# 3D Visual Grounding on ScanRefer (val)

Source: [3D Visual Grounding on ScanRefer (val)](<https://www.sota2.com/research/sota/3d-visual-grounding-on-scanrefer-val>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-07-07T09:19:08.668Z

Evaluation of 3D Visual Grounding performance on the validation split of the ScanRefer dataset, typically measured using accuracy metrics at different IoU thresholds (0.25 and 0.50).

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 Visual Grounding |
| Dataset (including any stated split) | ScanRefer (val) |
| Metric | Overall Accuracy @ IoU 0.50 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 293 of 293 |
| 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

- [Overall Accuracy @ IoU 0.50](<https://www.sota2.com/research/sota/3d-visual-grounding-on-scanrefer-val.md?metric=Overall%20Accuracy%20%40%20IoU%200.50>): Higher is better; unit: PERCENT.

- [Overall Accuracy @ IoU 0.25](<https://www.sota2.com/research/sota/3d-visual-grounding-on-scanrefer-val.md?metric=Overall%20Accuracy%20%40%20IoU%200.25>): Higher is better; unit: PERCENT.

- [Unique Accuracy @ IoU 0.25](<https://www.sota2.com/research/sota/3d-visual-grounding-on-scanrefer-val.md?metric=Unique%20Accuracy%20%40%20IoU%200.25>): Higher is better; unit: PERCENT.

- [Unique Accuracy @ IoU 0.50](<https://www.sota2.com/research/sota/3d-visual-grounding-on-scanrefer-val.md?metric=Unique%20Accuracy%20%40%20IoU%200.50>): Higher is better; unit: PERCENT.

- [Multiple Accuracy @ IoU 0.25](<https://www.sota2.com/research/sota/3d-visual-grounding-on-scanrefer-val.md?metric=Multiple%20Accuracy%20%40%20IoU%200.25>): Higher is better; unit: PERCENT.

- [Multiple Accuracy @ IoU 0.50](<https://www.sota2.com/research/sota/3d-visual-grounding-on-scanrefer-val.md?metric=Multiple%20Accuracy%20%40%20IoU%200.50>): Higher is better; unit: PERCENT.

- [Average IoU](<https://www.sota2.com/research/sota/3d-visual-grounding-on-scanrefer-val.md?metric=Average%20IoU>): Higher is better; unit: SCALAR.

- [Average Distance](<https://www.sota2.com/research/sota/3d-visual-grounding-on-scanrefer-val.md?metric=Average%20Distance>): Lower is better; unit: OTHER.

- [Inference Speed (FPS)](<https://www.sota2.com/research/sota/3d-visual-grounding-on-scanrefer-val.md?metric=Inference%20Speed%20(FPS)>): Higher is better; unit: TIME.

- [Unique Accuracy](<https://www.sota2.com/research/sota/3d-visual-grounding-on-scanrefer-val.md?metric=Unique%20Accuracy>): Higher is better; unit: PERCENT.

- [Multiple Accuracy](<https://www.sota2.com/research/sota/3d-visual-grounding-on-scanrefer-val.md?metric=Multiple%20Accuracy>): Higher is better; unit: PERCENT.

