Recti-Q: Feature-Space Rectification for Out-of-Distribution-Robust Quantized Perception in Edge Robotics

A new study from arXiv (2607.18540v1) reveals that post-training quantization (PTQ) on edge robotics platforms creates a Quantization-Induced Robustness Gap, degrading reliability under distribution shifts like sensor noise and severe weather despite preserving clean in-distribution accuracy. The researchers propose Recti-Q, a lightweight feature-space rectification framework that freezes the quantized backbone and trains a small classifier-head LoRA adapter using only source data, recovering lost robustness with less than 1% parameter overhead (as small as 6 KB) and preserving over 99% of PTQ memory savings.

arXiv:2607.18540v1 Announce Type: new Abstract: Robotic perception pipelines increasingly rely on large vision backbones deployed on SWaP-constrained edge platforms, making post-training quantization PTQ attractive for real-time inference. However, while PTQ often preserves clean in-distribution accuracy, we show that it can substantially degrade reliability under deployment-relevant distribution shifts e.g., sensor noise, severe weather, and novel operating environments , creating a Quantization-Induced Robustness Gap. Across foundational vision benchmarks ImageNet-C and PACS , 4-bit PTQ models exhibit pronounced robustness degradation despite negligible ID accuracy loss. To address this, we propose Recti-Q, a lightweight feature-space rectification framework that freezes the quantized backbone and trains a small classifier-head LoRA adapter using only source data. Recti-Q is architecture-agnostic across CNNs and Transformers, supports efficient teacher-free training, and recovers a significant portion of the lost robustness, in some cases matching or exceeding FP32 performance. At less than 1% parameter overhead as small as 6 KB , Recti-Q preserves over 99% of PTQ memory savings, adds negligible compute, and enables low-bandwidth Over-The-Air OTA resilience patching for deployed robotic fleets operating in unpredictable physical environments.