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

arXiv: https://arxiv.org/abs/2607.18540 · HTML: https://mlsanity.com/paper/2607.18540/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 31 references that are in the map.
- First posted on arXiv 20 Jul 2026.

## Builds on (most cited first, 23 of 23)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Benchmarking Neural Network Robustness to Common Corruptions and Perturbations](https://mlsanity.com/paper/1903.12261/) (2019)
- [Deeper, Broader and Artier Domain Generalization](https://mlsanity.com/paper/1710.03077/) (2017)
- [Just Train Twice: Improving Group Robustness without Training Group Information](https://mlsanity.com/paper/2107.09044/) (2021)
- [Quantized Neural Networks: Training Neural Networks with Low Precision Weights and Activations](https://mlsanity.com/paper/1609.07061/) (2016)
- [Data-Free Quantization Through Weight Equalization and Bias Correction](https://mlsanity.com/paper/1906.04721/) (2019)
- [Post-Training Quantization for Vision Transformer](https://mlsanity.com/paper/2106.14156/) (2021)
- [Adversarially Robust Distillation](https://mlsanity.com/paper/1905.09747/) (2019)
- [Robust Quantization: One Model to Rule Them All](https://mlsanity.com/paper/2002.07686/) (2020)
- [RepairLLaMA: Efficient Representations and Fine-Tuned Adapters for Program Repair](https://mlsanity.com/paper/2312.15698/) (2023)
- [Zero-shot Adversarial Quantization](https://mlsanity.com/paper/2103.15263/) (2021)
- [TorchAO: PyTorch-Native Training-to-Serving Model Optimization](https://mlsanity.com/paper/2507.16099/) (2025)
- [Parameter-Efficient Fine-Tuning for Pre-Trained Vision Models: A Survey and Benchmark](https://mlsanity.com/paper/2402.02242/) (2024)
- [Comprehensive Survey of Model Compression and Speed up for Vision Transformers](https://mlsanity.com/paper/2404.10407/) (2024)
- [Design Space Exploration of Low-Bit Quantized Neural Networks for Visual Place Recognition](https://mlsanity.com/paper/2312.09028/) (2023)
- [Quadapter: Adapter for GPT-2 Quantization](https://mlsanity.com/paper/2211.16912/) (2022)
- [Out-Of-Domain Unlabeled Data Improves Generalization](https://mlsanity.com/paper/2310.00027/) (2023)
- [Annotation-Free Group Robustness via Loss-Based Resampling](https://mlsanity.com/paper/2312.04893/) (2023)
- [Robustness Feature Adapter for Efficient Adversarial Training](https://mlsanity.com/paper/2508.17680/) (2025)
- [Quantized Distillation: Optimizing Driver Activity Recognition Models for Resource-Constrained Environments](https://mlsanity.com/paper/2311.05970/) (2023)
- [Quantifying and Modeling Driving Styles in Trajectory Forecasting](https://mlsanity.com/paper/2503.04994/) (2025)

## Abstract (arXiv)

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.