- [Overall Accuracy](<https://www.sota2.com/research/sota/3d-visual-grounding-on-scanrefer-val.md?metric=Overall%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Overall Accuracy @ IoU 0.50 | Result date | Report paper |
| --- | --- | --- | --- |
| InstanceRefer | 66.83 | Not recorded | 68fa6c2b1e209131cea32841 |
| Multi-View Transformer | 66.45 | Not recorded | 68fa6c2b1e209131cea32841 |
| Multi-View Transformer | 63.04 | Not recorded | 68fa6c2b1e209131cea32841 |
| LLaVA-3D | 58.6 | Not recorded | 68fa78461e209131cea778b5 |
| GPT4Scene | 57 | Not recorded | 6a152977dbd7150506f7e41e |
| TGNN + BERT | 56.8 | Not recorded | 68fa6c2b1e209131cea32841 |
| CVP | 55.4 | Not recorded | 6938feb66cfd488761a53e5c |
| 3D-IDE | 54.53 | Not recorded | 69d48f987d0172140ec74337 |
| MCM-VG | 53.6 | Not recorded | 69f2e22655a3c32af89fc6ed |
| APEIRIA | 53.2 | Not recorded | 6a1e6407dbd7150506f8560a |
| TGNN | 53.01 | Not recorded | 68fa6c2b1e209131cea32841 |
| Video-3D-LLM | 51.7 | Not recorded | 6938feb66cfd488761a53e5c |
| Video-3D LLM | 51.7 | Not recorded | 69d48f987d0172140ec74337 |
| Video-3D LLM | 51.7 | Not recorded | 6a152977dbd7150506f7e41e |
| Video-3D LLM | 51.7 | Not recorded | 6a1e6407dbd7150506f8560a |
| Inst3D-LMM | 51.6 | Not recorded | 68fa79861e209131cea7e04b |
| Inst3D-LMM | 51.6 | Not recorded | 6a1e6407dbd7150506f8560a |
| PQ3D | 51.2 | Not recorded | 69d48f987d0172140ec74337 |
| PQ3D | 51.2 | Not recorded | 6a1e6407dbd7150506f8560a |
| APEIRIA | 51.2 | Not recorded | 6a1e6407dbd7150506f8560a |
| 3DRS\* | 50.83 | Not recorded | 69d48f987d0172140ec74337 |
| Chat-Scene | 50.23 | Not recorded | 68fa74f61e209131cea64e15 |
| Chat-Scene | 50.2 | Not recorded | 68fa79861e209131cea7e04b |
| Chat-Scene | 50.2 | Not recorded | 68fa78461e209131cea778b5 |
| Chat-Scene | 50.2 | Not recorded | 6940e7cf456146936284fb0a |
| Chat-Scene | 50.2 | Not recorded | 6938feb66cfd488761a53e5c |
| ChatScene | 50.2 | Not recorded | 69d48f987d0172140ec74337 |
| Chat-Scene | 50.2 | Not recorded | 6a152977dbd7150506f7e41e |
| Chat-Scene | 50.2 | Not recorded | 6a1e6407dbd7150506f8560a |
| SeqVLM | 49.6 | Not recorded | 69afa5b82d603e6de3e43e03 |
| SeqVLM | 49.6 | Not recorded | 69f2e22655a3c32af89fc6ed |
| ConcreteNet | 49.5 | Not recorded | 68fa79861e209131cea7e04b |
| SPAZER | 48.8 | Not recorded | 69afa5b82d603e6de3e43e03 |
| GPS | 48.1 | Not recorded | 6940e7cf456146936284fb0a |
| SceneVerse | 48.1 | Not recorded | 6a1e6407dbd7150506f8560a |
| Lemon | 48 | Not recorded | 6940e7cf456146936284fb0a |
| D-LISA | 46.9 | Not recorded | 68fa764b1e209131cea6c368 |
| TSP3D | 46.71 | Not recorded | 68fa77591e209131cea72954 |
| TSP3D | 46.7 | Not recorded | 69afa5b82d603e6de3e43e03 |
| baseline | 46.66 | Not recorded | 69d48f987d0172140ec74337 |
| ConcreteNet | 46.53 | Not recorded | 68fa74f61e209131cea64e15 |
| ConcreteNet | 46.5 | Not recorded | 6967232ee25ad2fa9d1ec2fb |
| ConcreteNet | 46.5 | Not recorded | 693a50236cfd488761a54b90 |
| ConcreteNet | 46.5 | Not recorded | 69afa5b82d603e6de3e43e03 |
| ConcreteNet | 46.5 | Not recorded | 6a152977dbd7150506f7e41e |
| 3D-VisTA | 46.2 | Not recorded | 6a4c880785dba69fef8b2694 |
| 3D-VisTA | 45.8 | Not recorded | 68fa74f61e209131cea64e15 |
| 3D-VisTA | 45.8 | Not recorded | 6967232ee25ad2fa9d1ec2fb |
| 3D-VisTA | 45.8 | Not recorded | 6940e7cf456146936284fb0a |
| 3D-VisTA | 45.8 | Not recorded | 693a50236cfd488761a54b90 |
| 3D-VisTA | 45.8 | Not recorded | 69afa5b82d603e6de3e43e03 |
| 3D-VisTA | 45.8 | Not recorded | 69d48f987d0172140ec74337 |
| 3D-VisTA | 45.8 | Not recorded | 6a0fe31df984739ed6739dc8 |
| 3D-VisTA | 45.8 | Not recorded | 6a1e6407dbd7150506f8560a |
| MCLN | 45.7 | Not recorded | 6967232ee25ad2fa9d1ec2fb |
| MCLN | 45.7 | Not recorded | 693a50236cfd488761a54b90 |
| MCLN | 45.7 | Not recorded | 69afa5b82d603e6de3e43e03 |
| MCLN | 45.7 | Not recorded | 6a0fe31df984739ed6739dc8 |
| MCLN | 45.7 | Not recorded | 6a152977dbd7150506f7e41e |
| G2TAM | 45.7 | Not recorded | 6a4c880785dba69fef8b2694 |
| G³-LQ | 45.58 | Not recorded | 68fa77591e209131cea72954 |
| MCLN | 45.53 | Not recorded | 68fa77591e209131cea72954 |
| MCLN | 45.53 | Not recorded | 69f2e22655a3c32af89fc6ed |
| DOrA | 44.8 | Not recorded | 68fa74f61e209131cea64e15 |
| DOrA | 44.8 | Not recorded | 68fa764b1e209131cea6c368 |
| G³-LQ | 44.72 | Not recorded | 68fa77591e209131cea72954 |
| M3DRef-CLIP | 44.7 | Not recorded | 68fa79861e209131cea7e04b |
| M3DRef-CLIP | 44.7 | Not recorded | 68fa74f61e209131cea64e15 |
| M3DRef-CLIP | 44.7 | Not recorded | 68fa78461e209131cea778b5 |
| G3-LQ | 44.7 | Not recorded | 6967232ee25ad2fa9d1ec2fb |
| G3-LQ | 44.7 | Not recorded | 693a50236cfd488761a54b90 |
| M3DRef-CLIP | 44.7 | Not recorded | 6938feb66cfd488761a53e5c |
| G3-LQ | 44.7 | Not recorded | 69afa5b82d603e6de3e43e03 |
| M3DRef-CLIP | 44.7 | Not recorded | 69d48f987d0172140ec74337 |
| G3-LQ | 44.7 | Not recorded | 6a0fe31df984739ed6739dc8 |
| G3-LQ | 44.7 | Not recorded | 6a152977dbd7150506f7e41e |
| M3DRef-CLIP | 44.7 | Not recorded | 6a1e6407dbd7150506f8560a |
| Grounded 3D-LLM | 44.1 | Not recorded | 68fa79861e209131cea7e04b |
| Grounded 3D-LLM | 44.1 | Not recorded | 68fa78461e209131cea778b5 |
| Grounded 3D-LLM | 44.1 | Not recorded | 6938feb66cfd488761a53e5c |

