# Hierarchical Table Question Answering on HiTab Source: [Hierarchical Table Question Answering on HiTab]() Collection index: [llms.txt]() Data updated: 2026-06-11T07:26:49.439Z Evaluation of model performance on the HiTab dataset for Hierarchical Table Question Answering, measured using Accuracy. 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 | Hierarchical Table Question Answering | | Dataset (including any stated split) | HiTab | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 19 of 19 | | 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](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Formula-R1-Plus | 87.24 | Not recorded | 68fa78f91e209131cea7b6bd | | E5 | 85.08 | Not recorded | 68fa78f91e209131cea7b6bd | | SS-CoT | 79.1 | Not recorded | 68fa78f91e209131cea7b6bd | | TabAF | 78.41 | Not recorded | 68fa78f91e209131cea7b6bd | | TableGPT2 | 70.27 | Not recorded | 68fa78f91e209131cea7b6bd | | TabClaw | 69.8 | Not recorded | 6a28f31a90966ade919c15e2 | | Formula-R1 | 69.74 | Not recorded | 68fa78f91e209131cea7b6bd | | API-Assisted | 69.3 | Not recorded | 68fa78f91e209131cea7b6bd | | SheetAgent | 69.1 | Not recorded | 6a28f31a90966ade919c15e2 | | Binder | 65.8 | Not recorded | 6a28f31a90966ade919c15e2 | | SpreadsheetLLM | 64.6 | Not recorded | 6a28f31a90966ade919c15e2 | | Vanilla | 63.6 | Not recorded | 6a28f31a90966ade919c15e2 | | Chain-of-Table | 63.4 | Not recorded | 6a28f31a90966ade919c15e2 | | ReAct | 63.4 | Not recorded | 6a28f31a90966ade919c15e2 | | TableLlama | 60.48 | Not recorded | 68fa78f91e209131cea7b6bd | | NormTab | 60.3 | Not recorded | 6a28f31a90966ade919c15e2 | | FORTAP | 47 | Not recorded | 68fa78f91e209131cea7b6bd | | TAPEX-Large | 45.6 | Not recorded | 68fa78f91e209131cea7b6bd | | TableLLM | 43.88 | Not recorded | 68fa78f91e209131cea7b6bd | ### Formula-R1-Plus Accuracy: 87.24. Result ID: 69c29ed947f34ce924ab4461. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen 2.5-Coder 7B","Evaluation Protocol":"Finetuning-Based"}. This record does not assert independent reproduction. ### E5 Accuracy: 85.08. Result ID: 69c29ed947f34ce924ab4458. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"GPT-4","Evaluation Protocol":"Prompting-Based"}. This record does not assert independent reproduction. ### SS-CoT Accuracy: 79.1. Result ID: 69c29ed947f34ce924ab4459. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama-3.1 70B","Evaluation Protocol":"Prompting-Based"}. This record does not assert independent reproduction. ### TabAF Accuracy: 78.41. Result ID: 69c29ed947f34ce924ab445f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen 2.5-Coder 7B","Evaluation Protocol":"Finetuning-Based"}. This record does not assert independent reproduction. ### TableGPT2 Accuracy: 70.27. Result ID: 69c29ed947f34ce924ab445e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen 2.5 7B","Evaluation Protocol":"Finetuning-Based"}. This record does not assert independent reproduction. ### TabClaw Accuracy: 69.8. Result ID: 6a2a632efbc373592c271a54. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Ours","Backbone":"DeepSeek-v3.2"}. This record does not assert independent reproduction. ### Formula-R1 Accuracy: 69.74. Result ID: 69c29ed947f34ce924ab4460. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen 2.5-Coder 7B","Evaluation Protocol":"Finetuning-Based"}. This record does not assert independent reproduction. ### API-Assisted Accuracy: 69.3. Result ID: 69c29ed947f34ce924ab4457. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"CodeX","Evaluation Protocol":"Prompting-Based"}. This record does not assert independent reproduction. ### SheetAgent Accuracy: 69.1. Result ID: 6a2a632efbc373592c271a53. [Reporting paper]() Recorded evaluation configuration: {"Category":"Autonomous Agent","Backbone":"DeepSeek-v3.2"}. This record does not assert independent reproduction. ### Binder Accuracy: 65.8. Result ID: 6a2a632efbc373592c271a4e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Context Augmentation","Backbone":"DeepSeek-v3.2"}. This record does not assert independent reproduction. ### SpreadsheetLLM Accuracy: 64.6. Result ID: 6a2a632efbc373592c271a51. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Context Augmentation","Backbone":"DeepSeek-v3.2"}. This record does not assert independent reproduction. ### Vanilla Accuracy: 63.6. Result ID: 6a2a632efbc373592c271a4d. [Reporting paper]() Recorded evaluation configuration: {"Category":"Direct Inference","Backbone":"DeepSeek-v3.2"}. This record does not assert independent reproduction. ### Chain-of-Table Accuracy: 63.4. Result ID: 6a2a632efbc373592c271a4f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Context Augmentation","Backbone":"DeepSeek-v3.2"}. This record does not assert independent reproduction. ### ReAct Accuracy: 63.4. Result ID: 6a2a632efbc373592c271a52. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Autonomous Agent","Backbone":"DeepSeek-v3.2"}. This record does not assert independent reproduction. ### TableLlama Accuracy: 60.48. Result ID: 69c29ed947f34ce924ab445c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama-2 7B","Evaluation Protocol":"Finetuning-Based"}. This record does not assert independent reproduction. ### NormTab Accuracy: 60.3. Result ID: 6a2a632efbc373592c271a50. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Context Augmentation","Backbone":"DeepSeek-v3.2"}. This record does not assert independent reproduction. ### FORTAP Accuracy: 47. Result ID: 69c29ed947f34ce924ab445a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"BERT+LSTM","Evaluation Protocol":"Finetuning-Based"}. This record does not assert independent reproduction. ### TAPEX-Large Accuracy: 45.6. Result ID: 69c29ed947f34ce924ab445b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"BART-Large","Evaluation Protocol":"Finetuning-Based"}. This record does not assert independent reproduction. ### TableLLM Accuracy: 43.88. Result ID: 69c29ed947f34ce924ab445d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen 7B","Evaluation Protocol":"Finetuning-Based"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.