# Targeted Data Generation: Finding and Fixing Model Weaknesses

arXiv: https://arxiv.org/abs/2305.17804 · HTML: https://mlsanity.com/paper/2305.17804/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 28 references that are in the map.
- First posted on arXiv 28 May 2023.

## Builds on (most cited first, 18 of 18)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [A large annotated corpus for learning natural language inference](https://mlsanity.com/paper/1508.05326/) (2015)
- [Beyond Accuracy: Behavioral Testing of NLP models with CheckList](https://mlsanity.com/paper/2005.04118/) (2020)
- [Just Train Twice: Improving Group Robustness without Training Group Information](https://mlsanity.com/paper/2107.09044/) (2021)
- [Dynabench: Rethinking Benchmarking in NLP](https://mlsanity.com/paper/2104.14337/) (2021)
- [No Subclass Left Behind: Fine-Grained Robustness in Coarse-Grained Classification Problems](https://mlsanity.com/paper/2011.12945/) (2020)
- [GPT3Mix: Leveraging Large-scale Language Models for Text Augmentation](https://mlsanity.com/paper/2104.08826/) (2021)
- [Not Enough Data? Deep Learning to the Rescue!](https://mlsanity.com/paper/1911.03118/) (2019)
- [Augmented SBERT: Data Augmentation Method for Improving Bi-Encoders for Pairwise Sentence Scoring Tasks](https://mlsanity.com/paper/2010.08240/) (2020)
- [Human-in-the-Loop for Data Collection: a Multi-Target Counter Narrative Dataset to Fight Online Hate Speech](https://mlsanity.com/paper/2107.08720/) (2021)
- [The Spotlight: A General Method for Discovering Systematic Errors in Deep Learning Models](https://mlsanity.com/paper/2107.00758/) (2021)
- [Robustness to Spurious Correlations via Human Annotations](https://mlsanity.com/paper/2007.06661/) (2020)
- [Can neural networks understand monotonicity reasoning?](https://mlsanity.com/paper/1906.06448/) (2019)
- [Detect and Perturb: Neutral Rewriting of Biased and Sensitive Text via Gradient-based Decoding](https://mlsanity.com/paper/2109.11708/) (2021)
- [SEAL : Interactive Tool for Systematic Error Analysis and Labeling](https://mlsanity.com/paper/2210.05839/) (2022)
- [Maximum Weighted Loss Discrepancy](https://mlsanity.com/paper/1906.03518/) (2019)
- [Collaborative Development of NLP models](https://mlsanity.com/paper/2305.12219/) (2023)

## Built on by (most cited first, 15 of 15)

- [A Survey on Knowledge Distillation of Large Language Models](https://mlsanity.com/paper/2402.13116/) (2024)
- [Usable XAI: 10 Strategies Towards Exploiting Explainability in the LLM Era](https://mlsanity.com/paper/2403.08946/) (2024)
- [Large Language Models as Annotators: Enhancing Generalization of NLP Models at Minimal Cost](https://mlsanity.com/paper/2306.15766/) (2023)
- [Data Augmentation using Large Language Models: Data Perspectives, Learning Paradigms and Challenges](https://mlsanity.com/paper/2403.02990/) (2024)
- [MedEval: A Multi-Level, Multi-Task, and Multi-Domain Medical Benchmark for Language Model Evaluation](https://mlsanity.com/paper/2310.14088/) (2023)
- [DISCERN: Decoding Systematic Errors in Natural Language for Text Classifiers](https://mlsanity.com/paper/2410.22239/) (2024)
- [SYNTHEVAL: Hybrid Behavioral Testing of NLP Models with Synthetic CheckLists](https://mlsanity.com/paper/2408.17437/) (2024)
- [LLM-enhanced Self-training for Cross-domain Constituency Parsing](https://mlsanity.com/paper/2311.02660/) (2023)
- [Explainable LLM-driven Multi-dimensional Distillation for E-Commerce Relevance Learning](https://mlsanity.com/paper/2411.13045/) (2024)
- [Synthetic Artifact Auditing: Tracing LLM-Generated Synthetic Data Usage in Downstream Applications](https://mlsanity.com/paper/2502.00808/) (2025)
- [Improving the OOD Performance of Closed-Source LLMs on NLI Through Strategic Data Selection](https://mlsanity.com/paper/2505.20209/) (2025)
- [Targeted Distillation for Sentiment Analysis](https://mlsanity.com/paper/2503.03225/) (2025)
- [Targeted Image Data Augmentation Increases Basic Skills Captioning Robustness](https://mlsanity.com/paper/2309.15991/) (2023)
- [Cross-domain Chinese Sentence Pattern Parsing](https://mlsanity.com/paper/2402.16311/) (2024)
- [What Is Wrong with My Model? Identifying Systematic Problems with Semantic Data Slicing](https://mlsanity.com/paper/2409.09261/) (2024)

## Abstract (arXiv)

Even when aggregate accuracy is high, state-of-the-art NLP models often fail systematically on specific subgroups of data, resulting in unfair outcomes and eroding user trust. Additional data collection may not help in addressing these weaknesses, as such challenging subgroups may be unknown to users, and underrepresented in the existing and new data. We propose Targeted Data Generation (TDG), a framework that automatically identifies challenging subgroups, and generates new data for those subgroups using large language models (LLMs) with a human in the loop. TDG estimates the expected benefit and potential harm of data augmentation for each subgroup, and selects the ones most likely to improve within group performance without hurting overall performance. In our experiments, TDG significantly improves the accuracy on challenging subgroups for state-of-the-art sentiment analysis and natural language inference models, while also improving overall test accuracy.