### InstanceRefer

Overall Accuracy @ IoU 0.50: 66.83. Result ID: 698ee75367eb08d78c866d17.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c2b1e209131cea32841.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a701e209131cea280ce.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Multi-View Transformer

Overall Accuracy @ IoU 0.50: 66.45. Result ID: 698ee75367eb08d78c866d19.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c2b1e209131cea32841.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c2b1e209131cea32841.md>)

[Implementation](<https://github.com/sega-hsj/MVT-3DVG>)

Recorded evaluation configuration: {"view num":"4"}. This record does not assert independent reproduction.

### Multi-View Transformer

Overall Accuracy @ IoU 0.50: 63.04. Result ID: 698ee75367eb08d78c866d18.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c2b1e209131cea32841.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c2b1e209131cea32841.md>)

[Implementation](<https://github.com/sega-hsj/MVT-3DVG>)

Recorded evaluation configuration: {"view num":"1"}. This record does not assert independent reproduction.

### LLaVA-3D

Overall Accuracy @ IoU 0.50: 58.6. Result ID: 699583963badfaffe67c006a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78461e209131cea778b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78461e209131cea778b5.md>)

[Implementation](<https://zcmax.github.io/projects/LLaVA-3D>)

Recorded evaluation configuration: {"Model Type":"Two-Stage 3D LMMs","Protocol":"Two-stage application"}. This record does not assert independent reproduction.

### GPT4Scene

Overall Accuracy @ IoU 0.50: 57. Result ID: 6a159b43fbc373592c0e8cea.

[Reporting paper](<https://www.sota2.com/research/paper/6a152977dbd7150506f7e41e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77591e209131cea72932.md>)

Recorded evaluation configuration: {"Venue":"ICLR’26","Zero-Shot":"false","Agent":"Qwen2-VL-7B"}. This record does not assert independent reproduction.

### TGNN + BERT

Overall Accuracy @ IoU 0.50: 56.8. Result ID: 698ee75367eb08d78c866d16.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c2b1e209131cea32841.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### CVP

Overall Accuracy @ IoU 0.50: 55.4. Result ID: 699871e65698665ec4d2123e.

[Reporting paper](<https://www.sota2.com/research/paper/6938feb66cfd488761a53e5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6938feb66cfd488761a53e5c.md>)

Recorded evaluation configuration: {"Method Category":"3D LMMs"}. This record does not assert independent reproduction.

### 3D-IDE

Overall Accuracy @ IoU 0.50: 54.53. Result ID: 69d4b43c47f34ce924bc30b1.

[Reporting paper](<https://www.sota2.com/research/paper/69d48f987d0172140ec74337.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d48f987d0172140ec74337.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### MCM-VG

Overall Accuracy @ IoU 0.50: 53.6. Result ID: 69f2ff9f47f34ce924d8f0be.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22655a3c32af89fc6ed.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f2e22655a3c32af89fc6ed.md>)

Recorded evaluation configuration: {"Modality":"3D+2D","Supervision":"Zero-Shot"}. This record does not assert independent reproduction.

### APEIRIA

Overall Accuracy @ IoU 0.50: 53.2. Result ID: 6a1eb2d4fbc373592c199989.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e6407dbd7150506f8560a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a1e6407dbd7150506f8560a.md>)

Recorded evaluation configuration: {"Output":"Text","modular inference enhancement":"true"}. This record does not assert independent reproduction.

### TGNN

Overall Accuracy @ IoU 0.50: 53.01. Result ID: 698ee75367eb08d78c866d15.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c2b1e209131cea32841.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Video-3D-LLM

Overall Accuracy @ IoU 0.50: 51.7. Result ID: 699871e65698665ec4d2123d.

[Reporting paper](<https://www.sota2.com/research/paper/6938feb66cfd488761a53e5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77a41e209131cea7494d.md>)

[Implementation](<https://github.com/LaVi-Lab/Video-3D-LLM>)

Recorded evaluation configuration: {"Method Category":"3D LMMs"}. This record does not assert independent reproduction.

### Video-3D LLM

Overall Accuracy @ IoU 0.50: 51.7. Result ID: 69d4b43c47f34ce924bc30ae.

[Reporting paper](<https://www.sota2.com/research/paper/69d48f987d0172140ec74337.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77a41e209131cea7494d.md>)

[Implementation](<https://github.com/LaVi-Lab/Video-3D-LLM>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Video-3D LLM

Overall Accuracy @ IoU 0.50: 51.7. Result ID: 6a159b43fbc373592c0e8ce9.

[Reporting paper](<https://www.sota2.com/research/paper/6a152977dbd7150506f7e41e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77a41e209131cea7494d.md>)

[Implementation](<https://github.com/LaVi-Lab/Video-3D-LLM>)

Recorded evaluation configuration: {"Venue":"CVPR’25","Zero-Shot":"false","Agent":"LLaVA-Video 7B"}. This record does not assert independent reproduction.

