Semantic Segmentation on ScanNet (val)
80.7mIoUConcerto
Evaluation Results
| Method | Links | |||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ConcertoLearnable Parameters=124.8M, Percentage of Parameters=100%, Fine-tuning Protocol=Full fine-tuning2025.10 | 80.7 | — | — | — | — | — | — | — | — | — | — | — | 87.4 | 93.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DOSProtocol=Fine-Tuning, Additional Data=true2025.12 | 79.7 | — | — | — | — | — | — | — | — | — | — | — | 86.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ChorusEvaluation Protocol=fine-tuned2025.12 | 79.4 | — | — | — | — | — | — | — | — | — | — | — | 87.7 | 92.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SonataProtocol=Fine-Tuning, Additional Data=true2025.12 | 79.4 | — | — | — | — | — | — | — | — | — | — | — | 86.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SonataLearnable Parameters=124.8M, Percentage of Parameters=100%, Fine-tuning Protocol=Full fine-tuning2025.10 | 79.4 | — | — | — | — | — | — | — | — | — | — | — | 86.1 | 92.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| D-DITRProtocol=Fine-Tuning, Additional Data=false2025.12 | 79.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Sonata + DINOv2.5Evaluation Protocol=decoder2025.03 | 79.19 | — | — | — | — | — | — | — | — | — | — | — | 86.66 | 92.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Sonata + DINOv2Evaluation Protocol=decoder2025.03 | 79.12 | — | — | — | — | — | — | — | — | — | — | — | 87.23 | 92.47 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SonataEvaluation Protocol=decoder2025.03 | 79.07 | — | — | — | — | — | — | — | — | — | — | — | 86.57 | 92.68 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PointINSProtocol=Fine-tuning2026.03 | 79 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DOSProtocol=Fine-Tuning, Additional Data=false2025.12 | 78.7 | — | — | — | — | — | — | — | — | — | — | — | 86.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DOSProtocol=Fine-tuning2026.03 | 78.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SonataEvaluation Protocol=fine-tuned2025.12 | 78.6 | — | — | — | — | — | — | — | — | — | — | — | 86.6 | 92.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Concrt-LParams=207.7M, SSL & MD=true, Resolution=τ* (2cm), Evaluation Protocol=linear probing2026.05 | 78.6 | — | — | — | — | — | — | — | — | — | — | — | 87.4 | 92.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MSMProtocol=Fine-Tuning, Additional Data=false2025.12 | 78.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MSCEvaluation Protocol=fine-tuned2025.12 | 78.2 | — | — | — | — | — | — | — | — | — | — | — | 85.3 | 92.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MSCLearnable Parameters=124.8M, Percentage of Parameters=100%, Fine-tuning Protocol=Full fine-tuning2025.10 | 78.2 | — | — | — | — | — | — | — | — | — | — | — | 85.3 | 92.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv3Protocol=Supervised, Additional Data=false2025.12 | 77.6 | — | — | — | — | — | — | — | — | — | — | — | 85 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv3Learnable Parameters=124.8M, Percentage of Parameters=100%, Fine-tuning Protocol=Full fine-tuning2025.10 | 77.6 | — | — | — | — | — | — | — | — | — | — | — | 85 | 92 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv3Evaluation protocol=supervised2026.03 | 77.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv3 (sup.)Protocol=Supervised2026.03 | 77.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv3Params=46.1M, SSL & MD=false, Resolution=τ* (2cm)2026.05 | 77.6 | — | — | — | — | — | — | — | — | — | — | — | 85 | 92 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv3#Params=46.1M, Limited Scenes (Pct.)=Full2025.12 | 77.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv3Evaluation Protocol=supervised2025.12 | 77.4 | — | — | — | — | — | — | — | — | — | — | — | 84.8 | 92 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SonataEvaluation Protocol=decoder2025.12 | 77.3 | — | — | — | — | — | — | — | — | — | — | — | 85.9 | 92 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SonataProtocol=Fine-tuning2026.03 | 77.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HUNetBackbone=HUNet, Training=Supervised2025.04 | 77 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PonderV2#Params.=41.0M2023.10 | 77 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DOSProtocol=Decoder2026.03 | 76.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LitePT-S*#Params=16.0M, Limited Scenes (Pct.)=Full2025.12 | 76.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PointINSProtocol=Decoder2026.03 | 76.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UtoniaParams=157.7M, SSL & MD=true, Resolution=τ* (2cm), Evaluation Protocol=linear probing2026.05 | 76.6 | — | — | — | — | — | — | — | — | — | — | — | 85.1 | 94.