# Painting Grading on HanMo-Bench Source: [Painting Grading on HanMo-Bench]() Collection index: [llms.txt]() Data updated: 2026-03-12T09:14:11.785Z Evaluation of model performance on grading paintings using the HanMo-Bench dataset, measured by metrics such as MAE, RMSE, Accuracy, and BERTScore. 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 | Painting Grading | | Dataset (including any stated split) | HanMo-Bench | | Metric | MAE | | Unit | SCALAR | | Direction | Lower is better | | Results loaded | 25 of 25 | | 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 - [MAE](): Lower is better; unit: SCALAR. - [RMSE](): Lower is better; unit: SCALAR. - [Accuracy](): Higher is better; unit: PERCENT. - [BERTScore](): Higher is better; unit: SCALAR. ## Recorded results | Method | MAE | Result date | Report paper | | --- | --- | --- | --- | | HanMoVLM (RFT) | 0.631 | Not recorded | 69b24889d030084c22e2f895 | | HanMoVLM (SFT) | 0.645 | Not recorded | 69b24889d030084c22e2f895 | | HanMoVLM (RFT) | 0.6632 | Not recorded | 69b24889d030084c22e2f895 | | HanMoVLM (SFT) | 0.685 | Not recorded | 69b24889d030084c22e2f895 | | Gemini-3-Pro | 0.9416 | Not recorded | 69b24889d030084c22e2f895 | | Gemini-3-Flash | 0.9466 | Not recorded | 69b24889d030084c22e2f895 | | Gemini-3-Flash | 0.9633 | Not recorded | 69b24889d030084c22e2f895 | | Gemini-3-Pro | 0.9783 | Not recorded | 69b24889d030084c22e2f895 | | Gemini-3-Pro | 0.9866 | Not recorded | 69b24889d030084c22e2f895 | | Gemini-3-Flash | 0.9883 | Not recorded | 69b24889d030084c22e2f895 | | Qwen3-VL-Instruct | 1.505 | Not recorded | 69b24889d030084c22e2f895 | | Qwen3-VL-Instruct | 1.5124 | Not recorded | 69b24889d030084c22e2f895 | | Qwen3-VL-Instruct | 1.5582 | Not recorded | 69b24889d030084c22e2f895 | | GPT-4 | 1.7491 | Not recorded | 69b24889d030084c22e2f895 | | GPT-5.1 | 1.7671 | Not recorded | 69b24889d030084c22e2f895 | | GPT-5.1 | 1.7742 | Not recorded | 69b24889d030084c22e2f895 | | GPT-4 | 1.8023 | Not recorded | 69b24889d030084c22e2f895 | | GPT-5.1 | 1.8204 | Not recorded | 69b24889d030084c22e2f895 | | InternVL-3.5 | 1.8866 | Not recorded | 69b24889d030084c22e2f895 | | Qwen3-VL-Instruct | 1.9341 | Not recorded | 69b24889d030084c22e2f895 | | Qwen3-VL-Instruct | 1.9493 | Not recorded | 69b24889d030084c22e2f895 | | InternVL-3.5 | 1.9591 | Not recorded | 69b24889d030084c22e2f895 | | Qwen3-VL-Instruct | 1.9734 | Not recorded | 69b24889d030084c22e2f895 | | GLM4.6V-Flash | 2.0495 | Not recorded | 69b24889d030084c22e2f895 | | STEP3-VL | 2.295 | Not recorded | 69b24889d030084c22e2f895 | ### HanMoVLM (RFT) MAE: 0.631. Result ID: 69b2832d6dbb7075b95d5e35. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"8B"}. This record does not assert independent reproduction. ### HanMoVLM (SFT) MAE: 0.645. Result ID: 69b2832d6dbb7075b95d5e34. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"8B"}. This record does not assert independent reproduction. ### HanMoVLM (RFT) MAE: 0.6632. Result ID: 69b2832d6dbb7075b95d5e2d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"4B"}. This record does not assert independent reproduction. ### HanMoVLM (SFT) MAE: 0.685. Result ID: 69b2832d6dbb7075b95d5e2c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"4B"}. This record does not assert independent reproduction. ### Gemini-3-Pro MAE: 0.9416. Result ID: 69b2832d6dbb7075b95d5e22. [Reporting paper]() Recorded evaluation configuration: {"Size":"-","Expert CoT":"true"}. This record does not assert independent reproduction. ### Gemini-3-Flash MAE: 0.9466. Result ID: 69b2832d6dbb7075b95d5e1f. [Reporting paper]() Recorded evaluation configuration: {"Size":"-","Expert CoT":"true"}. This record does not assert independent