### Video-3D LLM

Overall Accuracy @ IoU 0.50: 51.7. Result ID: 6a1eb2d4fbc373592c199984.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e6407dbd7150506f8560a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77a41e209131cea7494d.md>)

[Implementation](<https://github.com/LaVi-Lab/Video-3D-LLM>)

Recorded evaluation configuration: {"Output":"Head","modular inference enhancement":"false"}. This record does not assert independent reproduction.

### Inst3D-LMM

Overall Accuracy @ IoU 0.50: 51.6. Result ID: 698b52678d67a41d0a6d119c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79861e209131cea7e04b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79861e209131cea7e04b.md>)

[Implementation](<https://github.com/hanxunyu/Inst3D-LMM>)

Recorded evaluation configuration: {"Method Paradigm":"Generalist"}. This record does not assert independent reproduction.

### Inst3D-LMM

Overall Accuracy @ IoU 0.50: 51.6. Result ID: 6a1eb2d4fbc373592c199987.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e6407dbd7150506f8560a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79861e209131cea7e04b.md>)

[Implementation](<https://github.com/hanxunyu/Inst3D-LMM>)

Recorded evaluation configuration: {"Output":"Text","modular inference enhancement":"false"}. This record does not assert independent reproduction.

### PQ3D

Overall Accuracy @ IoU 0.50: 51.2. Result ID: 69d4b43c47f34ce924bc30ab.

[Reporting paper](<https://www.sota2.com/research/paper/69d48f987d0172140ec74337.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74491e209131cea60c57.md>)

[Implementation](<https://www.sota2.com/pq3d.github.io>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### PQ3D

Overall Accuracy @ IoU 0.50: 51.2. Result ID: 6a1eb2d4fbc373592c199983.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e6407dbd7150506f8560a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74491e209131cea60c57.md>)

[Implementation](<https://www.sota2.com/pq3d.github.io>)

Recorded evaluation configuration: {"Output":"Head","modular inference enhancement":"false"}. This record does not assert independent reproduction.

### APEIRIA

Overall Accuracy @ IoU 0.50: 51.2. Result ID: 6a1eb2d4fbc373592c199988.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e6407dbd7150506f8560a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a1e6407dbd7150506f8560a.md>)

Recorded evaluation configuration: {"Output":"Text","modular inference enhancement":"false"}. This record does not assert independent reproduction.

### 3DRS\*

Overall Accuracy @ IoU 0.50: 50.83. Result ID: 69d4b43c47f34ce924bc30b0.

[Reporting paper](<https://www.sota2.com/research/paper/69d48f987d0172140ec74337.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79031e209131cea7baa1.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Chat-Scene

Overall Accuracy @ IoU 0.50: 50.23. Result ID: 69936c7e3a6c31484915d666.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74f61e209131cea64e15.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74f61e209131cea64e15.md>)

[Implementation](<https://github.com/Chat-3D/Chat-3D-v2>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Chat-Scene

Overall Accuracy @ IoU 0.50: 50.2. Result ID: 698b52678d67a41d0a6d119b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79861e209131cea7e04b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74f61e209131cea64e15.md>)

[Implementation](<https://github.com/Chat-3D/Chat-3D-v2>)

Recorded evaluation configuration: {"Method Paradigm":"Generalist"}. This record does not assert independent reproduction.

### Chat-Scene

Overall Accuracy @ IoU 0.50: 50.2. Result ID: 699583963badfaffe67c0069.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78461e209131cea778b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74f61e209131cea64e15.md>)

[Implementation](<https://github.com/Chat-3D/Chat-3D-v2>)

Recorded evaluation configuration: {"Model Type":"Two-Stage 3D LMMs"}. This record does not assert independent reproduction.

### Chat-Scene

Overall Accuracy @ IoU 0.50: 50.2. Result ID: 69984c6f5698665ec4a9c5fc.

[Reporting paper](<https://www.sota2.com/research/paper/6940e7cf456146936284fb0a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74f61e209131cea64e15.md>)

[Implementation](<https://github.com/Chat-3D/Chat-3D-v2>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Chat-Scene

Overall Accuracy @ IoU 0.50: 50.2. Result ID: 699871e65698665ec4d2123b.

[Reporting paper](<https://www.sota2.com/research/paper/6938feb66cfd488761a53e5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74f61e209131cea64e15.md>)

[Implementation](<https://github.com/Chat-3D/Chat-3D-v2>)

Recorded evaluation configuration: {"Method Category":"3D LMMs"}. This record does not assert independent reproduction.

### ChatScene

Overall Accuracy @ IoU 0.50: 50.2. Result ID: 69d4b43c47f34ce924bc30ac.

[Reporting paper](<https://www.sota2.com/research/paper/69d48f987d0172140ec74337.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74f61e209131cea64e15.md>)

[Implementation](<https://github.com/Chat-3D/Chat-3D-v2>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Chat-Scene

Overall Accuracy @ IoU 0.50: 50.2. Result ID: 6a159b43fbc373592c0e8ce8.

