# Claim Verification on SciTab-OD Source: [Claim Verification on SciTab-OD]() Collection index: [llms.txt]() Data updated: 2026-04-21T09:57:49.824Z Evaluation of model performance on the SciTab-OD dataset for the Claim Verification task, reporting Macro F1 score. 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 | Claim Verification | | Dataset (including any stated split) | SciTab-OD | | Metric | Macro F1 | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 28 of 28 | | 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 - [Macro F1](): Higher is better; unit: PERCENT. ## Recorded results | Method | Macro F1 | Result date | Report paper | | --- | --- | --- | --- | | Llama-400B | 77 | Not recorded | 69e704cb7d0172140ec7efa0 | | MACE | 76 | Not recorded | 69e704cb7d0172140ec7efa0 | | Llama-70B | 74 | Not recorded | 69e704cb7d0172140ec7efa0 | | Qwen-235B | 74 | Not recorded | 69e704cb7d0172140ec7efa0 | | DeepSeek-V3.1 | 74 | Not recorded | 69e704cb7d0172140ec7efa0 | | DeepSeek-70B | 73 | Not recorded | 69e704cb7d0172140ec7efa0 | | Llama-400B | 73 | Not recorded | 69e704cb7d0172140ec7efa0 | | DeepSeek-70B | 72 | Not recorded | 69e704cb7d0172140ec7efa0 | | Qwen-72B | 72 | Not recorded | 69e704cb7d0172140ec7efa0 | | GPT-4o | 71 | Not recorded | 69e704cb7d0172140ec7efa0 | | Claude-3.5-Sonnet | 71 | Not recorded | 69e704cb7d0172140ec7efa0 | | Claude-3.5-Sonnet | 71 | Not recorded | 69e704cb7d0172140ec7efa0 | | MACE | 71 | Not recorded | 69e704cb7d0172140ec7efa0 | | Qwen-72B | 69 | Not recorded | 69e704cb7d0172140ec7efa0 | | Gemma-12B | 68 | Not recorded | 69e704cb7d0172140ec7efa0 | | Qwen-235B | 68 | Not recorded | 69e704cb7d0172140ec7efa0 | | MACE | 66 | Not recorded | 69e704cb7d0172140ec7efa0 | | Mistral-7B | 65 | Not recorded | 69e704cb7d0172140ec7efa0 | | Llama-70B | 65 | Not recorded | 69e704cb7d0172140ec7efa0 | | Gemma-12B | 64 | Not recorded | 69e704cb7d0172140ec7efa0 | | GPT-4o | 60 | Not recorded | 69e704cb7d0172140ec7efa0 | | MACE | 60 | Not recorded | 69e704cb7d0172140ec7efa0 | | Llama-8B | 59 | Not recorded | 69e704cb7d0172140ec7efa0 | | Mistral-7B | 58 | Not recorded | 69e704cb7d0172140ec7efa0 | | Llama-3B | 57 | Not recorded | 69e704cb7d0172140ec7efa0 | | Llama-8B | 56 | Not recorded | 69e704cb7d0172140ec7efa0 | | DeepSeek-V3.1 | 56 | Not recorded | 69e704cb7d0172140ec7efa0 | | Llama-3B | 37 | Not recorded | 69e704cb7d0172140ec7efa0 | ### Llama-400B Macro F1: 77. Result ID: 69e74a0f47f34ce924ce0181. [Reporting paper]() Recorded evaluation configuration: {"Reasoning":"w COT"}. This record does not assert independent reproduction. ### MACE Macro F1: 76. Result ID: 69e74a0f47f34ce924ce018c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qw-235B","Configuration":"Pm + Em + Vm"}. This record does not assert independent reproduction. ### Llama-70B Macro F1: 74. Result ID: 69e74a0f47f34ce924ce0179. [Reporting paper]() Recorded evaluation configuration: {"Reasoning":"w COT"}. This record does not assert independent reproduction. ### Qwen-235B Macro F1: 74. Result ID: 69e74a0f47f34ce924ce017f. [Reporting paper]() Recorded evaluation configuration: {"Reasoning":"w COT"}. This record does not assert independent reproduction. ### DeepSeek-V3.1 Macro F1: 74. Result ID: 69e74a0f47f34ce924ce0183. [Reporting paper]() Recorded evaluation configuration: {"Reasoning":"w COT"}. This record does not assert independent reproduction. ### DeepSeek-70B Macro F1: 73. Result ID: 69e74a0f47f34ce924ce017c. [Reporting paper]() Recorded evaluation configuration: {"Reasoning":"wo COT"}. This record does not assert independent reproduction. ### Llama-400B Macro F1: 73. Result ID: 69e74a0f47f34ce924ce0182. [Reporting paper]() Recorded evaluation configuration: {"Reasoning":"wo COT"}. This record does not assert independent reproduction. ### DeepSeek-70B Macro F1: 72. Result ID: 69e74a0f47f34ce924ce017b. [Reporting paper]() Recorded evaluation configuration: {"Reasoning":"w COT"}. This record does not assert independent reproduction. ### Qwen-72B Macro F1: 72. Result ID: 69e74a0f47f34ce924ce017d. [Reporting paper]() Recorded evaluation configuration: {"Reasoning":"w COT"}. This record does not assert independent reproduction. ### GPT-4o Macro F1: 71. Result ID: 69e74a0f47f34ce924ce0185. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Reasoning":"w COT"}. This record does not assert independent reproduction. ### Claude-3.5-Sonnet Macro F1: 71. Result ID: 69e74a0f47f34ce924ce0187. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Reasoning":"w COT"}. This record does not assert independent reproduction. ### Claude-3.5-Sonnet Macro F1: 71. Result ID: 69e74a0f47f34ce924ce0188. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Reasoning":"wo COT"}. This record does not assert independent reproduction. ### MACE Macro F1: 71. Result ID: 69e74a0f47f34ce924ce018b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qw-72B","Configuration":"Pm + Em + Vm"}. This record does not assert independent reproduction. ### Qwen-72B Macro F1: 69. Result ID: 69e74a0f47f34ce924ce017e. [Reporting paper]() Recorded evaluation configuration: {"Reasoning":"wo COT"}. This record does not assert independent reproduction. ### Gemma-12B Macro F1: 68. Result ID: 69e74a0f47f34ce924ce0177. [Reporting paper]() Recorded evaluation configuration: {"Reasoning":"w COT"}. This record does not assert independent reproduction. ### Qwen-235B Macro F1: 68. Result ID: 69e74a0f47f34ce924ce0180. [Reporting paper]() Recorded evaluation configuration: {"Reasoning":"wo COT"}. This record does not assert independent reproduction. ### MACE Macro F1: 66. Result ID: 69e74a0f47f34ce924ce018a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Ll-8B","Configuration":"Pm + Em + Vm'"}. This record does not assert independent reproduction. ### Mistral-7B Macro F1: 65. Result ID: 69e74a0f47f34ce924ce0173. [Reporting paper]() Recorded evaluation configuration: {"Reasoning":"w COT"}. This record does not assert independent reproduction. ### Llama-70B Macro F1: 65. Result ID: 69e74a0f47f34ce924ce017a. [Reporting paper]() Recorded evaluation configuration: {"Reasoning":"wo COT"}. This record does not assert independent reproduction. ### Gemma-12B Macro F1: 64. Result ID: 69e74a0f47f34ce924ce0178. [Reporting paper]() Recorded evaluation configuration: {"Reasoning":"wo COT"}. This record does not assert independent reproduction. ### GPT-4o Macro F1: 60. Result ID: 69e74a0f47f34ce924ce0186. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Reasoning":"wo COT"}. This record does not assert independent reproduction. ### MACE Macro F1: 60. Result ID: 69e74a0f47f34ce924ce0189. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Mt-7B","Configuration":"Pm + Em + Vm'"}. This record does not assert independent reproduction. ### Llama-8B Macro F1: 59. Result ID: 69e74a0f47f34ce924ce0176. [Reporting paper]() Recorded evaluation configuration: {"Reasoning":"wo COT"}. This record does not assert independent reproduction. ### Mistral-7B Macro F1: 58. Result ID: 69e74a0f47f34ce924ce0174. [Reporting paper]() Recorded evaluation configuration: {"Reasoning":"wo COT"}. This record does not assert independent reproduction. ### Llama-3B Macro F1: 57. Result ID: 69e74a0f47f34ce924ce0171. [Reporting paper]() Recorded evaluation configuration: {"Reasoning":"w COT"}. This record does not assert independent reproduction. ### Llama-8B Macro F1: 56. Result ID: 69e74a0f47f34ce924ce0175. [Reporting paper]() Recorded evaluation configuration: {"Reasoning":"w COT"}. This record does not assert independent reproduction. ### DeepSeek-V3.1 Macro F1: 56. Result ID: 69e74a0f47f34ce924ce0184. [Reporting paper]() Recorded evaluation configuration: {"Reasoning":"wo COT"}. This record does not assert independent reproduction. ### Llama-3B Macro F1: 37. Result ID: 69e74a0f47f34ce924ce0172. [Reporting paper]() Recorded evaluation configuration: {"Reasoning":"wo COT"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.