AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Better Fine-Tuning via Instance Weighting for Text Classification

Zhi Wang, Wei Bi, Yan Wang, Xiaojiang Liu

AAAI 2019 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Statistical transfer learning

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 soft-mixing instance weighting to fine-tuningIW-Fit fine-tuning · Amazon to Yelpaccuracy (%): 52.46 -> 53.54Improved
add soft-mixing instance weighting to fine-tuningIW-Fit fine-tuning · Yelp-2015 to Yelp-2016accuracy (%): 57.18 -> 58.30Improved
add soft-mixing instance weighting to fine-tuningIW-Fit fine-tuning · Yelp-2015 to Yelp-2017accuracy (%): 55.08 -> 55.90Improved
use prediction-loss (gradient) weights instead of uniform (ablation)IW-Fit fine-tuning · Yelp-2015 to Yelp-2017accuracy (%): 55.08 -> 55.08No clear effect
use variance-of-loss weights instead of prediction-loss weights (ablation)IW-Fit fine-tuning · Amazon to Yelpaccuracy (%): 52.80 -> 53.20Improved
use variance-of-loss weights instead of prediction-loss weights (ablation)IW-Fit fine-tuning · Yelp-2015 to Yelp-2016accuracy (%): 57.82 -> 57.60Hurt
soft-mixing instead of hard-mixing of the two metrics (ablation)IW-Fit fine-tuning · Amazon to Yelpaccuracy (%): 53.44 -> 53.54Improved
fine-tune over pre-training baseline: fine-tuned Uniform vs All (joint training) (ablation)Fine-tuned Uniform · Amazon to Yelpaccuracy (%): 50.94 -> 52.46Improved
replace LSTM with Text-CNN encoder, soft-mixing vs uniform (ablation)IW-Fit fine-tuning · Amazon to Yelpaccuracy (%): 47.94 -> 48.56Improved
fix encoder (fine-tune only FC) vs fine-tune all, Soft-mixing (ablation)IW-Fit fine-tuning · Amazon to Yelpaccuracy (%): 53.54 -> 48.23Hurt
fix word embedding only, Soft-mixing (ablation)IW-Fit fine-tuning · Yelp-2015 to Yelp-2016accuracy (%): 58.30 -> 57.60Hurt
vary target fine-tuning data to 25%, soft-mixing vs uniform (ablation)IW-Fit fine-tuning · Amazon to Yelpaccuracy (%): 49.72 -> 50.46Improved

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