[Reporting paper](<https://www.sota2.com/research/paper/6a152977dbd7150506f7e41e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74f61e209131cea64e15.md>)

[Implementation](<https://github.com/Chat-3D/Chat-3D-v2>)

Recorded evaluation configuration: {"Venue":"NIPS’24","Zero-Shot":"false","Agent":"Vicuna-7B"}. This record does not assert independent reproduction.

### Chat-Scene

Overall Accuracy @ IoU 0.50: 50.2. Result ID: 6a1eb2d4fbc373592c199986.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e6407dbd7150506f8560a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74f61e209131cea64e15.md>)

[Implementation](<https://github.com/Chat-3D/Chat-3D-v2>)

Recorded evaluation configuration: {"Output":"Text","modular inference enhancement":"false"}. This record does not assert independent reproduction.

### SeqVLM

Overall Accuracy @ IoU 0.50: 49.6. Result ID: 69b00e636dbb7075b959e16e.

[Reporting paper](<https://www.sota2.com/research/paper/69afa5b82d603e6de3e43e03.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b011e209131cea862fa.md>)

Recorded evaluation configuration: {"Venue":"ACM’25","Agent Rep.":"✗","Agent VLM":"Doubao-1.5","Prior Info. Bbox":"✓","Prior Info. Class":"✓","Paradigm":"Open-Vocabulary Zero-Shot"}. This record does not assert independent reproduction.

### SeqVLM

Overall Accuracy @ IoU 0.50: 49.6. Result ID: 69f2ff9f47f34ce924d8f0bd.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22655a3c32af89fc6ed.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b011e209131cea862fa.md>)

Recorded evaluation configuration: {"Modality":"3D+2D","Supervision":"Zero-Shot"}. This record does not assert independent reproduction.

### ConcreteNet

Overall Accuracy @ IoU 0.50: 49.5. Result ID: 698b52678d67a41d0a6d118f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79861e209131cea7e04b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74291e209131cea601ed.md>)

Recorded evaluation configuration: {"Method Paradigm":"Closed-set, full sup."}. This record does not assert independent reproduction.

### SPAZER

Overall Accuracy @ IoU 0.50: 48.8. Result ID: 69b00e636dbb7075b959e16f.

[Reporting paper](<https://www.sota2.com/research/paper/69afa5b82d603e6de3e43e03.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7f014.md>)

Recorded evaluation configuration: {"Venue":"NeurIPS’25","Agent Rep.":"✗","Agent VLM":"GPT-4o","Prior Info. Bbox":"✓","Prior Info. Class":"✓","Paradigm":"Open-Vocabulary Zero-Shot"}. This record does not assert independent reproduction.

### GPS

Overall Accuracy @ IoU 0.50: 48.1. Result ID: 69984c6f5698665ec4a9c5fb.

[Reporting paper](<https://www.sota2.com/research/paper/6940e7cf456146936284fb0a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74e41e209131cea64757.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### SceneVerse

Overall Accuracy @ IoU 0.50: 48.1. Result ID: 6a1eb2d4fbc373592c199980.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e6407dbd7150506f8560a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74e41e209131cea64757.md>)

Recorded evaluation configuration: {"Output":"Head","modular inference enhancement":"false"}. This record does not assert independent reproduction.

### Lemon

Overall Accuracy @ IoU 0.50: 48. Result ID: 69984c6f5698665ec4a9c5fd.

[Reporting paper](<https://www.sota2.com/research/paper/6940e7cf456146936284fb0a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6940e7cf456146936284fb0a.md>)

Recorded evaluation configuration: {"training":"fine-tuned"}. This record does not assert independent reproduction.

### D-LISA

Overall Accuracy @ IoU 0.50: 46.9. Result ID: 699437053a6c3148491bce65.

[Reporting paper](<https://www.sota2.com/research/paper/68fa764b1e209131cea6c368.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa764b1e209131cea6c368.md>)

[Implementation](<https://github.com/haomengz/D-LISA>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### TSP3D

Overall Accuracy @ IoU 0.50: 46.71. Result ID: 6994821d3a6c3148491db2e9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77591e209131cea72954.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77591e209131cea72954.md>)

[Implementation](<https://github.com/GWxuan/TSP3D>)

Recorded evaluation configuration: {"Input":"3D","Stage":"Single-stage"}. This record does not assert independent reproduction.

### TSP3D

Overall Accuracy @ IoU 0.50: 46.7. Result ID: 69b00e636dbb7075b959e169.

[Reporting paper](<https://www.sota2.com/research/paper/69afa5b82d603e6de3e43e03.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77591e209131cea72954.md>)

[Implementation](<https://github.com/GWxuan/TSP3D>)

Recorded evaluation configuration: {"Venue":"CVPR’25","Agent Rep.":"✗","Agent VLM":"✗","Prior Info. Bbox":"-","Prior Info. Class":"-","Paradigm":"Fully Supervised Learning-Based"}. This record does not assert independent reproduction.

### baseline

Overall Accuracy @ IoU 0.50: 46.66. Result ID: 69d4b43c47f34ce924bc30af.

[Reporting paper](<https://www.sota2.com/research/paper/69d48f987d0172140ec74337.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77a41e209131cea7494d.md>)

[Implementation](<https://github.com/LaVi-Lab/Video-3D-LLM>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### ConcreteNet

Overall Accuracy @ IoU 0.50: 46.53. Result ID: 69936c7e3a6c31484915d664.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74f61e209131cea64e15.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74291e209131cea601ed.md>)

Recorded evaluation configuration: {"Venue":"ECCV24"}. This record does not assert independent reproduction.

