Why LLMs Are Bad at Synthetic Table Generation (and what to do about it)

• URL: http://arxiv.org/abs/2406.14541v3
• Date: Thu, 13 Mar 2025 21:19:46 GMT
• Title: Why LLMs Are Bad at Synthetic Table Generation (and what to do about it)
• Authors: Shengzhe Xu, Cho-Ting Lee, Mandar Sharma, Raquib Bin Yousuf, Nikhil Muralidhar, Naren Ramakrishnan,
• Abstract summary: Synthetic data generation is integral to ML pipelines, e.g., to augment training data, replace sensitive information, and even to power advanced platforms like DeepSeek.<n>While LLMs fine-tuned for synthetic data generation are gaining traction, synthetic table generation remains under-explored compared to text and image synthesis.<n>This paper shows that LLMs, whether used as-is or after traditional fine-tuning, are inadequate for generating synthetic tables.
• Score: 11.266896863556124
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Synthetic data generation is integral to ML pipelines, e.g., to augment training data, replace sensitive information, and even to power advanced platforms like DeepSeek. While LLMs fine-tuned for synthetic data generation are gaining traction, synthetic table generation -- a critical data type in business and science -- remains under-explored compared to text and image synthesis. This paper shows that LLMs, whether used as-is or after traditional fine-tuning, are inadequate for generating synthetic tables. Their autoregressive nature, combined with random order permutation during fine-tuning, hampers the modeling of functional dependencies and prevents capturing conditional mixtures of distributions essential for real-world constraints. We demonstrate that making LLMs permutation-aware can mitigate these issues.

