Decision-FLIP: Budget-Free Structured Pruning for Gaussian Occupancy Models
Quick Answer
Decision-FLIP introduces a budget-free pruning framework for Gaussian occupancy models, achieving 12-79% row removal and reducing latency by 7-45%.
Quick Take
This method outperforms traditional pruning techniques, maintaining model performance with minimal semantic mIoU change.
Key Points
- Exact Mean-FLIP scores rows based on voxel decision changes upon removal.
- Automatically selected cuts led to 8.0% whole-model savings and 16.6% peak-memory savings.
- Random row removal negatively impacted GaussianFormer-1 and OccFormer models.
- Survivor count is budget-free, focusing on stable checkpoint-row identity.
- Method integrates scores over each row's lifecycle for streaming banks.
Paper Resources
📖 Reader Mode
~2 min readAbstract:Occupancy predictors rely on large persistent banks of Gaussians or learned queries whose size is fixed by hand. Common pruning proxies, such as opacity, activation frequency, or parameter magnitude, do not measure whether removing a row changes the model's voxel decisions. We introduce Decision-FLIP, a budget-free compilation framework whose scoring and structure selection need only unlabeled calibration inputs. Exact Mean-FLIP scores each persistent row by the number of voxel argmax decisions that change when the row is removed from the readout; a guarded Otsu split of the log-score distribution selects the survivor count without a target and abstains when the split is unsupported; and the survivors are physically compiled into a smaller checkpoint, so camera sampling, attention, and aggregation shrink with the bank. For streaming banks, scores are integrated over each row's lifecycle through warp, death, and refill. Across four Gaussian-bank model-dataset settings and one persistent-query transfer, the automatically selected cuts remove 12-79% of eligible rows and cut bank/query-dependent latency by 7-45% across settings with component profiles, with whole-model savings of up to 8.0% on the primary endpoints bounded by fixed-cost image, voxel, and sensor modules and peak-memory savings of up to 16.6%. Under matched one-epoch recovery controls, semantic mIoU changes by at most 0.18 point. Same-$K$ controls show that the ranking, not only the bank size, matters: random removal collapses GaussianFormer-1 and OccFormer, while label-free Mean-FLIP matches or outperforms supervised Taylor ranking on all five settings where that control was run. The method applies when removable primitives keep a stable checkpoint-row identity; it is budget-free in survivor count, not parameter-free.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2610.20917 [cs.CV] |
| (or arXiv:2610.20917v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.20917 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Jingxing Li [view email]
[v1]
Wed, 14 Oct 2026 20:48:09 UTC (359 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →ProMoE-FL: Prototype-conditioned Mixture of Experts for Multimodal Federated Learning with Missing Modalities
ProMoE-FL introduces a Prototype-conditioned Mixture-of-Experts framework for multimodal federated learning, effectively addressing missing modalities. It outperforms existing methods on four chest X-ray datasets, demonstrating superior feature synthesis capabilities in both homogeneous and heterogeneous settings.