### ConcreteNet

Overall Accuracy @ IoU 0.50: 46.5. Result ID: 6997c2aa5698665ec4197d66.

[Reporting paper](<https://www.sota2.com/research/paper/6967232ee25ad2fa9d1ec2fb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74291e209131cea601ed.md>)

Recorded evaluation configuration: {"Fine-tuned on Training Set":true,"Source":"ECCV’24"}. This record does not assert independent reproduction.

### ConcreteNet

Overall Accuracy @ IoU 0.50: 46.5. Result ID: 699869005698665ec4c87138.

[Reporting paper](<https://www.sota2.com/research/paper/693a50236cfd488761a54b90.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74291e209131cea601ed.md>)

Recorded evaluation configuration: {"Supervision Level":"Fully-Supervised","Venue":"ECCV’24","Agent":"—"}. This record does not assert independent reproduction.

### ConcreteNet

Overall Accuracy @ IoU 0.50: 46.5. Result ID: 69b00e636dbb7075b959e168.

[Reporting paper](<https://www.sota2.com/research/paper/69afa5b82d603e6de3e43e03.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74291e209131cea601ed.md>)

Recorded evaluation configuration: {"Venue":"ECCV’24","Agent Rep.":"✗","Agent VLM":"✗","Prior Info. Bbox":"-","Prior Info. Class":"-","Paradigm":"Fully Supervised Learning-Based"}. This record does not assert independent reproduction.

### ConcreteNet

Overall Accuracy @ IoU 0.50: 46.5. Result ID: 6a159b43fbc373592c0e8ce7.

[Reporting paper](<https://www.sota2.com/research/paper/6a152977dbd7150506f7e41e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74291e209131cea601ed.md>)

Recorded evaluation configuration: {"Venue":"ECCV’24","Zero-Shot":"false","Agent":"-"}. This record does not assert independent reproduction.

### 3D-VisTA

Overall Accuracy @ IoU 0.50: 46.2. Result ID: 6a4cc47360b88bb339c9697a.

[Reporting paper](<https://www.sota2.com/research/paper/6a4c880785dba69fef8b2694.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70461e209131cea4a1dd.md>)

[Implementation](<http://3d-vista.github.io/>)

Recorded evaluation configuration: {"w/o. LLM/VLM":"true","w/o. PC":"false","Zero-Shot":"false"}. This record does not assert independent reproduction.

### 3D-VisTA

Overall Accuracy @ IoU 0.50: 45.8. Result ID: 69936c7e3a6c31484915d663.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74f61e209131cea64e15.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70461e209131cea4a1dd.md>)

[Implementation](<http://3d-vista.github.io/>)

Recorded evaluation configuration: {"Venue":"ICCV23"}. This record does not assert independent reproduction.

### 3D-VisTA

Overall Accuracy @ IoU 0.50: 45.8. Result ID: 6997c2aa5698665ec4197d63.

[Reporting paper](<https://www.sota2.com/research/paper/6967232ee25ad2fa9d1ec2fb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70461e209131cea4a1dd.md>)

[Implementation](<http://3d-vista.github.io/>)

Recorded evaluation configuration: {"Fine-tuned on Training Set":true,"Source":"ICCV’23"}. This record does not assert independent reproduction.

### 3D-VisTA

Overall Accuracy @ IoU 0.50: 45.8. Result ID: 69984c6f5698665ec4a9c5fa.

[Reporting paper](<https://www.sota2.com/research/paper/6940e7cf456146936284fb0a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70461e209131cea4a1dd.md>)

[Implementation](<http://3d-vista.github.io/>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### 3D-VisTA

Overall Accuracy @ IoU 0.50: 45.8. Result ID: 699869005698665ec4c87135.

[Reporting paper](<https://www.sota2.com/research/paper/693a50236cfd488761a54b90.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70461e209131cea4a1dd.md>)

[Implementation](<http://3d-vista.github.io/>)

Recorded evaluation configuration: {"Supervision Level":"Fully-Supervised","Venue":"ICCV’23","Agent":"—"}. This record does not assert independent reproduction.

### 3D-VisTA

Overall Accuracy @ IoU 0.50: 45.8. Result ID: 69b00e636dbb7075b959e165.

[Reporting paper](<https://www.sota2.com/research/paper/69afa5b82d603e6de3e43e03.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70461e209131cea4a1dd.md>)

[Implementation](<http://3d-vista.github.io/>)

Recorded evaluation configuration: {"Venue":"ICCV’23","Agent Rep.":"✗","Agent VLM":"✗","Prior Info. Bbox":"-","Prior Info. Class":"-","Paradigm":"Fully Supervised Learning-Based"}. This record does not assert independent reproduction.

### 3D-VisTA

Overall Accuracy @ IoU 0.50: 45.8. Result ID: 69d4b43c47f34ce924bc30a7.

[Reporting paper](<https://www.sota2.com/research/paper/69d48f987d0172140ec74337.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70461e209131cea4a1dd.md>)

[Implementation](<http://3d-vista.github.io/>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### 3D-VisTA

Overall Accuracy @ IoU 0.50: 45.8. Result ID: 6a10073afbc373592c08a34c.

