# Writing flaw detection on BenchMarker Writing - Out of Domain, Human Source: [Writing flaw detection on BenchMarker Writing - Out of Domain, Human]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:30:51.008Z Evaluation of models on detecting writing flaws within the Human subset of the BenchMarker Writing Out-of-Domain dataset, measured by Accuracy, F1 Score, and Cohen's Kappa. 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 | Writing flaw detection | | Dataset (including any stated split) | BenchMarker Writing - Out of Domain, Human | | 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. - [F1 Score](): Higher is better; unit: PERCENT. - [Cohen's Kappa](): Higher is better; unit: SCALAR. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Gemma-3 | 92.6 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | SAQUET | 92.6 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | Always Not Flawed | 92.5 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | Gemma-3 | 92.3 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | Qwen-3 | 90.8 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | Qwen-3 | 90.2 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | Command R | 88.5 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | Command R+ | 88.1 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | GPT-5 | 87.8 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | Qwen-3 | 87.6 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | GPT-5 | 87.1 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | Gemini 2.5 | 85.9 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | LLaMA-3.1 | 85.9 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | Gemini 2.5 | 85.3 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | GPT-5 | 84.1 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | Claude 4.5 | 83.1 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | LLaMA-3.2 | 81.4 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | Qwen-3 | 80.1 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | Qwen-3 | 79.9 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | Qwen-3 | 79.3 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | Claude 4.5 | 76.1 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | Gemma-3 | 73.6 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | LLaMA-3.1 | 72.7 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | Gemini 2.5 | 71.3 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | Random (50/50) | 50 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | LLaMA-3.2 | 42.3 | Not recorded | 69896a28c97a0b3e3b5dc5ac | | Always Flawed | 7.5 | Not recorded | 69896a28c97a0b3e3b5dc5ac | ### Gemma-3 Accuracy: 92.6. Result ID: 699646735698665ec48a5641. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"27B"}. This record does not assert independent reproduction. ### SAQUET Accuracy: 92.6. Result ID: 699646735698665ec48a5646. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Always Not Flawed Accuracy: 92.5. Result ID: 699646735698665ec48a5648. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Gemma-3 Accuracy: 92.3. Result ID: 699646735698665ec48a5640. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"12B"}. This record does not assert independent reproduction. ### Qwen-3 Accuracy: 90.8. Result ID: 699646735698665ec48a563a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"1.7B"}. This record does not assert independent reproduction. ### Qwen-3 Accuracy: 90.2. Result ID: 699646735698665ec48a5639. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"0.6B"}. This record does not assert independent reproduction. ### Command R Accuracy: 88.5. Result ID: 699646735698665ec48a5637. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Command R+ Accuracy: 88.1. Result ID: 699646735698665ec48a5638. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GPT-5 Accuracy: 87.8. Result ID: 699646735698665ec48a5632. [Reporting paper]() Recorded evaluation configuration: {"Model Variant":"Nano"}. This record does not assert independent reproduction. ### Qwen-3 Accuracy: 87.6. Result ID: 699646735698665ec48a563b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"4B"}. This record does not assert independent reproduction. ### GPT-5 Accuracy: 87.1. Result ID: 699646735698665ec48a5634. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Gemini 2.5 Accuracy: 85.9. Result ID: 699646735698665ec48a5631. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Variant":"Pro"}. This record does not assert independent reproduction. ### LLaMA-3.1 Accuracy: 85.9. Result ID: 699646735698665ec48a5644. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Parameters":"8B"}. This record does not assert independent reproduction. ### Gemini 2.5 Accuracy: 85.3. Result ID: 699646735698665ec48a5630. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Variant":"Flash"}. This record does not assert independent reproduction. ### GPT-5 Accuracy: 84.1. Result ID: 699646735698665ec48a5633. [Reporting paper]() Recorded evaluation configuration: {"Model Variant":"Mini"}. This record does not assert independent reproduction. ### Claude 4.5 Accuracy: 83.1. Result ID: 699646735698665ec48a5636. [Reporting paper]() Recorded evaluation configuration: {"Model Variant":"Sonnet"}. This record does not assert independent reproduction. ### LLaMA-3.2 Accuracy: 81.4. Result ID: 699646735698665ec48a5643. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Parameters":"3B"}. This record does not assert independent reproduction. ### Qwen-3 Accuracy: 80.1. Result ID: 699646735698665ec48a563e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"32B"}. This record does not assert independent reproduction. ### Qwen-3 Accuracy: 79.9. Result ID: 699646735698665ec48a563d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"14B"}. This record does not assert independent reproduction. ### Qwen-3 Accuracy: 79.3. Result ID: 699646735698665ec48a563c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"8B"}. This record does not assert independent reproduction. ### Claude 4.5 Accuracy: 76.1. Result ID: 699646735698665ec48a5635. [Reporting paper]() Recorded evaluation configuration: {"Model Variant":"Haiku"}. This record does not assert independent reproduction. ### Gemma-3 Accuracy: 73.6. Result ID: 699646735698665ec48a563f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"4B"}. This record does not assert independent reproduction. ### LLaMA-3.1 Accuracy: 72.7. Result ID: 699646735698665ec48a5645. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Parameters":"70B"}. This record does not assert independent reproduction. ### Gemini 2.5 Accuracy: 71.3. Result ID: 699646735698665ec48a562f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Variant":"Lite"}. This record does not assert independent reproduction. ### Random (50/50) Accuracy: 50. Result ID: 699646735698665ec48a5647. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### LLaMA-3.2 Accuracy: 42.3. Result ID: 699646735698665ec48a5642. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Parameters":"1B"}. This record does not assert independent reproduction. ### Always Flawed Accuracy: 7.5. Result ID: 699646735698665ec48a5649. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.