# Spatial Reasoning on MMSI-Bench MindJourney Subset (162 questions) (test) Source: [Spatial Reasoning on MMSI-Bench MindJourney Subset (162 questions) (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:31:58.517Z Evaluation of model performance on the specific 162-question MindJourney subset of the MMSI-Bench test split, focusing on detailed Spatial Reasoning capabilities across positional relationships, attributes, and motion. 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 | Spatial Reasoning | | Dataset (including any stated split) | MMSI-Bench MindJourney Subset (162 questions) (test) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 19 of 19 | | 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. - [Positional Relationship (Cam-Cam)](): Higher is better; unit: PERCENT. - [Positional Relationship (Obj-Obj)](): Higher is better; unit: PERCENT. - [Positional Relationship (Reg-Reg)](): Higher is better; unit: PERCENT. - [Positional Relationship (Cam-Obj)](): Higher is better; unit: PERCENT. - [Positional Relationship (Obj-Reg)](): Higher is better; unit: PERCENT. - [Positional Relationship (Cam-Reg)](): Higher is better; unit: PERCENT. - [Attribute (Measurable)](): Higher is better; unit: PERCENT. - [Attribute (Appearance)](): Higher is better; unit: PERCENT. - [Motion (Camera)](): Higher is better; unit: PERCENT. - [Motion (Object)](): Higher is better; unit: PERCENT. - [Multi-Step Reasoning (MSR)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | ViSA | 0.358 | Not recorded | 69365b78d6b0dd3342934b9c | | MindJourney | 0.3395 | Not recorded | 69365b78d6b0dd3342934b9c | | Random | 0.3333 | Not recorded | 69365b78d6b0dd3342934b9c | | Random | 0.3333 | Not recorded | 69365b78d6b0dd3342934b9c | | Random | 0.3333 | Not recorded | 69365b78d6b0dd3342934b9c | | MindJourney | 0.3272 | Not recorded | 69365b78d6b0dd3342934b9c | | ViSA | 0.3272 | Not recorded | 69365b78d6b0dd3342934b9c | | MindJourney | 0.3272 | Not recorded | 69365b78d6b0dd3342934b9c | | MindJourney | 0.3272 | Not recorded | 69365b78d6b0dd3342934b9c | | ViSA | 0.321 | Not recorded | 69365b78d6b0dd3342934b9c | | ViSA | 0.3148 | Not recorded | 69365b78d6b0dd3342934b9c | | Random | 0.3025 | Not recorded | 69365b78d6b0dd3342934b9c | | Random | 0.2963 | Not recorded | 69365b78d6b0dd3342934b9c | | MindJourney | 0.2963 | Not recorded | 69365b78d6b0dd3342934b9c | | InternVL3-14B | 0.2716 | Not recorded | 69365b78d6b0dd3342934b9c | | MindJourney | 0.2716 | Not recorded | 69365b78d6b0dd3342934b9c | | Random | 0.2716 | Not recorded | 69365b78d6b0dd3342934b9c | | ViSA | 0.2654 | Not recorded | 69365b78d6b0dd3342934b9c | | ViSA | 0.2593 | Not recorded | 69365b78d6b0dd3342934b9c | ### ViSA Accuracy: 0.358. Result ID: 699886fb5698665ec4e866f9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Beam depth (γ)":1,"top-k":1}. This record does not assert independent reproduction. ### MindJourney Accuracy: 0.3395. Result ID: 699886fb5698665ec4e86700. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Beam depth (γ)":2,"top-k":2}. This record does not assert independent reproduction. ### Random Accuracy: 0.3333. Result ID: 699886fb5698665ec4e866f3. [Reporting paper]() Recorded evaluation configuration: {"Beam depth (γ)":1,"top-k":1}. This record does not assert independent reproduction. ### Random Accuracy: 0.3333. Result ID: 699886fb5698665ec4e866f5. [Reporting paper]() Recorded evaluation configuration: {"Beam depth (γ)":1,"top-k":3}. This record does not assert independent reproduction. ### Random Accuracy: 0.3333. Result ID: 699886fb5698665ec4e866fd. [Reporting paper]() Recorded evaluation configuration: {"Beam depth (γ)":2,"top-k":2}. This record does not assert independent reproduction. ### MindJourney Accuracy: 0.3272. Result ID: 699886fb5698665ec4e866f6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Beam depth (γ)":1,"top-k":1}. This record does not assert independent reproduction. ### ViSA Accuracy: 0.3272. Result ID: 699886fb5698665ec4e866fb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Beam depth (γ)":1,"top-k":3}. This record does not assert independent reproduction. ### MindJourney Accuracy: 0.3272. Result ID: 699886fb5698665ec4e866ff. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Beam depth (γ)":2,"top-k":1}. This record does not assert independent reproduction. ### MindJourney Accuracy: 0.3272. Result ID: 699886fb5698665ec4e86701. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Beam depth (γ)":2,"top-k":3}. This record does not assert independent reproduction. ### ViSA Accuracy: 0.321. Result ID: 699886fb5698665ec4e86702. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Beam depth (γ)":2,"top-k":1}. This record does not assert independent reproduction. ### ViSA Accuracy: 0.3148. Result ID: 699886fb5698665ec4e86703. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Beam depth (γ)":2,"top-k":2}. This record does not assert independent reproduction. ### Random Accuracy: 0.3025. Result ID: 699886fb5698665ec4e866fc. [Reporting paper]() Recorded evaluation configuration: {"Beam depth (γ)":2,"top-k":1}. This record does not assert independent reproduction. ### Random Accuracy: 0.2963. Result ID: 699886fb5698665ec4e866f4. [Reporting paper]() Recorded evaluation configuration: {"Beam depth (γ)":1,"top-k":2}. This record does not assert independent reproduction. ### MindJourney Accuracy: 0.2963. Result ID: 699886fb5698665ec4e866f8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Beam depth (γ)":1,"top-k":3}. This record does not assert independent reproduction. ### InternVL3-14B Accuracy: 0.2716. Result ID: 699886fb5698665ec4e866f2. [Reporting paper]() Recorded evaluation configuration: {"mode":"baseline"}. This record does not assert independent reproduction. ### MindJourney Accuracy: 0.2716. Result ID: 699886fb5698665ec4e866f7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Beam depth (γ)":1,"top-k":2}. This record does not assert independent reproduction. ### Random Accuracy: 0.2716. Result ID: 699886fb5698665ec4e866fe. [Reporting paper]() Recorded evaluation configuration: {"Beam depth (γ)":2,"top-k":3}. This record does not assert independent reproduction. ### ViSA Accuracy: 0.2654. Result ID: 699886fb5698665ec4e86704. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Beam depth (γ)":2,"top-k":3}. This record does not assert independent reproduction. ### ViSA Accuracy: 0.2593. Result ID: 699886fb5698665ec4e866fa. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Beam depth (γ)":1,"top-k":2}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.