[Reporting paper](<https://www.sota2.com/research/paper/6a0fe31df984739ed6739dc8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70461e209131cea4a1dd.md>)

[Implementation](<http://3d-vista.github.io/>)

Recorded evaluation configuration: {"GT Pointcloud":"YES","Supervision":"Full"}. This record does not assert independent reproduction.

### 3D-VisTA

Overall Accuracy @ IoU 0.50: 45.8. Result ID: 6a1eb2d4fbc373592c19997f.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e6407dbd7150506f8560a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70461e209131cea4a1dd.md>)

[Implementation](<http://3d-vista.github.io/>)

Recorded evaluation configuration: {"Output":"Head","modular inference enhancement":"false"}. This record does not assert independent reproduction.

### MCLN

Overall Accuracy @ IoU 0.50: 45.7. Result ID: 6997c2aa5698665ec4197d65.

[Reporting paper](<https://www.sota2.com/research/paper/6967232ee25ad2fa9d1ec2fb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74111e209131cea5f974.md>)

Recorded evaluation configuration: {"Fine-tuned on Training Set":true,"Source":"ECCV’24"}. This record does not assert independent reproduction.

### MCLN

Overall Accuracy @ IoU 0.50: 45.7. Result ID: 699869005698665ec4c87137.

[Reporting paper](<https://www.sota2.com/research/paper/693a50236cfd488761a54b90.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74111e209131cea5f974.md>)

Recorded evaluation configuration: {"Supervision Level":"Fully-Supervised","Venue":"ECCV’24","Agent":"—"}. This record does not assert independent reproduction.

### MCLN

Overall Accuracy @ IoU 0.50: 45.7. Result ID: 69b00e636dbb7075b959e167.

[Reporting paper](<https://www.sota2.com/research/paper/69afa5b82d603e6de3e43e03.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74111e209131cea5f974.md>)

Recorded evaluation configuration: {"Venue":"ECCV’24","Agent Rep.":"✗","Agent VLM":"✗","Prior Info. Bbox":"-","Prior Info. Class":"-","Paradigm":"Fully Supervised Learning-Based"}. This record does not assert independent reproduction.

### MCLN

Overall Accuracy @ IoU 0.50: 45.7. Result ID: 6a10073afbc373592c08a34e.

[Reporting paper](<https://www.sota2.com/research/paper/6a0fe31df984739ed6739dc8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74111e209131cea5f974.md>)

Recorded evaluation configuration: {"GT Pointcloud":"YES","Supervision":"Full"}. This record does not assert independent reproduction.

### MCLN

Overall Accuracy @ IoU 0.50: 45.7. Result ID: 6a159b43fbc373592c0e8ce6.

[Reporting paper](<https://www.sota2.com/research/paper/6a152977dbd7150506f7e41e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74111e209131cea5f974.md>)

Recorded evaluation configuration: {"Venue":"ECCV’24","Zero-Shot":"false","Agent":"-"}. This record does not assert independent reproduction.

### G2TAM

Overall Accuracy @ IoU 0.50: 45.7. Result ID: 6a4cc47360b88bb339c9697f.

[Reporting paper](<https://www.sota2.com/research/paper/6a4c880785dba69fef8b2694.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a4c880785dba69fef8b2694.md>)

Recorded evaluation configuration: {"w/o. LLM/VLM":"true","w/o. PC":"true","Zero-Shot":"true"}. This record does not assert independent reproduction.

### G³-LQ

Overall Accuracy @ IoU 0.50: 45.58. Result ID: 6994821d3a6c3148491db2e2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77591e209131cea72954.md>)

Recorded evaluation configuration: {"Input":"3D","Stage":"Two-Stage"}. This record does not assert independent reproduction.

### MCLN

Overall Accuracy @ IoU 0.50: 45.53. Result ID: 6994821d3a6c3148491db2e3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77591e209131cea72954.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74111e209131cea5f974.md>)

Recorded evaluation configuration: {"Input":"3D","Stage":"Two-Stage"}. This record does not assert independent reproduction.

### MCLN

Overall Accuracy @ IoU 0.50: 45.53. Result ID: 69f2ff9f47f34ce924d8f0b6.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22655a3c32af89fc6ed.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74111e209131cea5f974.md>)

Recorded evaluation configuration: {"Modality":"3D","Supervision":"Fully-Supervised"}. This record does not assert independent reproduction.

### DOrA

Overall Accuracy @ IoU 0.50: 44.8. Result ID: 69936c7e3a6c31484915d665.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74f61e209131cea64e15.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa771e1e209131cea71315.md>)

Recorded evaluation configuration: {"Venue":"ArXiv24"}. This record does not assert independent reproduction.

### DOrA

Overall Accuracy @ IoU 0.50: 44.8. Result ID: 699437053a6c3148491bce64.

[Reporting paper](<https://www.sota2.com/research/paper/68fa764b1e209131cea6c368.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa771e1e209131cea71315.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### G³-LQ

Overall Accuracy @ IoU 0.50: 44.72. Result ID: 6994821d3a6c3148491db2e7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77591e209131cea72954.md>)

Recorded evaluation configuration: {"Input":"3D","Stage":"Single-stage"}. This record does not assert independent reproduction.

