# Natural Language Understanding on italic cot custom Source: [Natural Language Understanding on italic cot custom]() Collection index: [llms.txt]() Data updated: 2026-05-11T10:11:09.658Z Evaluation of Natural Language Understanding performance on the custom 'italic cot custom' dataset variant, measured by the 'flexible\_extract' metric. 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 | Natural Language Understanding | | Dataset (including any stated split) | italic cot custom | | Metric | Flexible Extract | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 16 of 16 | | 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 - [Flexible Extract](): Higher is better; unit: PERCENT. ## Recorded results | Method | Flexible Extract | Result date | Report paper | | --- | --- | --- | --- | | GPT-5 nano | 86.8 | Not recorded | 6a0162badbc836b87d82332a | | Ministral-3-8B-Instruct-2512-BF16 | 81.3 | Not recorded | 6a0162badbc836b87d82332a | | gemma-3-12b-it | 79 | Not recorded | 6a0162badbc836b87d82332a | | Qwen3-8B | 77.4 | Not recorded | 6a0162badbc836b87d82332a | | gpt-oss-20b | 76.9 | Not recorded | 6a0162badbc836b87d82332a | | Qwen3-4B | 71.4 | Not recorded | 6a0162badbc836b87d82332a | | EngGPT2-16B-A3B | 68.3 | Not recorded | 6a0162badbc836b87d82332a | | LLaMAntino-3-ANITA-8B-Inst-DPO-ITA | 67.3 | Not recorded | 6a0162badbc836b87d82332a | | Velvet-14B | 66.6 | Not recorded | 6a0162badbc836b87d82332a | | gemma-3-4b-it | 65.6 | Not recorded | 6a0162badbc836b87d82332a | | Llama-3.1-8B-Instruct | 65.3 | Not recorded | 6a0162badbc836b87d82332a | | FastwebMIIA-7B | 53.7 | Not recorded | 6a0162badbc836b87d82332a | | Moonlight-16B-A3B-Instruct | 53 | Not recorded | 6a0162badbc836b87d82332a | | Llama-3.2-3B-Instruct | 49.6 | Not recorded | 6a0162badbc836b87d82332a | | deepseek-moe-16b-chat | 36.8 | Not recorded | 6a0162badbc836b87d82332a | | Minerva-7B-instruct-v1.0 | 32.9 | Not recorded | 6a0162badbc836b87d82332a | ### GPT-5 nano Flexible Extract: 86.8. Result ID: 6a01ab3c47f34ce924e41eb1. [Reporting paper]() Recorded evaluation configuration: {"strategy":"Chain-of-Thought","configuration":"custom/structural improvements"}. This record does not assert independent reproduction. ### Ministral-3-8B-Instruct-2512-BF16 Flexible Extract: 81.3. Result ID: 6a01ab3c47f34ce924e41eb0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"strategy":"Chain-of-Thought","configuration":"custom/structural improvements"}. This record does not assert independent reproduction. ### gemma-3-12b-it Flexible Extract: 79. Result ID: 6a01ab3c47f34ce924e41eaf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"strategy":"Chain-of-Thought","configuration":"custom/structural improvements"}. This record does not assert independent reproduction. ### Qwen3-8B Flexible Extract: 77.4. Result ID: 6a01ab3c47f34ce924e41eab. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"strategy":"Chain-of-Thought","configuration":"custom/structural improvements"}. This record does not assert independent reproduction. ### gpt-oss-20b Flexible Extract: 76.9. Result ID: 6a01ab3c47f34ce924e41eb3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"strategy":"Chain-of-Thought","configuration":"custom/structural improvements"}. This record does not assert independent reproduction. ### Qwen3-4B Flexible Extract: 71.4. Result ID: 6a01ab3c47f34ce924e41eaa. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"strategy":"Chain-of-Thought","configuration":"custom/structural improvements"}. This record does not assert independent reproduction. ### EngGPT2-16B-A3B Flexible Extract: 68.3. Result ID: 6a01ab3c47f34ce924e41ea5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"strategy":"Chain-of-Thought","configuration":"custom/structural improvements"}. This record does not assert independent reproduction. ### LLaMAntino-3-ANITA-8B-Inst-DPO-ITA Flexible Extract: 67.3. Result ID: 6a01ab3c47f34ce924e41ea9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"strategy":"Chain-of-Thought","configuration":"custom/structural improvements"}. This record does not assert independent reproduction. ### Velvet-14B Flexible Extract: 66.6. Result ID: 6a01ab3c47f34ce924e41ea8. [Reporting paper]() Recorded evaluation configuration: {"strategy":"Chain-of-Thought","configuration":"custom/structural improvements"}. This record does not assert independent reproduction. ### gemma-3-4b-it Flexible Extract: 65.6. Result ID: 6a01ab3c47f34ce924e41eae. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"strategy":"Chain-of-Thought","configuration":"custom/structural improvements"}. This record does not assert independent reproduction. ### Llama-3.1-8B-Instruct Flexible Extract: 65.3. Result ID: 6a01ab3c47f34ce924e41ead. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"strategy":"Chain-of-Thought","configuration":"custom/structural improvements"}. This record does not assert independent reproduction. ### FastwebMIIA-7B Flexible Extract: 53.7. Result ID: 6a01ab3c47f34ce924e41ea6. [Reporting paper]() Recorded evaluation configuration: {"strategy":"Chain-of-Thought","configuration":"custom/structural improvements"}. This record does not assert independent reproduction. ### Moonlight-16B-A3B-Instruct Flexible Extract: 53. Result ID: 6a01ab3c47f34ce924e41eb4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"strategy":"Chain-of-Thought","configuration":"custom/structural improvements"}. This record does not assert independent reproduction. ### Llama-3.2-3B-Instruct Flexible Extract: 49.6. Result ID: 6a01ab3c47f34ce924e41eac. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"strategy":"Chain-of-Thought","configuration":"custom/structural improvements"}. This record does not assert independent reproduction. ### deepseek-moe-16b-chat Flexible Extract: 36.8. Result ID: 6a01ab3c47f34ce924e41eb2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"strategy":"Chain-of-Thought","configuration":"custom/structural improvements"}. This record does not assert independent reproduction. ### Minerva-7B-instruct-v1.0 Flexible Extract: 32.9. Result ID: 6a01ab3c47f34ce924e41ea7. [Reporting paper]() Recorded evaluation configuration: {"strategy":"Chain-of-Thought","configuration":"custom/structural improvements"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.