# Balanced Training Data Augmentation for Aspect-Based Sentiment Analysis

arXiv: https://arxiv.org/abs/2507.09485 · HTML: https://mlsanity.com/paper/2507.09485/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 53 references that are in the map.
- First posted on arXiv 13 Jul 2025.

## Builds on (most cited first, 25 of 25)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Semi-Supervised Classification with Graph Convolutional Networks](https://mlsanity.com/paper/1609.02907/) (2016)
- [Graph Attention Networks](https://mlsanity.com/paper/1710.10903/) (2017)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Efficient Estimation of Word Representations in Vector Space](https://mlsanity.com/paper/1301.3781/) (2013)
- [Character-level Convolutional Networks for Text Classification](https://mlsanity.com/paper/1509.01626/) (2015)
- [Unsupervised Data Augmentation for Consistency Training](https://mlsanity.com/paper/1904.12848/) (2019)
- [EDA: Easy Data Augmentation Techniques for Boosting Performance on Text Classification Tasks](https://mlsanity.com/paper/1901.11196/) (2019)
- [Beyond English-Centric Multilingual Machine Translation](https://mlsanity.com/paper/2010.11125/) (2020)
- [BERT Post-Training for Review Reading Comprehension and Aspect-based Sentiment Analysis](https://mlsanity.com/paper/1904.02232/) (2019)
- [Conditional BERT Contextual Augmentation](https://mlsanity.com/paper/1812.06705/) (2018)
- [Relational Graph Attention Network for Aspect-based Sentiment Analysis](https://mlsanity.com/paper/2004.12362/) (2020)
- [Data Noising as Smoothing in Neural Network Language Models](https://mlsanity.com/paper/1703.02573/) (2017)
- [Aspect Based Sentiment Analysis with Gated Convolutional Networks](https://mlsanity.com/paper/1805.07043/) (2018)
- [Aspect-based Sentiment Classification with Aspect-specific Graph Convolutional Networks](https://mlsanity.com/paper/1909.03477/) (2019)
- [Transformation Networks for Target-Oriented Sentiment Classification](https://mlsanity.com/paper/1805.01086/) (2018)
- [ZEN 2.0: Continue Training and Adaption for N-gram Enhanced Text Encoders](https://mlsanity.com/paper/2105.01279/) (2021)
- [Large language models for aspect-based sentiment analysis](https://mlsanity.com/paper/2310.18025/) (2023)
- [hyperdoc2vec: Distributed Representations of Hypertext Documents](https://mlsanity.com/paper/1805.03793/) (2018)
- [BiSyn-GAT+: Bi-Syntax Aware Graph Attention Network for Aspect-based Sentiment Analysis](https://mlsanity.com/paper/2204.03117/) (2022)
- [Improving Aspect-based Sentiment Analysis with Gated Graph Convolutional Networks and Syntax-based Regulation](https://mlsanity.com/paper/2010.13389/) (2020)
- [Exploring Model Editing for LLM-based Aspect-Based Sentiment Classification](https://mlsanity.com/paper/2503.15117/) (2025)
- [Large Language Models Enhanced by Plug and Play Syntactic Knowledge for Aspect-based Sentiment Analysis](https://mlsanity.com/paper/2506.12991/) (2025)
- [ZZU-NLP at SIGHAN-2024 dimABSA Task: Aspect-Based Sentiment Analysis with Coarse-to-Fine In-context Learning](https://mlsanity.com/paper/2407.15341/) (2024)

## Built on by (most cited first, 5 of 5)

- [Text Reinforcement for Multimodal Time Series Forecasting](https://mlsanity.com/paper/2509.00687/) (2025)
- [Training Data Selection with Gradient Orthogonality for Efficient Domain Adaptation](https://mlsanity.com/paper/2602.06359/) (2026)
- [Fine-grained Verbal Attack Detection via a Hierarchical Divide-and-Conquer Framework](https://mlsanity.com/paper/2601.06907/) (2026)
- [LEAD: Layer-wise Expert-aligned Decoding for Faithful Radiology Report Generation](https://mlsanity.com/paper/2602.04617/) (2026)
- [SURGE: An Event-Centric Social Media Sentiment Time Series Benchmark with Interaction Structure](https://mlsanity.com/paper/2605.21198/) (2026)

## Abstract (arXiv)

Aspect-based sentiment analysis (ABSA) is a crucial fine-grained task in social media scenarios to identify the sentiment polarity of specific aspect terms in a sentence. Although many existing studies leverage large language models (LLMs) to perform ABSA due to their strong context understanding capabilities, they still face challenges to learn the context information in the running text because of the short text, as well as the small and unbalanced labeled training data, where most data are labeled with positive sentiment. Data augmentation (DA) is a feasible strategy for providing richer contextual information, especially when using LLMs to create synthetic training data, but faces challenges in ensuring a high quality of the augmented data.In this paper, we propose an LLM-based ABSA approach with training data augmentation.Specifically, an LLM is prompted to generate augmented training data based on the original training data, so as to construct a new training data with larger size and balanced label distributions to better train an ABSA model. Meanwhile, in order to improve the quality of the augmented data, we propose a reinforcement learning approach to optimize the data augmentation. LLM.Experiment results and further analyses on English benchmark datasets for ABSA demonstrate the effectiveness of our approach, where superior performance is observed over strong baselines and most existing studies.