# Can we Trust Unreliable Voxels? Exploring 3D Semantic Occupancy Prediction under Label Noise

arXiv: https://arxiv.org/abs/2603.06279 · HTML: https://mlsanity.com/paper/2603.06279/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 35 references that are in the map.
- First posted on arXiv 6 Mar 2026.

## Builds on (most cited first, 22 of 22)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [A Closer Look at Memorization in Deep Networks](https://mlsanity.com/paper/1706.05394/) (2017)
- [SparseOcc: Rethinking Sparse Latent Representation for Vision-Based Semantic Occupancy Prediction](https://mlsanity.com/paper/2404.09502/) (2024)
- [SCPNet: Semantic Scene Completion on Point Cloud](https://mlsanity.com/paper/2303.06884/) (2023)
- [UniOcc: A Unified Benchmark for Occupancy Forecasting and Prediction in Autonomous Driving](https://mlsanity.com/paper/2503.24381/) (2025)
- [Understanding Self-Distillation in the Presence of Label Noise](https://mlsanity.com/paper/2301.13304/) (2023)
- [QuadricFormer: Scene as Superquadrics for 3D Semantic Occupancy Prediction](https://mlsanity.com/paper/2506.10977/) (2025)
- [From Semi-supervised to Omni-supervised Room Layout Estimation Using Point Clouds](https://mlsanity.com/paper/2301.13865/) (2023)
- [Language Driven Occupancy Prediction](https://mlsanity.com/paper/2411.16072/) (2024)
- [ProtoOcc: Accurate, Efficient 3D Occupancy Prediction Using Dual Branch Encoder-Prototype Query Decoder](https://mlsanity.com/paper/2412.08774/) (2024)
- [AutoOcc: Automatic Open-Ended Semantic Occupancy Annotation via Vision-Language Guided Gaussian Splatting](https://mlsanity.com/paper/2502.04981/) (2025)
- [VoxDet: Rethinking 3D Semantic Occupancy Prediction as Dense Object Detection](https://mlsanity.com/paper/2506.04623/) (2025)
- [Federated Learning with Extremely Noisy Clients via Negative Distillation](https://mlsanity.com/paper/2312.12703/) (2023)
- [Semantic Causality-Aware Vision-Based 3D Occupancy Prediction](https://mlsanity.com/paper/2509.08388/) (2025)
- [OCCUQ: Exploring Efficient Uncertainty Quantification for 3D Occupancy Prediction](https://mlsanity.com/paper/2503.10605/) (2025)
- [Online Diffusion-Based 3D Occupancy Prediction at the Frontier with Probabilistic Map Reconciliation](https://mlsanity.com/paper/2409.10681/) (2024)
- [Particle-based Instance-aware Semantic Occupancy Mapping in Dynamic Environments](https://mlsanity.com/paper/2409.11975/) (2024)
- [Active Negative Loss: A Robust Framework for Learning with Noisy Labels](https://mlsanity.com/paper/2412.02373/) (2024)
- [H3O: Hyper-Efficient 3D Occupancy Prediction with Heterogeneous Supervision](https://mlsanity.com/paper/2503.04059/) (2025)
- [Variation-Bounded Loss for Noise-Tolerant Learning](https://mlsanity.com/paper/2511.12143/) (2025)
- [Robust Noisy Label Learning via Two-Stream Sample Distillation](https://mlsanity.com/paper/2404.10499/) (2024)
- [VPOcc: Exploiting Vanishing Point for 3D Semantic Occupancy Prediction](https://mlsanity.com/paper/2408.03551/) (2024)

## Built on by (most cited first, 1 of 1)

- [RadarGSOcc: Doppler-Aware 3D Semantic Occupancy Prediction Using 4D Radar Tensor with Geometric-Semantic Feature Extraction](https://mlsanity.com/paper/2610.18686/) (2026)

## Abstract (arXiv)

3D semantic occupancy prediction is a cornerstone of robotic perception, yet real-world voxel annotations are inherently corrupted by structural artifacts and dynamic trailing effects. This raises a critical but underexplored question: can autonomous systems safely rely on such unreliable occupancy supervision? To systematically investigate this issue, we establish OccNL, the first benchmark dedicated to 3D occupancy under occupancy-asymmetric and dynamic trailing noise. Our analysis reveals a fundamental domain gap: state-of-the-art 2D label noise learning strategies collapse catastrophically in sparse 3D voxel spaces, exposing a critical vulnerability in existing paradigms. To address this challenge, we propose DPR-Occ, a principled label-noise-robust framework that constructs reliable supervision through dual-source partial label reasoning. By synergizing temporal model memory with representation-level structural affinity, DPR-Occ dynamically expands and prunes candidate label sets to preserve true semantics while suppressing noise propagation. Extensive experiments on SemanticKITTI demonstrate that DPR-Occ prevents geometric and semantic collapse under extreme corruption. Notably, even at 90% label noise, our method achieves significant performance gains (up to 2.57% mIoU and 13.91% IoU) over existing label noise learning baselines adapted to the 3D occupancy prediction task. By bridging label noise learning and 3D perception, OccNL and DPR-Occ provide a reliable foundation for safety-critical robotic perception in dynamic environments. The benchmark and source code will be made publicly available at https://github.com/mylwx/OccNL.