# Table Question Answering on ROBUT-WTQ Column Extension (test) Source: [Table Question Answering on ROBUT-WTQ Column Extension (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:34:23.952Z Evaluation of model performance on the test split of the ROBUT-WTQ dataset, specifically testing robustness against column extension perturbations for Table Question Answering. 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 | Table Question Answering | | Dataset (including any stated split) | ROBUT-WTQ Column Extension (test) | | Metric | Accuracy (Pre-perturbation) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 10 of 10 | | 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 - [Accuracy (Pre-perturbation)](): Higher is better; unit: PERCENT. - [Accuracy (Post-perturbation)](): Higher is better; unit: PERCENT. - [Robustness Accuracy (R-ACC)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy (Pre-perturbation) | Result date | Report paper | | --- | --- | --- | --- | | TAPEX | 92 | Not recorded | 68fa6fdc1e209131cea47b48 | | OmniTab | 91.2 | Not recorded | 68fa6fdc1e209131cea47b48 | | TAPAS | 89.8 | Not recorded | 68fa6fdc1e209131cea47b48 | | TableFormer | 86 | Not recorded | 68fa6fdc1e209131cea47b48 | | GPT-3 | 80.2 | Not recorded | 68fa6fdc1e209131cea47b48 | | OmniTab | 63.7 | Not recorded | 68fa6fdc1e209131cea47b48 | | TAPEX | 61.2 | Not recorded | 68fa6fdc1e209131cea47b48 | | TableFormer | 52.5 | Not recorded | 68fa6fdc1e209131cea47b48 | | TAPAS | 50.9 | Not recorded | 68fa6fdc1e209131cea47b48 | | GPT-3 | 47.9 | Not recorded | 68fa6fdc1e209131cea47b48 | ### TAPEX Accuracy (Pre-perturbation): 92. Result ID: 6990ef833a6c314849046827. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation Protocol":"Fine-tuned"}. This record does not assert independent reproduction. ### OmniTab Accuracy (Pre-perturbation): 91.2. Result ID: 6990ef833a6c314849046828. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation Protocol":"Fine-tuned"}. This record does not assert independent reproduction. ### TAPAS Accuracy (Pre-perturbation): 89.8. Result ID: 6990ef833a6c314849046825. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation Protocol":"Fine-tuned"}. This record does not assert independent reproduction. ### TableFormer Accuracy (Pre-perturbation): 86. Result ID: 6990ef833a6c314849046826. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation Protocol":"Fine-tuned"}. This record does not assert independent reproduction. ### GPT-3 Accuracy (Pre-perturbation): 80.2. Result ID: 6990ef833a6c314849046829. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation Protocol":"Few-shot","Sample Size":"200"}. This record does not assert independent reproduction. ### OmniTab Accuracy (Pre-perturbation): 63.7. Result ID: 6990ef833a6c3148490467f1. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation Protocol":"Fine-tuned"}. This record does not assert independent reproduction. ### TAPEX Accuracy (Pre-perturbation): 61.2. Result ID: 6990ef833a6c3148490467f0. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation Protocol":"Fine-tuned"}. This record does not assert independent reproduction. ### TableFormer Accuracy (Pre-perturbation): 52.5. Result ID: 6990ef833a6c3148490467ef. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation Protocol":"Fine-tuned"}. This record does not assert independent reproduction. ### TAPAS Accuracy (Pre-perturbation): 50.9. Result ID: 6990ef833a6c3148490467ee. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation Protocol":"Fine-tuned"}. This record does not assert independent reproduction. ### GPT-3 Accuracy (Pre-perturbation): 47.9. Result ID: 6990ef833a6c3148490467f2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation Protocol":"Few-shot"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.