# Multiple Choice Answering on VIEW2SPACE v1 Source: [Multiple Choice Answering on VIEW2SPACE v1]() Collection index: [llms.txt]() Data updated: 2026-03-18T08:52:00.276Z Evaluation of model performance on the VIEW2SPACE v1 dataset for the Multiple Choice Answering task, typically measured using Accuracy. 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 | Multiple Choice Answering | | Dataset (including any stated split) | VIEW2SPACE v1 | | 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. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Ours (Grounded CoT) | 64.93 | Not recorded | 69ba318d0636e5bf4afc0831 | | GPT-5 | 59.86 | Not recorded | 69ba318d0636e5bf4afc0831 | | Ours (CoT) | 59.57 | Not recorded | 69ba318d0636e5bf4afc0831 | | GPT-5-mini | 49 | Not recorded | 69ba318d0636e5bf4afc0831 | | Qwen3-VL-8B | 37.41 | Not recorded | 69ba318d0636e5bf4afc0831 | | GPT-5.2 | 37.07 | Not recorded | 69ba318d0636e5bf4afc0831 | | Ours (Instruct-tuning) | 36.5 | Not recorded | 69ba318d0636e5bf4afc0831 | | Qwen3-VL-4B | 35.19 | Not recorded | 69ba318d0636e5bf4afc0831 | | Molmo2-8B | 34.57 | Not recorded | 69ba318d0636e5bf4afc0831 | | Ours (Grounded CoT, Direct Answer at Test) | 33.95 | Not recorded | 69ba318d0636e5bf4afc0831 | | GPT-4o | 33.93 | Not recorded | 69ba318d0636e5bf4afc0831 | | GPT-5-nano | 33.36 | Not recorded | 69ba318d0636e5bf4afc0831 | | Qwen3-VL-2B | 31.43 | Not recorded | 69ba318d0636e5bf4afc0831 | | Molmo2-4B | 31.36 | Not recorded | 69ba318d0636e5bf4afc0831 | | Gemma-3-12B-it | 31.14 | Not recorded | 69ba318d0636e5bf4afc0831 | | Internvl2.5-8B | 30.93 | Not recorded | 69ba318d0636e5bf4afc0831 | | RoBobrain2.0 | 30.79 | Not recorded | 69ba318d0636e5bf4afc0831 | | Openview | 30.72 | Not recorded | 69ba318d0636e5bf4afc0831 | | Qwen2.5-VL-7B | 30.29 | Not recorded | 69ba318d0636e5bf4afc0831 | | MINDCUBE-Plain | 30.21 | Not recorded | 69ba318d0636e5bf4afc0831 | | Mantis-8B (SigLip) | 30.07 | Not recorded | 69ba318d0636e5bf4afc0831 | | DeepSeek-VL2-Small | 29.43 | Not recorded | 69ba318d0636e5bf4afc0831 | | MINDCUBE-Aug | 29 | Not recorded | 69ba318d0636e5bf4afc0831 | | Random (chance) | 28.59 | Not recorded | 69ba318d0636e5bf4afc0831 | | Random (frequency) | 28.59 | Not recorded | 69ba318d0636e5bf4afc0831 | | Spatial-MLLM | 27.93 | Not recorded | 69ba318d0636e5bf4afc0831 | | Idefics2-8B | 26.64 | Not recorded | 69ba318d0636e5bf4afc0831 | ### Ours (Grounded CoT) Accuracy: 64.93. Result ID: 69ba679947f34ce924a46e7a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Fine-tuned","Backbone":"Qwen3VL-4B"}. This record does not assert independent reproduction. ### GPT-5 Accuracy: 59.86. Result ID: 69ba679947f34ce924a46e75. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Closed-source MLLMs"}. This record does not assert independent reproduction. ### Ours (CoT) Accuracy: 59.57. Result ID: 69ba679947f34ce924a46e79. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Fine-tuned","Backbone":"Qwen3VL-4B"}. This record does not assert independent reproduction. ### GPT-5-mini Accuracy: 49. Result ID: 69ba679947f34ce924a46e74. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Closed-source MLLMs"}. This record does not assert independent reproduction. ### Qwen3-VL-8B Accuracy: 37.41. Result ID: 69ba679947f34ce924a46e6a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open-source MLLMs"}. This record does not assert independent reproduction. ### GPT-5.2 Accuracy: 37.07. Result ID: 69ba679947f34ce924a46e76. [Reporting paper]() Recorded evaluation configuration: {"Category":"Closed-source MLLMs"}. This record does not assert independent reproduction. ### Ours (Instruct-tuning) Accuracy: 36.5. Result ID: 69ba679947f34ce924a46e77. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Fine-tuned","Backbone":"Qwen3VL-4B"}. This record does not assert independent reproduction. ### Qwen3-VL-4B Accuracy: 35.19. Result ID: 69ba679947f34ce924a46e69. