# 3D/4D Video Question Answering on STI-Bench Source: [3D/4D Video Question Answering on STI-Bench]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:31:49.707Z Evaluation of model performance on the STI-Bench dataset for 3D/4D Video Question Answering tasks, typically measured by 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 | 3D/4D Video Question Answering | | Dataset (including any stated split) | STI-Bench | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 12 of 12 | | 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 | | --- | --- | --- | --- | | Gemini-2.5-Pro | 41.4 | Not recorded | 6948d086423ea42265729f78 | | GPT-5 | 39.3 | Not recorded | 6948d086423ea42265729f78 | | 4D-RGPT-8B | 37.6 | Not recorded | 6948d086423ea42265729f78 | | SpaceR-7B | 37 | Not recorded | 6948d086423ea42265729f78 | | VideoLLaMA3-7B | 35.2 | Not recorded | 6948d086423ea42265729f78 | | GPT-4o | 34.8 | Not recorded | 6948d086423ea42265729f78 | | NVILA-Lite-8B | 33.8 | Not recorded | 6948d086423ea42265729f78 | | ViLaSR-7B | 33.4 | Not recorded | 6948d086423ea42265729f78 | | Qwen2.5-VL-7B | 32.1 | Not recorded | 6948d086423ea42265729f78 | | SpatialReasoner-7B | 31 | Not recorded | 6948d086423ea42265729f78 | | LLaVA-NeXT-Video-7B | 29.9 | Not recorded | 6948d086423ea42265729f78 | | LLaVA-OneVision-7B | 29 | Not recorded | 6948d086423ea42265729f78 | ### Gemini-2.5-Pro Accuracy: 41.4. Result ID: 69982c705698665ec4876640. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GPT-5 Accuracy: 39.3. Result ID: 69982c705698665ec487663f. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### 4D-RGPT-8B Accuracy: 37.6. Result ID: 69982c705698665ec4876649. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"parameters":"8B"}. This record does not assert independent reproduction. ### SpaceR-7B Accuracy: 37. Result ID: 69982c705698665ec4876647. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"parameters":"7B"}. This record does not assert independent reproduction. ### VideoLLaMA3-7B Accuracy: 35.2. Result ID: 69982c705698665ec4876642. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"parameters":"7B"}. This record does not assert independent reproduction. ### GPT-4o Accuracy: 34.8. Result ID: 69982c705698665ec487663e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### NVILA-Lite-8B Accuracy: 33.8. Result ID: 69982c705698665ec4876648. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"parameters":"8B"}. This record does not assert independent reproduction. ### ViLaSR-7B Accuracy: 33.4. Result ID: 69982c705698665ec4876645. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"parameters":"7B"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B Accuracy: 32.1. Result ID: 69982c705698665ec4876641. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"parameters":"7B"}. This record does not assert independent reproduction. ### SpatialReasoner-7B Accuracy: 31. Result ID: 69982c705698665ec4876646. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"parameters":"7B"}. This record does not assert independent reproduction. ### LLaVA-NeXT-Video-7B Accuracy: 29.9. Result ID: 69982c705698665ec4876644. [Reporting paper]() Recorded evaluation configuration: {"parameters":"7B"}. This record does not assert independent reproduction. ### LLaVA-OneVision-7B Accuracy: 29. Result ID: 69982c705698665ec4876643. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"parameters":"7B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.