InstructABSA: Instruction Learning for Aspect Based Sentiment Analysis
We introduce InstructABSA, an instruction learning paradigm for Aspect-Based Sentiment Analysis (ABSA) subtasks. Our method introduces positive, negative, and neutral examples to each training sample, and instruction tune the model (Tk-Instruct) for ABSA subtasks, yielding significant performance improvements. Experimental results on the Sem Eval 2014, 15, and 16 datasets demonstrate that InstructABSA outperforms the previous state-of-the-art (SOTA) approaches on Term Extraction (ATE), Sentiment Classification(ATSC) and Sentiment Pair Extraction (ASPE) subtasks. In particular, InstructABSA outperforms the previous state-of-the-art (SOTA) on the Rest14 ATE subtask by 5.69% points, the Rest15 ATSC subtask by 9.59% points, and the Lapt14 AOPE subtask by 3.37% points, surpassing 7x larger models. We also get competitive results on AOOE, AOPE, and AOSTE subtasks indicating strong generalization ability to all subtasks. Exploring sample efficiency reveals that just 50% train data is required to get competitive results with other instruction tuning approaches. Lastly, we assess the quality of instructions and observe that InstructABSA's performance experiences a decline of ~10% when adding misleading examples.
PDF AbstractCode
Datasets
Task | Dataset | Model | Metric Name | Metric Value | Global Rank | Benchmark |
---|---|---|---|---|---|---|
Aspect-Based Sentiment Analysis (ABSA) | SemEval-2014 Task-4 | InstructABSA | Restaurant (Acc) | 89.76 | # 3 | |
Laptop (Acc) | 88.37 | # 1 | ||||
Mean Acc (Restaurant + Laptop) | 89.065 | # 1 | ||||
Aspect Extraction | SemEval-2014 Task-4 | InstructABSA | Laptop (F1) | 92.30 | # 1 | |
Mean F1 (Laptop + Restaurant) | 92.53 | # 1 | ||||
Restaurant (F1) | 92.76 | # 1 | ||||
Aspect-Based Sentiment Analysis (ABSA) | SemEval 2014 Task 4 Laptop | InstructABSA | F1 | 79.34 | # 1 | |
Aspect Extraction | SemEval 2014 Task 4 Sub Task 1 | InstructABSA | Laptop (F1) | 92.30 | # 1 | |
Aspect-Based Sentiment Analysis (ABSA) | SemEval 2014 Task 4 Sub Task 1 | InstructABSA | Laptop (F1) | 92.30 | # 1 | |
Restaurant (F1) | 92.76 | # 1 | ||||
Sentiment Analysis | SemEval 2014 Task 4 Subtask 1+2 | InstructABSA | F1 | 79.34 | # 1 | |
Aspect-Based Sentiment Analysis (ABSA) | SemEval 2014 Task 4 Subtask 1+2 | InstructABSA | F1 | 79.34 | # 2 |