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open-source MLLMs"}. This record does not assert independent reproduction. ### Molmo2-8B Accuracy: 34.57. Result ID: 69ba679947f34ce924a46e6c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open-source Spatial Models"}. This record does not assert independent reproduction. ### Ours (Grounded CoT, Direct Answer at Test) Accuracy: 33.95. Result ID: 69ba679947f34ce924a46e78. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Fine-tuned","Backbone":"Qwen3VL-4B"}. This record does not assert independent reproduction. ### GPT-4o Accuracy: 33.93. Result ID: 69ba679947f34ce924a46e72. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Closed-source MLLMs"}. This record does not assert independent reproduction. ### GPT-5-nano Accuracy: 33.36. Result ID: 69ba679947f34ce924a46e73. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Closed-source MLLMs"}. This record does not assert independent reproduction. ### Qwen3-VL-2B Accuracy: 31.43. Result ID: 69ba679947f34ce924a46e68. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open-source MLLMs"}. This record does not assert independent reproduction. ### Molmo2-4B Accuracy: 31.36. Result ID: 69ba679947f34ce924a46e6b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open-source Spatial Models"}. This record does not assert independent reproduction. ### Gemma-3-12B-it Accuracy: 31.14. Result ID: 69ba679947f34ce924a46e65. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open-source MLLMs"}. This record does not assert independent reproduction. ### Internvl2.5-8B Accuracy: 30.93. Result ID: 69ba679947f34ce924a46e62. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Open-source MLLMs"}. This record does not assert independent reproduction. ### RoBobrain2.0 Accuracy: 30.79. Result ID: 69ba679947f34ce924a46e6d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open-source Spatial Models"}. This record does not assert independent reproduction. ### Openview Accuracy: 30.72. Result ID: 69ba679947f34ce924a46e6f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open-source Spatial Models"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B Accuracy: 30.29. Result ID: 69ba679947f34ce924a46e67. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open-source MLLMs"}. This record does not assert independent reproduction. ### MINDCUBE-Plain Accuracy: 30.21. Result ID: 69ba679947f34ce924a46e71. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open-source Spatial Models"}. This record does not assert independent reproduction. ### Mantis-8B (SigLip) Accuracy: 30.07. Result ID: 69ba679947f34ce924a46e63. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open-source MLLMs"}. This record does not assert independent reproduction. ### DeepSeek-VL2-Small Accuracy: 29.43. Result ID: 69ba679947f34ce924a46e64. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Open-source MLLMs"}. This record does not assert independent reproduction. ### MINDCUBE-Aug Accuracy: 29. Result ID: 69ba679947f34ce924a46e70. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open-source Spatial Models"}. This record does not assert independent reproduction. ### Random (chance) Accuracy: 28.59. Result ID: 69ba679947f34ce924a46e60. [Reporting paper]() Recorded evaluation configuration: {"Category":"Baseline"}. This record does not assert independent reproduction. ### Random (frequency) Accuracy: 28.59. Result ID: 69ba679947f34ce924a46e61. [Reporting paper]() Recorded evaluation configuration: {"Category":"Baseline"}. This record does not assert independent reproduction. ### Spatial-MLLM Accuracy: 27.93. Result ID: 69ba679947f34ce924a46e6e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open-source Spatial Models"}. This record does not assert independent reproduction. ### Idefics2-8B Accuracy: 26.64. Result ID: 69ba679947f34ce924a46e66. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Open-source MLLMs"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.