sinclaircoder / fs-absa

FS-ABSA: A Simple yet Effective Framework for Few-Shot Aspect-Based Sentiment Analisys

Quick links

• Overview
• Requirements
• Run FS-ABSA
  • Fully-supervised Setting
  • Non-English Low-resource Setting
• Main Results
• Ablation Study
• Any Question?
• Citation

Overview

In this work, we introduce a simple yet effective framework called FS-ABSA, which involves domain-adaptive pre-training and textinfilling fine-tuning. Specifically,

• we approach the End-to-End ABSA task as a text-infilling problem.
• we perform domain-adaptive pre-training with the text-infilling objective, narrowing the two gaps, i.e., domain gap and objective gap, and consequently facilitating the knowledge transfer.

Requirements

To run the code, please install all the dependency packages by using the following command:

pip install -r requirements.txt

NOTE: All experiments are conducted on NVIDIA RTX 3090 (and Linux OS). Different versions of packages and GPU may lead to different results.

Run FS-ABSA

NOTE: All experiment scripts are with multiple runs (three seeds).

Fully-supervised Setting

## English Dataset: 14lap
$ bash script/run_aspe_14lap.sh

## English Dataset: 14res
$ bash script/run_aspe_14res.sh

## Dutch Dataset: 16res
$ bash script/run_aspe_dutch.sh

## French Dataset: 16res
$ bash script/run_aspe_french.sh

Few-Shot Setting

## English Dataset: 14lap
$ bash script/run_aspe_fewshot_14lap.sh

## English Dataset: 14res
$ bash script/run_aspe_fewshot_14res.sh

## Dutch Dataset: 16res
$ bash script/run_aspe_fewshot_dutch.sh

## French Dataset: 16res
$ bash script/run_aspe_fewshot_french.sh

Main Results

Results on 14-Lap and 14-Res under different training data size scenarios

Comparison with SOTA under the full data setting

Results in two low-resource languages under different training data sizes

Ablation Study

Citation

Any Questions?

If you have any questions related to this work, you can open an issue with details or feel free to email Zengzhi([email protected]), Qiming([email protected]).

Acknowledgements

Our code is based on ABSA-QUAD. Thanks for their work.