Related papers

• Large Language Models for Data Synthesis [17.333852085464176]
  Large Language Models (LLMs) have potential as flexible, high-dimensional priors over real-world distributions.<n>We introduce LLM Synthor, a framework for data synthesis that transforms LLMs into structure-aware simulators guided by distributional feedback.<n>By minimizing discrepancies in the summary statistics space, the iterative synthesis loop aligns real and synthetic data.
  arXiv  Detail & Related papers  (2025-05-20T13:35:38Z)
• MetaSynth: Meta-Prompting-Driven Agentic Scaffolds for Diverse Synthetic Data Generation [10.231668557630577]
  We propose a method for generating synthetic data that enhances diversity through meta-prompting. We successfully adapt a well-trained LLM to two specialized domains-Finance and Biomedicine. Our findings suggest that a few million tokens of diverse synthetic data without mixing any real data, is sufficient for effective domain adaptation.
  arXiv  Detail & Related papers  (2025-04-17T01:25:15Z)
• Synthetic Data Generation Using Large Language Models: Advances in Text and Code [0.0]
  Large language models (LLMs) have unlocked new possibilities for generating synthetic training data in both natural language and code. We show how these methods enrich low-resource tasks such as classification and question answering. We address challenges like factual inaccuracies in generated text, lack of stylistic realism, and the risk of bias amplification.
  arXiv  Detail & Related papers  (2025-03-18T08:34:03Z)
• LLM-TabFlow: Synthetic Tabular Data Generation with Inter-column Logical Relationship Preservation [49.898152180805454]
  This study is the first to explicitly address inter-column relationship preservation in synthetic tabular data generation. LLM-TabFlow is a novel approach that captures complex inter-column relationships and compress data, while using Score-based Diffusion to model the distribution of the compressed data in latent space. Our results show that LLM-TabFlow outperforms all baselines, fully preserving inter-column relationships while achieving the best balance between data fidelity, utility, and privacy.
  arXiv  Detail & Related papers  (2025-03-04T00:47:52Z)
• Few-shot LLM Synthetic Data with Distribution Matching [37.55363714371521]
  Large language models (LLMs) produce high-quality synthetic data to enhance the performance of smaller models. LLMs-generated synthetic data often differs from the real data in key language attributes. We introduce SynAlign: a synthetic data generation and filtering framework based on key attribute distribution matching.
  arXiv  Detail & Related papers  (2025-02-09T16:43:32Z)
• Can a Large Language Model Learn Matrix Functions In Context? [3.7478782183628634]
  Large Language Models (LLMs) have demonstrated the ability to solve complex tasks through In-Context Learning (ICL) This paper explores the capacity of LLMs to solve non-linear numerical computations, with specific emphasis on functions of the Singular Value Decomposition.
  arXiv  Detail & Related papers  (2024-11-24T00:33:43Z)
• Understanding Synthetic Context Extension via Retrieval Heads [51.8869530817334]
  We investigate fine-tuning on synthetic data for three long-context tasks that require retrieval and reasoning. We find that models trained on synthetic data fall short of the real data, but surprisingly, the mismatch can be interpreted. Our results shed light on how to interpret synthetic data fine-tuning performance and how to approach creating better data for learning real-world capabilities over long contexts.
  arXiv  Detail & Related papers  (2024-10-29T17:55:00Z)
• Misinforming LLMs: vulnerabilities, challenges and opportunities [4.54019093815234]
  Large Language Models (LLMs) have made significant advances in natural language processing, but their underlying mechanisms are often misunderstood. This paper argues that current LLM architectures are inherently untrustworthy due to their reliance on correlations of sequential patterns of word embedding vectors. Research into combining generative transformer-based models with fact bases and logic programming languages may lead to the development of trustworthy LLMs.
  arXiv  Detail & Related papers  (2024-08-02T10:35:49Z)
• SELF-GUIDE: Better Task-Specific Instruction Following via Self-Synthetic Finetuning [70.21358720599821]
  Large language models (LLMs) hold the promise of solving diverse tasks when provided with appropriate natural language prompts. We propose SELF-GUIDE, a multi-stage mechanism in which we synthesize task-specific input-output pairs from the student LLM. We report an absolute improvement of approximately 15% for classification tasks and 18% for generation tasks in the benchmark's metrics.
  arXiv  Detail & Related papers  (2024-07-16T04:41:58Z)
• Unveiling the Flaws: Exploring Imperfections in Synthetic Data and Mitigation Strategies for Large Language Models [89.88010750772413]
  Synthetic data has been proposed as a solution to address the issue of high-quality data scarcity in the training of large language models (LLMs) Our work delves into these specific flaws associated with question-answer (Q-A) pairs, a prevalent type of synthetic data, and presents a method based on unlearning techniques to mitigate these flaws. Our work has yielded key insights into the effective use of synthetic data, aiming to promote more robust and efficient LLM training.
  arXiv  Detail & Related papers  (2024-06-18T08:38:59Z)
• SynthesizRR: Generating Diverse Datasets with Retrieval Augmentation [55.2480439325792]
  We study the synthesis of six datasets, covering topic classification, sentiment analysis, tone detection, and humor. We find that SynthesizRR greatly improves lexical and semantic diversity, similarity to human-written text, and distillation performance.
  arXiv  Detail & Related papers  (2024-05-16T12:22:41Z)
• EPIC: Effective Prompting for Imbalanced-Class Data Synthesis in Tabular Data Classification via Large Language Models [39.347666307218006]
  Large language models (LLMs) have demonstrated remarkable in-context learning capabilities across diverse applications. We introduce EPIC, a novel approach that leverages balanced, grouped data samples and consistent formatting with unique variable mapping to guide LLMs in generating accurate synthetic data across all classes, even for imbalanced datasets.
  arXiv  Detail & Related papers  (2024-04-15T17:49:16Z)
• Large Language Models Can Automatically Engineer Features for Few-Shot Tabular Learning [35.03338699349037]
  We propose a novel in-context learning framework, FeatLLM, which employs Large Language Models as feature engineers. FeatLLM generates high-quality rules, significantly (10% on average) outperforming alternatives such as TabLLM and STUNT.
  arXiv  Detail & Related papers  (2024-04-15T06:26:08Z)
• Mitigating Catastrophic Forgetting in Large Language Models with Self-Synthesized Rehearsal [49.24054920683246]
  Large language models (LLMs) suffer from catastrophic forgetting during continual learning. We propose a framework called Self-Synthesized Rehearsal (SSR) that uses the LLM to generate synthetic instances for rehearsal.
  arXiv  Detail & Related papers  (2024-03-02T16:11:23Z)
• Data Science with LLMs and Interpretable Models [19.4969442162327]
  Large language models (LLMs) are remarkably good at working with interpretable models. We show that LLMs can describe, interpret, and debug Generalized Additive Models (GAMs)
  arXiv  Detail & Related papers  (2024-02-22T12:04:15Z)
• Knowledge Fusion of Large Language Models [73.28202188100646]
  This paper introduces the notion of knowledge fusion for large language models (LLMs) We externalize their collective knowledge and unique strengths, thereby elevating the capabilities of the target model beyond those of any individual source LLM. Our findings confirm that the fusion of LLMs can improve the performance of the target model across a range of capabilities such as reasoning, commonsense, and code generation.
  arXiv  Detail & Related papers  (2024-01-19T05:02:46Z)
• Curated LLM: Synergy of LLMs and Data Curation for tabular augmentation in low-data regimes [57.62036621319563]
  We introduce CLLM, which leverages the prior knowledge of Large Language Models (LLMs) for data augmentation in the low-data regime. We demonstrate the superior performance of CLLM in the low-data regime compared to conventional generators.
  arXiv  Detail & Related papers  (2023-12-19T12:34:46Z)
• Simultaneous Machine Translation with Large Language Models [51.470478122113356]
  We investigate the possibility of applying Large Language Models to SimulMT tasks. We conducted experiments using the textttLlama2-7b-chat model on nine different languages from the MUST-C dataset. The results show that LLM outperforms dedicated MT models in terms of BLEU and LAAL metrics.
  arXiv  Detail & Related papers  (2023-09-13T04:06:47Z)
• LLMs Understand Glass-Box Models, Discover Surprises, and Suggest Repairs [10.222281712562705]
  We show that large language models (LLMs) are remarkably good at working with interpretable models. By adopting a hierarchical approach to reasoning, LLMs can provide comprehensive model-level summaries. We present the package $textttTalkToEBM$ as an open-source LLM-GAM interface.
  arXiv  Detail & Related papers  (2023-08-02T13:59:35Z)
• Synthetic data, real errors: how (not) to publish and use synthetic data [86.65594304109567]
  We show how the generative process affects the downstream ML task. We introduce Deep Generative Ensemble (DGE) to approximate the posterior distribution over the generative process model parameters.
  arXiv  Detail & Related papers  (2023-05-16T07:30:29Z)
• Augmenting Interpretable Models with LLMs during Training [73.40079895413861]
  We propose Augmented Interpretable Models (Aug-imodels) to build efficient and interpretable models. Aug-imodels use LLMs during fitting but not during inference, allowing complete transparency. We explore two instantiations of Aug-imodels in natural-language processing: (i) Aug-GAM, which augments a generalized additive model with decoupled embeddings from an LLM and (ii) Aug-Tree, which augments a decision tree with LLM feature expansions.
  arXiv  Detail & Related papers  (2022-09-23T18:36:01Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.