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LitePT-S#Params=12.7M, Limited Scenes (Pct.)=Full2025.12 | 76.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Sonata + DINOv2.5Evaluation Protocol=linear2025.03 | 76.44 | — | — | — | — | — | — | — | — | — | — | — | 85.68 | 91.33 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SparseUNet + PPT#Params.=41.0M2023.10 | 76.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Point-MoE-LParams=100M, Activated=60M, Precise Evaluator (None)=None, Joint indoor-outdoor training=true, Training batch composition=Standard, Normalization type (layernorm)=No2025.05 | 76.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PSAProtocol=Fine-tuning2026.03 | 76.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| InvariaParams=25.5M, SSL & MD=false, Resolution=τ* (2cm)2026.05 | 76.2 | — | — | — | — | — | — | — | — | — | — | — | 84.3 | 91.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Sonata + DINOv2Evaluation Protocol=linear2025.03 | 75.91 | — | — | — | — | — | — | — | — | — | — | — | 85.36 | 91.25 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MSCLearnable Parameters=39.2M, Percentage of Parameters=100%, Fine-tuning Protocol=Full fine-tuning2025.10 | 75.7 | — | — | — | — | — | — | — | — | — | — | — | 83.4 | 91.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| InvariaParams=25.5M, SSL & MD=false, Resolution=2τ* (4cm)2026.05 | 75.7 | — | — | — | — | — | — | — | — | — | — | — | 84 | 90.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MSP (Ours)Initialization=pre-trained2023.05 | 75.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Sp.Mink.Params=39.2M, SSL & MD=false, Resolution=τ* (2cm)2026.05 | 75.6 | — | — | — | — | — | — | — | — | — | — | — | 83.5 | 91.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MSCBackbone=SparseUNet, Pre-training=Masked Scene Contrast2023.03 | 75.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SparseUNet + MSC#Params.=39.2M2023.10 | 75.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SonataProtocol=Decoder2026.03 | 75.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SegContrastProtocol=Fine-tuning2026.03 | 75.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| InvariaParams=25.5M, SSL & MD=false, Resolution=0.5τ* (1cm)2026.05 | 75.5 | — | — | — | — | — | — | — | — | — | — | — | 83.8 | 91 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv2#Params.=12.8M2023.10 | 75.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv2#Params=12.8M, Limited Scenes (Pct.)=Full2025.12 | 75.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MSCBackbone=SparseUNet, Pct.=100%2023.03 | 75.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MSCBackbone=SparseUNet, Pts.=Full2023.03 | 75.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| NOMAEProtocol=Fine-tuning2026.03 | 75.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ChorusEvaluation Protocol=linear probing2025.12 | 75.2 | — | — | — | — | — | — | — | — | — | — | — | 84.8 | 90.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PPT-LParams=98M, Activated=98M, Precise Evaluator (None)=None, Joint indoor-outdoor training=true, Training batch composition=Standard, Normalization type (layernorm)=No2025.05 | 75.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ChorusEvaluation Protocol=decoder2025.12 | 75 | — | — | — | — | — | — | — | — | — | — | — | 83.1 | 90.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| StratifiedFormerType=p, Params.=18.8M, Training strategy=Expert2024.03 | 74.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| StratifiedFormer#Params.=18.8M2023.10 | 74.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LitePT-S*#Params=16.0M, Limited Annotations (Pts.)=2002025.12 | 74.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PCBackbone=SparseUNet, Pre-training=PointContrast2023.03 | 74.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PointContrastInitialization=pre-trained2023.05 | 74.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SparseUNet + PC#Params.=39.2M2023.10 | 74.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DOSProtocol=Linear Probing, Additional Data=true2025.12 | 73.9 | — | — | — | — | — | — | — | — | — | — | — | 83.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DOS*Evaluation protocol=linear probing2026.03 | 73.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CSCBackbone=SparseUNet, Pct.=100%2023.03 | 73.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CSCBackbone=SparseUNet, Pts.=Full2023.03 | 73.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CSCBackbone=SparseUNet, Pre-training=Contrastive Scene Contexts2023.03 | 73.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CSCInitialization=pre-trained2023.05 | 73.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SparseUNet + CSC#Params.