reproduction. ### Gemini-3-Flash MAE: 0.9633. Result ID: 69b2832d6dbb7075b95d5e1e. [Reporting paper]() Recorded evaluation configuration: {"Size":"-","Thinking":"true"}. This record does not assert independent reproduction. ### Gemini-3-Pro MAE: 0.9783. Result ID: 69b2832d6dbb7075b95d5e21. [Reporting paper]() Recorded evaluation configuration: {"Size":"-","Thinking":"true"}. This record does not assert independent reproduction. ### Gemini-3-Pro MAE: 0.9866. Result ID: 69b2832d6dbb7075b95d5e20. [Reporting paper]() Recorded evaluation configuration: {"Size":"-"}. This record does not assert independent reproduction. ### Gemini-3-Flash MAE: 0.9883. Result ID: 69b2832d6dbb7075b95d5e1d. [Reporting paper]() Recorded evaluation configuration: {"Size":"-"}. This record does not assert independent reproduction. ### Qwen3-VL-Instruct MAE: 1.505. Result ID: 69b2832d6dbb7075b95d5e32. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"8B","Thinking":"true"}. This record does not assert independent reproduction. ### Qwen3-VL-Instruct MAE: 1.5124. Result ID: 69b2832d6dbb7075b95d5e33. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"8B","Expert CoT":"true"}. This record does not assert independent reproduction. ### Qwen3-VL-Instruct MAE: 1.5582. Result ID: 69b2832d6dbb7075b95d5e31. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"8B"}. This record does not assert independent reproduction. ### GPT-4 MAE: 1.7491. Result ID: 69b2832d6dbb7075b95d5e24. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Size":"-","Expert CoT":"true"}. This record does not assert independent reproduction. ### GPT-5.1 MAE: 1.7671. Result ID: 69b2832d6dbb7075b95d5e27. [Reporting paper]() Recorded evaluation configuration: {"Size":"-","Expert CoT":"true"}. This record does not assert independent reproduction. ### GPT-5.1 MAE: 1.7742. Result ID: 69b2832d6dbb7075b95d5e25. [Reporting paper]() Recorded evaluation configuration: {"Size":"-"}. This record does not assert independent reproduction. ### GPT-4 MAE: 1.8023. Result ID: 69b2832d6dbb7075b95d5e23. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Size":"-"}. This record does not assert independent reproduction. ### GPT-5.1 MAE: 1.8204. Result ID: 69b2832d6dbb7075b95d5e26. [Reporting paper]() Recorded evaluation configuration: {"Size":"-","Thinking":"true"}. This record does not assert independent reproduction. ### InternVL-3.5 MAE: 1.8866. Result ID: 69b2832d6dbb7075b95d5e30. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"8B"}. This record does not assert independent reproduction. ### Qwen3-VL-Instruct MAE: 1.9341. Result ID: 69b2832d6dbb7075b95d5e2b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"4B","Expert CoT":"true"}. This record does not assert independent reproduction. ### Qwen3-VL-Instruct MAE: 1.9493. Result ID: 69b2832d6dbb7075b95d5e2a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"4B","Thinking":"true"}. This record does not assert independent reproduction. ### InternVL-3.5 MAE: 1.9591. Result ID: 69b2832d6dbb7075b95d5e28. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"4B"}. This record does not assert independent reproduction. ### Qwen3-VL-Instruct MAE: 1.9734. Result ID: 69b2832d6dbb7075b95d5e29. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"4B"}. This record does not assert independent reproduction. ### GLM4.6V-Flash MAE: 2.0495. Result ID: 69b2832d6dbb7075b95d5e2f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"9B"}. This record does not assert independent reproduction. ### STEP3-VL MAE: 2.295. Result ID: 69b2832d6dbb7075b95d5e2e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Size":"10B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.