The reviewed record of science sign in
Pith

arxiv: 2406.14541 · v3 · pith:NBQXGMFI · submitted 2024-06-20 · cs.LG

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

Shengzhe Xu , Cho-Ting Lee , Mandar Sharma , Raquib Bin Yousuf , Nikhil Muralidhar , Naren Ramakrishnan This is my paper

Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:NBQXGMFIrecord.jsonopen to challenge →

classification cs.LG
keywords syntheticdatagenerationllmsfine-tuningtableadvancedas-is
0
0 comments X
read the original 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.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. LLM-TabLogic: Preserving Inter-Column Logical Relationships in Synthetic Tabular Data via Prompt-Guided Latent Diffusion

    cs.LG 2025-03 unverdicted novelty 7.0

    LLM-TabLogic extracts inter-column logical constraints using LLMs and conditions a score-based latent diffusion model on them to generate synthetic tabular data that preserves those relationships.

  2. Creating Artificial Students that Never Existed: Leveraging Large Language Models and CTGANs for Synthetic Data Generation

    cs.LG 2025-01 unverdicted novelty 3.0

    CTGAN and LLMs generate synthetic student data that passes statistical and predictive utility checks for learning analytics.

  3. Generative AI and Federated Learning for Intrusion Detection Systems: A Survey

    cs.CR 2026-07 unverdicted novelty 2.0

    A structured literature survey categorizing generative AI (autoencoders, GANs, diffusion models, LLMs) and federated learning uses in IDS, covering tasks like synthetic data generation and anomaly detection plus open ...