BERT and RoBERTa (ablation)Embroid with GPT-JT · Law taskstask win rate: 77\% of tasks; up to 6 points F1Improved
Use two science embedding spaces rather than general BERT and RoBERTa (ablation)Embroid with GPT-JT · Science taskstask win rate: 92\% of tasks; up to 4.3 points F1Improved
Add text-embedding-ada-002 embeddings (ablation)Embroid · 19 tasksF1 improvement: up to 13 points F1; average of 2 pointsImproved
Increase the number of prompts used by Embroid from one to three (ablation)Embroid · All tasks (average macro-F1), J1-Jumbomacro-F1: 60.4 -> 64.5Improved
Increase the number of prompts used by Embroid from one to three (ablation)Embroid · All tasks (average macro-F1), GPT-3.5macro-F1: 83.9 -> 86.0Improved
Use selective annotation with 100 labeled examples to construct the prompt, then apply EmbroidGPT-JT · All tasksF1 improvement: 88\% of tasks; average of 4.3 pointsImproved

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