=39.2M2023.10 | 73.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| InvariaParams=25.5M, SSL & MD=false, Resolution=3τ* (6cm)2026.05 | 73.8 | — | — | — | — | — | — | — | — | — | — | — | 82.7 | 90 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SonataEvaluation Protocol=linear probing2025.12 | 73.7 | — | — | — | — | — | — | — | — | — | — | — | 84.4 | 90.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MSP (Ours)Initialization=scratch2023.05 | 73.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv3-L**Params=97M, Activated=97M, Precise Evaluator (None)=None, Joint indoor-outdoor training=true, Training batch composition=Mix dataset batch, Normalization type (layernorm)=Yes2025.05 | 73.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PointINS*Evaluation protocol=linear probing2026.03 | 73.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv3-L*Params=97M, Activated=97M, Precise Evaluator (None)=None, Joint indoor-outdoor training=true, Training batch composition=Mix dataset batch, Normalization type (layernorm)=No2025.05 | 73.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Invaria-sParams=16.3M, SSL & MD=false, Resolution=τ* (2cm)2026.05 | 73.3 | — | — | — | — | — | — | — | — | — | — | — | 80.5 | 89.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Invaria-sParams=16.3M, SSL & MD=false, Resolution=0.5τ* (1cm)2026.05 | 73.1 | — | — | — | — | — | — | — | — | — | — | — | 80.2 | 89.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UtoniaParams=157.7M, SSL & MD=true, Resolution=0.5τ* (1cm), Evaluation Protocol=linear probing2026.05 | 73 | — | — | — | — | — | — | — | — | — | — | — | 81.9 | 95.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LitePT-S#Params=12.7M, Limited Annotations (Pts.)=2002025.12 | 72.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DOSProtocol=Linear Probing, Additional Data=false2025.12 | 72.8 | — | — | — | — | — | — | — | — | — | — | — | 83.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CSCLearnable Parameters=39.2M, Percentage of Parameters=100%, Fine-tuning Protocol=Full fine-tuning2025.10 | 72.8 | — | — | — | — | — | — | — | — | — | — | — | 81 | 90.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DOSProtocol=Linear Probing2026.03 | 72.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv3#Params=46.1M, Limited Annotations (Pts.)=2002025.12 | 72.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SonataParams=124.8M, SSL & MD=true, Resolution=τ* (2cm), Evaluation Protocol=linear probing2026.05 | 72.6 | — | — | — | — | — | — | — | — | — | — | — | 83.4 | 89.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SonataEvaluation Protocol=linear2025.03 | 72.52 | — | — | — | — | — | — | — | — | — | — | — | 83.11 | 89.74 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SonataProtocol=Linear Probing, Additional Data=true2025.12 | 72.5 | — | — | — | — | — | — | — | — | — | — | — | 83.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Sonata*Evaluation protocol=linear probing2026.03 | 72.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Concrt-LParams=207.7M, SSL & MD=true, Resolution=2τ* (4cm), Evaluation Protocol=linear probing2026.05 | 72.5 | — | — | — | — | — | — | — | — | — | — | — | 82.3 | 89.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PointINSProtocol=Linear Probing2026.03 | 72.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| 4DContrastPre-training=4D-informed2021.12 | 72.3 | — | — | — | — | — | — | — | — | — | — | — | 80.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SparseUNetLearnable Parameters=39.2M, Percentage of Parameters=100%, Fine-tuning Protocol=Full fine-tuning2025.10 | 72.3 | — | — | — | — | — | — | — | — | — | — | — | 80.2 | 90 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PCLearnable Parameters=39.2M, Percentage of Parameters=100%, Fine-tuning Protocol=Full fine-tuning2025.10 | 72.3 | — | — | — | — | — | — | — | — | — | — | — | 80.9 | 90.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SCBackbone=SparseUNet, Pct.=100%2023.03 | 72.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SCBackbone=SparseUNet, Pts.=Full2023.03 | 72.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SCBackbone=SparseUNet, Pre-training=Train from scratch2023.03 | 72.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PointContrastInitialization=scratch2023.05 | 72.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CSCInitialization=scratch2023.05 | 72.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SR-UNetBackbone=SR-UNet, Training=Supervised2025.04 | 72.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SparseUNet#Params.=39.2M2023.10 | 72.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MinkUNet#Params=39.2M, Limited Scenes (Pct.)=Full2025.12 | 72.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — |