# Common-sense Reasoning on BBH Source: [Common-sense Reasoning on BBH]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:31:01.933Z Evaluation of model performance specifically on the common-sense reasoning subset of tasks within the Big-Bench Hard (BBH) benchmark suite, typically measured by 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 | Common-sense Reasoning | | Dataset (including any stated split) | BBH | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 27 of 27 | | 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 | | --- | --- | --- | --- | | ReMiT | 58.27 | Not recorded | 6982d2e5a8b01e76cc8e98a7 | | MiniPLM | 57.1 | Not recorded | 6982d2e5a8b01e76cc8e98a7 | | Vanilla NTP | 56.5 | Not recorded | 6982d2e5a8b01e76cc8e98a7 | | RHO-1 | 55.32 | Not recorded | 6982d2e5a8b01e76cc8e98a7 | | ReMiT | 47.21 | Not recorded | 6982d2e5a8b01e76cc8e98a7 | | SFT | 45.56 | Not recorded | 696722efe25ad2fa9d1ebfc4 | | T-SPIN | 45.05 | Not recorded | 696722efe25ad2fa9d1ebfc4 | | T-SPIN | 44.97 | Not recorded | 696722efe25ad2fa9d1ebfc4 | | T-SPIN | 44.91 | Not recorded | 696722efe25ad2fa9d1ebfc4 | | T-SPIN | 44.9 | Not recorded | 696722efe25ad2fa9d1ebfc4 | | SPIN | 44.56 | Not recorded | 696722efe25ad2fa9d1ebfc4 | | Vanilla NTP | 44.37 | Not recorded | 6982d2e5a8b01e76cc8e98a7 | | Mistral-7B | 44.26 | Not recorded | 696722efe25ad2fa9d1ebfc4 | | MiniPLM | 44.2 | Not recorded | 6982d2e5a8b01e76cc8e98a7 | | T-SPIN | 44.06 | Not recorded | 696722efe25ad2fa9d1ebfc4 | | Pre-Trained | 43.39 | Not recorded | 6982d2e5a8b01e76cc8e98a7 | | Pre-Trained | 43.13 | Not recorded | 6982d2e5a8b01e76cc8e98a7 | | SPIN | 42.88 | Not recorded | 696722efe25ad2fa9d1ebfc4 | | SPIN | 42.5 | Not recorded | 696722efe25ad2fa9d1ebfc4 | | SPIN | 41.97 | Not recorded | 696722efe25ad2fa9d1ebfc4 | | SPIN | 41.4 | Not recorded | 696722efe25ad2fa9d1ebfc4 | | RHO-1 | 40.45 | Not recorded | 6982d2e5a8b01e76cc8e98a7 | | ReMiT | 32.07 | Not recorded | 6982d2e5a8b01e76cc8e98a7 | | Vanilla NTP | 30.87 | Not recorded | 6982d2e5a8b01e76cc8e98a7 | | MiniPLM | 30.38 | Not recorded | 6982d2e5a8b01e76cc8e98a7 | | RHO-1 | 29.33 | Not recorded | 6982d2e5a8b01e76cc8e98a7 | | Pre-Trained | 28.43 | Not recorded | 6982d2e5a8b01e76cc8e98a7 | ### ReMiT Accuracy: 58.27. Result ID: 699651375698665ec49611d4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Family":"SmolLM3-3B","Mid-Training":"True","Few-shot":"True"}. This record does not assert independent reproduction. ### MiniPLM Accuracy: 57.1. Result ID: 699651375698665ec49611d2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Family":"SmolLM3-3B","Mid-Training":"True","Few-shot":"True"}. This record does not assert independent reproduction. ### Vanilla NTP Accuracy: 56.5. Result ID: 699651375698665ec49611d1. [Reporting paper]() Recorded evaluation configuration: {"Model Family":"SmolLM3-3B","Mid-Training":"True","Few-shot":"True"}. This record does not assert independent reproduction. ### RHO-1 Accuracy: 55.32. Result ID: 699651375698665ec49611d3. [Reporting paper]() Recorded evaluation configuration: {"Model Family":"SmolLM3-3B","Mid-Training":"True","Few-shot":"True"}. This record does not assert independent reproduction. ### ReMiT Accuracy: 47.21. Result ID: 699651375698665ec49611d9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Family":"Youtu-LLM-2B","Mid-Training":"True","Few-shot":"True"}. This record does not assert independent reproduction. ### SFT Accuracy: 45.56. Result ID: 69968a7b5698665ec4d3114e. [Reporting paper]() Recorded evaluation configuration: {"Method":"Supervised Fine-Tuning"}. This record does not assert independent reproduction. ### T-SPIN Accuracy: 45.05. Result ID: 69968a7b5698665ec4d31156. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Iteration":"Iter2"}. This record does not assert independent reproduction. ### T-SPIN Accuracy: 44.97. Result ID: 69968a7b5698665ec4d31157. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Iteration":"Iter3"}. This record does not assert independent reproduction. ### T-SPIN Accuracy: 44.91. Result ID: 69968a7b5698665ec4d31155. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Iteration":"Iter1"}. This record does not assert independent reproduction. ### T-SPIN Accuracy: 44.9. Result ID: 69968a7b5698665ec4d31158. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Iteration":"Iter4"}. This record does not assert independent reproduction. ### SPIN Accuracy: 44.56. Result ID: 69968a7b5698665ec4d3114f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Iteration":"Iter0"}. This record does not assert independent reproduction. ### Vanilla NTP Accuracy: 44.37. Result ID: 699651375698665ec49611d6. [Reporting paper]() Recorded evaluation configuration: {"Model Family":"Youtu-LLM-2B","Mid-Training":"True","Few-shot":"True"}. This record does not assert independent reproduction. ### Mistral-7B Accuracy: 44.26. Result ID: 69968a7b5698665ec4d3114d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Configuration":"Base"}. This record does not assert independent reproduction. ### MiniPLM Accuracy: 44.2. Result ID: 699651375698665ec49611d7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Family":"Youtu-LLM-2B","Mid-Training":"True","Few-shot":"True"}. This record does not assert independent reproduction. ### T-SPIN Accuracy: 44.06. Result ID: 69968a7b5698665ec4d31154. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Iteration":"Iter0"}. This record does not assert independent reproduction. ### Pre-Trained Accuracy: 43.39. Result ID: 699651375698665ec49611d5. [Reporting paper]() Recorded evaluation configuration: {"Model Family":"Youtu-LLM-2B","Mid-Training":"Standard Checkpoint","Few-shot":"True"}. This record does not assert independent reproduction. ### Pre-Trained Accuracy: 43.13. Result ID: 699651375698665ec49611d0. [Reporting paper]() Recorded evaluation configuration: {"Model Family":"SmolLM3-3B","Mid-Training":"Standard Checkpoint","Few-shot":"True"}. This record does not assert independent reproduction. ### SPIN Accuracy: 42.88. Result ID: 69968a7b5698665ec4d31153. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Iteration":"Iter4"}. This record does not assert independent reproduction. ### SPIN Accuracy: 42.5. Result ID: 69968a7b5698665ec4d31150. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Iteration":"Iter1"}. This record does not assert independent reproduction. ### SPIN Accuracy: 41.97. Result ID: 69968a7b5698665ec4d31151. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Iteration":"Iter2"}. This record does not assert independent reproduction. ### SPIN Accuracy: 41.4. Result ID: 69968a7b5698665ec4d31152. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Iteration":"Iter3"}. This record does not assert independent reproduction. ### RHO-1 Accuracy: 40.45. Result ID: 699651375698665ec49611d8. [Reporting paper]() Recorded evaluation configuration: {"Model Family":"Youtu-LLM-2B","Mid-Training":"True","Few-shot":"True"}. This record does not assert independent reproduction. ### ReMiT Accuracy: 32.07. Result ID: 699651375698665ec49611cf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Family":"OLMo-1B","Mid-Training":"True","Few-shot":"True"}. This record does not assert independent reproduction. ### Vanilla NTP Accuracy: 30.87. Result ID: 699651375698665ec49611cc. [Reporting paper]() Recorded evaluation configuration: {"Model Family":"OLMo-1B","Mid-Training":"True","Few-shot":"True"}. This record does not assert independent reproduction. ### MiniPLM Accuracy: 30.38. Result ID: 699651375698665ec49611cd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Family":"OLMo-1B","Mid-Training":"True","Few-shot":"True"}. This record does not assert independent reproduction. ### RHO-1 Accuracy: 29.33. Result ID: 699651375698665ec49611ce. [Reporting paper]() Recorded evaluation configuration: {"Model Family":"OLMo-1B","Mid-Training":"True","Few-shot":"True"}. This record does not assert independent reproduction. ### Pre-Trained Accuracy: 28.43. Result ID: 699651375698665ec49611cb. [Reporting paper]() Recorded evaluation configuration: {"Model Family":"OLMo-1B","Mid-Training":"Standard Checkpoint","Few-shot":"True"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.