# Visual Reasoning on VisualProbe Easy Source: [Visual Reasoning on VisualProbe Easy]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:32:48.905Z Evaluation of model performance on the easy subset of the VisualProbe dataset for Visual Reasoning capabilities, reporting ACC and TSR. 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 | Visual Reasoning | | Dataset (including any stated split) | VisualProbe Easy | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 18 of 18 | | 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. - [TSR](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | RTWI | 65.4 | Not recorded | 6992a4749ebc499c97406be0 | | SC | 64 | Not recorded | 6992a4749ebc499c97406be0 | | ASC | 64 | Not recorded | 6992a4749ebc499c97406be0 | | ESC | 64 | Not recorded | 6992a4749ebc499c97406be0 | | Deepconf | 61.9 | Not recorded | 6992a4749ebc499c97406be0 | | RTWI | 60.4 | Not recorded | 6992a4749ebc499c97406be0 | | DeepEyes | 60.1 | Not recorded | 6992a4749ebc499c97406be0 | | CISC | 59.7 | Not recorded | 6992a4749ebc499c97406be0 | | Self-Cer. | 59 | Not recorded | 6992a4749ebc499c97406be0 | | SC | 56.8 | Not recorded | 6992a4749ebc499c97406be0 | | ASC | 56.8 | Not recorded | 6992a4749ebc499c97406be0 | | ESC | 56.8 | Not recorded | 6992a4749ebc499c97406be0 | | Deepconf | 56.8 | Not recorded | 6992a4749ebc499c97406be0 | | Base | 56 | Not recorded | 6992a4749ebc499c97406be0 | | Self-Cer. | 54 | Not recorded | 6992a4749ebc499c97406be0 | | CISC | 52.5 | Not recorded | 6992a4749ebc499c97406be0 | | Base | 47.6 | Not recorded | 6992a4749ebc499c97406be0 | | GPT-4o | 47.5 | Not recorded | 6992a4749ebc499c97406be0 | ### RTWI Accuracy: 65.4. Result ID: 699a5fad5698665ec4e79e09. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-VL Instruct"}. This record does not assert independent reproduction. ### SC Accuracy: 64. Result ID: 699a5fad5698665ec4e79e03. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-VL Instruct"}. This record does not assert independent reproduction. ### ASC Accuracy: 64. Result ID: 699a5fad5698665ec4e79e04. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-VL Instruct"}. This record does not assert independent reproduction. ### ESC Accuracy: 64. Result ID: 699a5fad5698665ec4e79e05. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-VL Instruct"}. This record does not assert independent reproduction. ### Deepconf Accuracy: 61.9. Result ID: 699a5fad5698665ec4e79e07. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-VL Instruct"}. This record does not assert independent reproduction. ### RTWI Accuracy: 60.4. Result ID: 699a5fad5698665ec4e79e01. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-VL Thinking"}. This record does not assert independent reproduction. ### DeepEyes Accuracy: 60.1. Result ID: 699a5fad5698665ec4e79df9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepEyes-7B"}. This record does not assert independent reproduction. ### CISC Accuracy: 59.7. Result ID: 699a5fad5698665ec4e79e06. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-VL Instruct"}. This record does not assert independent reproduction. ### Self-Cer. Accuracy: 59. Result ID: 699a5fad5698665ec4e79e08. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen3-VL Instruct"}. This record does not assert independent reproduction. ### SC Accuracy: 56.8. Result ID: 699a5fad5698665ec4e79dfb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-VL Thinking"}. This record does not assert independent reproduction. ### ASC Accuracy: 56.8. Result ID: 699a5fad5698665ec4e79dfc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-VL Thinking"}. This record does not assert independent reproduction. ### ESC Accuracy: 56.8. Result ID: 699a5fad5698665ec4e79dfd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-VL Thinking"}. This record does not assert independent reproduction. ### Deepconf Accuracy: 56.8. Result ID: 699a5fad5698665ec4e79dff. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-VL Thinking"}. This record does not assert independent reproduction. ### Base Accuracy: 56. Result ID: 699a5fad5698665ec4e79e02. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-VL Instruct"}. This record does not assert independent reproduction. ### Self-Cer. Accuracy: 54. Result ID: 699a5fad5698665ec4e79e00. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen3-VL Thinking"}. This record does not assert independent reproduction. ### CISC Accuracy: 52.5. Result ID: 699a5fad5698665ec4e79dfe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-VL Thinking"}. This record does not assert independent reproduction. ### Base Accuracy: 47.6. Result ID: 699a5fad5698665ec4e79dfa. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-VL Thinking"}. This record does not assert independent reproduction. ### GPT-4o Accuracy: 47.5. Result ID: 699a5fad5698665ec4e79df8. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"GPT-4o"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.