# 3D Question Answering on Beacon3D ScanNet (test) Source: [3D Question Answering on Beacon3D ScanNet (test)]() Collection index: [llms.txt]() Data updated: 2026-03-19T10:14:47.958Z Evaluation of 3D Question Answering performance on the test split of the Beacon3D ScanNet dataset, utilizing metrics focused on classification, spatial understanding, and overall object/case correctness. 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 Question Answering | | Dataset (including any stated split) | Beacon3D ScanNet (test) | | Metric | Class Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 7 of 7 | | 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 - [Class Accuracy](): Higher is better; unit: PERCENT. - [Attribute Accuracy](): Higher is better; unit: PERCENT. - [Geometric Accuracy](): Higher is better; unit: PERCENT. - [Spatial Accuracy](): Higher is better; unit: PERCENT. - [Existence Accuracy](): Higher is better; unit: PERCENT. - [Overall Accuracy (Case)](): Higher is better; unit: PERCENT. - [Overall Accuracy (Object)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Class Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Loc3R-VLM | 44.8 | Not recorded | 69bb830f0636e5bf4afc1aa6 | | Video-3D LLM | 40.1 | Not recorded | 69bb830f0636e5bf4afc1aa6 | | GPT4Scene | 38.1 | Not recorded | 69bb830f0636e5bf4afc1aa6 | | Chat-Scene | 36.4 | Not recorded | 69bb830f0636e5bf4afc1aa6 | | LLaVA-3D | 35.1 | Not recorded | 69bb830f0636e5bf4afc1aa6 | | SceneVerse | 26.4 | Not recorded | 69bb830f0636e5bf4afc1aa6 | | LEO | 16.4 | Not recorded | 69bb830f0636e5bf4afc1aa6 | ### Loc3R-VLM Class Accuracy: 44.8. Result ID: 69bbcc8e47f34ce924a650c3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Video-3D LLM Class Accuracy: 40.1. Result ID: 69bbcc8e47f34ce924a650c2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GPT4Scene Class Accuracy: 38.1. Result ID: 69bbcc8e47f34ce924a650c0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Chat-Scene Class Accuracy: 36.4. Result ID: 69bbcc8e47f34ce924a650bf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### LLaVA-3D Class Accuracy: 35.1. Result ID: 69bbcc8e47f34ce924a650c1. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### SceneVerse Class Accuracy: 26.4. Result ID: 69bbcc8e47f34ce924a650bd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### LEO Class Accuracy: 16.4. Result ID: 69bbcc8e47f34ce924a650be. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.