### M3DRef-CLIP

Overall Accuracy @ IoU 0.50: 44.7. Result ID: 698b52678d67a41d0a6d118e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79861e209131cea7e04b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70931e209131cea4bf5a.md>)

Recorded evaluation configuration: {"Method Paradigm":"Closed-set, full sup."}. This record does not assert independent reproduction.

### M3DRef-CLIP

Overall Accuracy @ IoU 0.50: 44.7. Result ID: 69936c7e3a6c31484915d662.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74f61e209131cea64e15.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70931e209131cea4bf5a.md>)

Recorded evaluation configuration: {"Venue":"ICCV23"}. This record does not assert independent reproduction.

### M3DRef-CLIP

Overall Accuracy @ IoU 0.50: 44.7. Result ID: 699583963badfaffe67c0066.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78461e209131cea778b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74211e209131cea5fdb4.md>)

Recorded evaluation configuration: {"Model Type":"Task-specific"}. This record does not assert independent reproduction.

### G3-LQ

Overall Accuracy @ IoU 0.50: 44.7. Result ID: 6997c2aa5698665ec4197d64.

[Reporting paper](<https://www.sota2.com/research/paper/6967232ee25ad2fa9d1ec2fb.md>)

Recorded evaluation configuration: {"Fine-tuned on Training Set":true,"Source":"CVPR’24"}. This record does not assert independent reproduction.

### G3-LQ

Overall Accuracy @ IoU 0.50: 44.7. Result ID: 699869005698665ec4c87136.

[Reporting paper](<https://www.sota2.com/research/paper/693a50236cfd488761a54b90.md>)

Recorded evaluation configuration: {"Supervision Level":"Fully-Supervised","Venue":"CVPR’24","Agent":"—"}. This record does not assert independent reproduction.

### M3DRef-CLIP

Overall Accuracy @ IoU 0.50: 44.7. Result ID: 699871e65698665ec4d21237.

[Reporting paper](<https://www.sota2.com/research/paper/6938feb66cfd488761a53e5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70931e209131cea4bf5a.md>)

Recorded evaluation configuration: {"Method Category":"Expert models"}. This record does not assert independent reproduction.

### G3-LQ

Overall Accuracy @ IoU 0.50: 44.7. Result ID: 69b00e636dbb7075b959e166.

[Reporting paper](<https://www.sota2.com/research/paper/69afa5b82d603e6de3e43e03.md>)

Recorded evaluation configuration: {"Venue":"CVPR’24","Agent Rep.":"✗","Agent VLM":"✗","Prior Info. Bbox":"-","Prior Info. Class":"-","Paradigm":"Fully Supervised Learning-Based"}. This record does not assert independent reproduction.

### M3DRef-CLIP

Overall Accuracy @ IoU 0.50: 44.7. Result ID: 69d4b43c47f34ce924bc30a6.

[Reporting paper](<https://www.sota2.com/research/paper/69d48f987d0172140ec74337.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70931e209131cea4bf5a.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### G3-LQ

Overall Accuracy @ IoU 0.50: 44.7. Result ID: 6a10073afbc373592c08a34d.

[Reporting paper](<https://www.sota2.com/research/paper/6a0fe31df984739ed6739dc8.md>)

Recorded evaluation configuration: {"GT Pointcloud":"YES","Supervision":"Full"}. This record does not assert independent reproduction.

### G3-LQ

Overall Accuracy @ IoU 0.50: 44.7. Result ID: 6a159b43fbc373592c0e8ce5.

[Reporting paper](<https://www.sota2.com/research/paper/6a152977dbd7150506f7e41e.md>)

Recorded evaluation configuration: {"Venue":"CVPR’24","Zero-Shot":"false","Agent":"-"}. This record does not assert independent reproduction.

### M3DRef-CLIP

Overall Accuracy @ IoU 0.50: 44.7. Result ID: 6a1eb2d4fbc373592c19997e.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e6407dbd7150506f8560a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70931e209131cea4bf5a.md>)

Recorded evaluation configuration: {"Output":"Head","modular inference enhancement":"false"}. This record does not assert independent reproduction.

### Grounded 3D-LLM

Overall Accuracy @ IoU 0.50: 44.1. Result ID: 698b52678d67a41d0a6d119a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79861e209131cea7e04b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75c11e209131cea69beb.md>)

Recorded evaluation configuration: {"Method Paradigm":"Generalist"}. This record does not assert independent reproduction.

### Grounded 3D-LLM

Overall Accuracy @ IoU 0.50: 44.1. Result ID: 699583963badfaffe67c0068.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78461e209131cea778b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75c11e209131cea69beb.md>)

Recorded evaluation configuration: {"Model Type":"Two-Stage 3D LMMs"}. This record does not assert independent reproduction.

### Grounded 3D-LLM

Overall Accuracy @ IoU 0.50: 44.1. Result ID: 699871e65698665ec4d2123a.

[Reporting paper](<https://www.sota2.com/research/paper/6938feb66cfd488761a53e5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75c11e209131cea69beb.md>)

Recorded evaluation configuration: {"Method Category":"3D LMMs"}. This record does not assert independent reproduction.

Page 1 of 4. This is an index page, not the full collection.

- [Next index page](<https://www.sota2.com/research/sota/3d-visual-grounding-on-scanrefer-val.md?metric=Overall%20Accuracy%20%40%20IoU%200.50&page=2>): Follow this link to discover more pages.
