# Tabular Data Synthesis on Aggregate of Five Tabular Datasets Source: [Tabular Data Synthesis on Aggregate of Five Tabular Datasets]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:51:37.226Z Evaluation of models performing Tabular Data Synthesis across an aggregate set comprising five distinct tabular datasets, comparing full training data against original training data splits. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Tabular Data Synthesis | | Dataset (including any stated split) | Aggregate of five tabular datasets (full train vs original train) | | Metric | Marginal Error | | Unit | SCALAR | | Direction | Lower is better | | Results loaded | 13 of 13 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Marginal Error](): Lower is better; unit: SCALAR. - [Correlation Score](): Higher is better; unit: SCALAR. - [Precision](): Higher is better; unit: PERCENT. - [Recall](): Higher is better; unit: PERCENT. - [C2ST](): Higher is better; unit: SCALAR. - [JSD](): Lower is better; unit: SCALAR. ## Recorded results | Method | Marginal Error | Result date | Report paper | | --- | --- | --- | --- | | TabSyn | 1.4 | Not recorded | 68fa77241e209131cea7158f | | TabMT | 4.46 | Not recorded | 68fa77241e209131cea7158f | | RTF | 5.31 | Not recorded | 68fa77241e209131cea7158f | | TABGEN-ICL | 9.14 | Not recorded | 68fa77241e209131cea7158f | | CLLM | 10.57 | Not recorded | 68fa77241e209131cea7158f | | TABGEN-ICL | 11.39 | Not recorded | 68fa77241e209131cea7158f | | STaSy | 12.35 | Not recorded | 68fa77241e209131cea7158f | | CLLM | 13.17 | Not recorded | 68fa77241e209131cea7158f | | TabDDPM | 14.04 | Not recorded | 68fa77241e209131cea7158f | | TVAE | 14.61 | Not recorded | 68fa77241e209131cea7158f | | GReaT | 15.53 | Not recorded | 68fa77241e209131cea7158f | | CTGAN | 16.36 | Not recorded | 68fa77241e209131cea7158f | | CoDi | 21.7 | Not recorded | 68fa77241e209131cea7158f | ### TabSyn Marginal Error: 1.4. Result ID: 699469ce3a6c3148491d13f0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Diffusion-based"}. This record does not assert independent reproduction. ### TabMT Marginal Error: 4.46. Result ID: 699469ce3a6c3148491d13f2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Autoregressive Models"}. This record does not assert independent reproduction. ### RTF Marginal Error: 5.31. Result ID: 699469ce3a6c3148491d13f1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Autoregressive Models"}. This record does not assert independent reproduction. ### TABGEN-ICL Marginal Error: 9.14. Result ID: 699469ce3a6c3148491d13f7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"LLM-Prompt-Only","Base Model":"GPT-4o"}. This record does not assert independent reproduction. ### CLLM Marginal Error: 10.57. Result ID: 699469ce3a6c3148491d13f6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"LLM-Prompt-Only","Base Model":"GPT-4o"}. This record does not assert independent reproduction. ### TABGEN-ICL Marginal Error: 11.39. Result ID: 699469ce3a6c3148491d13f5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"LLM-Prompt-Only","Base Model":"GPT-4o-mini"}. This record does not assert independent reproduction. ### STaSy Marginal Error: 12.35. Result ID: 699469ce3a6c3148491d13ed. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Diffusion-based"}. This record does not assert independent reproduction. ### CLLM Marginal Error: 13.17. Result ID: 699469ce3a6c3148491d13f4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"LLM-Prompt-Only","Base Model":"GPT-4o-mini"}. This record does not assert independent reproduction. ### TabDDPM Marginal Error: 14.04. Result ID: 699469ce3a6c3148491d13ef. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Diffusion-based"}. This record does not assert independent reproduction. ### TVAE Marginal Error: 14.61. Result ID: 699469ce3a6c3148491d13eb. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"VAE-based"}. This record does not assert independent reproduction. ### GReaT Marginal Error: 15.53. Result ID: 699469ce3a6c3148491d13f3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"LLM-Finetuned"}. This record does not assert independent reproduction. ### CTGAN Marginal Error: 16.36. Result ID: 699469ce3a6c3148491d13ec. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"GAN-based"}. This record does not assert independent reproduction. ### CoDi Marginal Error: 21.7. Result ID: 699469ce3a6c3148491d13ee. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Diffusion-based"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.