# When W4A4 Breaks Camouflaged Object Detection: Token-Group Dual-Constraint Activation Quantization

arXiv: https://arxiv.org/abs/2604.16855 · HTML: https://mlsanity.com/paper/2604.16855/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 32 of its 59 references that are in the map.
- First posted on arXiv 18 Apr 2026.

## Builds on (most cited first, 30 of 32)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers](https://mlsanity.com/paper/2210.17323/) (2022)
- [SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models](https://mlsanity.com/paper/2211.10438/) (2022)
- [Up or Down? Adaptive Rounding for Post-Training Quantization](https://mlsanity.com/paper/2004.10568/) (2020)
- [Parameter-Efficient Fine-Tuning Methods for Pretrained Language Models: A Critical Review and Assessment](https://mlsanity.com/paper/2312.12148/) (2023)
- [Anabranch Network for Camouflaged Object Segmentation](https://mlsanity.com/paper/2105.09451/) (2021)
- [Visual Attention Network](https://mlsanity.com/paper/2202.09741/) (2022)
- [Simultaneously Localize, Segment and Rank the Camouflaged Objects](https://mlsanity.com/paper/2103.04011/) (2021)
- [QDrop: Randomly Dropping Quantization for Extremely Low-bit Post-Training Quantization](https://mlsanity.com/paper/2203.05740/) (2022)
- [PTQ4ViT: Post-training quantization for vision transformers with twin uniform quantization](https://mlsanity.com/paper/2111.12293/) (2021)
- [RepQ-ViT: Scale Reparameterization for Post-Training Quantization of Vision Transformers](https://mlsanity.com/paper/2212.08254/) (2022)
- [Segment Anything Is Not Always Perfect: An Investigation of SAM on Different Real-world Applications](https://mlsanity.com/paper/2304.05750/) (2023)
- [Pretraining Large Language Models with NVFP4](https://mlsanity.com/paper/2509.25149/) (2025)
- [NoisyQuant: Noisy Bias-Enhanced Post-Training Activation Quantization for Vision Transformers](https://mlsanity.com/paper/2211.16056/) (2022)
- [CamoFormer: Masked Separable Attention for Camouflaged Object Detection](https://mlsanity.com/paper/2212.06570/) (2022)
- [PTQ4SAM: Post-Training Quantization for Segment Anything](https://mlsanity.com/paper/2405.03144/) (2024)
- [Instance-Aware Group Quantization for Vision Transformers](https://mlsanity.com/paper/2404.00928/) (2024)
- [A Simple yet Effective Network based on Vision Transformer for Camouflaged Object and Salient Object Detection](https://mlsanity.com/paper/2402.18922/) (2024)
- [FIMA-Q: Post-Training Quantization for Vision Transformers by Fisher Information Matrix Approximation](https://mlsanity.com/paper/2506.11543/) (2025)
- [ARCQuant: Boosting NVFP4 Quantization with Augmented Residual Channels for LLMs](https://mlsanity.com/paper/2601.07475/) (2026)
- [Open-Vocabulary Camouflaged Object Segmentation](https://mlsanity.com/paper/2311.11241/) (2023)
- [Improving SAM for Camouflaged Object Detection via Dual Stream Adapters](https://mlsanity.com/paper/2503.06042/) (2025)
- [When SAM2 Meets Video Camouflaged Object Segmentation: A Comprehensive Evaluation and Adaptation](https://mlsanity.com/paper/2409.18653/) (2024)
- [Controllable-LPMoE: Adapting to Challenging Object Segmentation via Dynamic Local Priors from Mixture-of-Experts](https://mlsanity.com/paper/2510.21114/) (2025)
- [ADFQ-ViT: Activation-Distribution-Friendly Post-Training Quantization for Vision Transformers](https://mlsanity.com/paper/2407.02763/) (2024)
- [Rethinking Detecting Salient and Camouflaged Objects in Unconstrained Scenes](https://mlsanity.com/paper/2412.10943/) (2024)
- [Evaluating SAM2 for Video Semantic Segmentation](https://mlsanity.com/paper/2512.01774/) (2025)
- [Towards Real Zero-Shot Camouflaged Object Segmentation without Camouflaged Annotations](https://mlsanity.com/paper/2410.16953/) (2024)
- [Open-Vocabulary Camouflaged Object Segmentation with Cascaded Vision Language Models](https://mlsanity.com/paper/2506.19300/) (2025)

## Abstract (arXiv)

Camouflaged object detection (COD) segments objects that intentionally blend with the background, so predictions depend on subtle texture and boundary cues. COD is often needed under tight on-device memory and latency budgets, making low-bit inference highly desirable. However, COD is unusually hard to quantize aggressively. We study post-training W4A4 quantization of Transformer-based COD and find a task-specific cliff: heavy-tailed background tokens dominate a shared activation range, inflating the step size and pushing weak-but-structured boundary cues into the zero bin. This exposes a token-local bottleneck -- remove cross-token range domination and bound the zero-bin mass under 4-bit activations. To address this, we introduce COD-TDQ, a COD-aware Token-group Dual-constraint activation Quantization method. COD-TDQ addresses this token-local bottleneck with two coupled steps: Direct-Sum Token-Group (DSTG) assigns token-group scales to suppress cross-token range domination, and Dual-Constraint Range Projection (DCRP) projects each token-group clip range to keep the step-to-dispersion ratio and the zero-bin mass bounded. Across four COD benchmarks and two baseline models (CFRN and ESCNet), COD-TDQ consistently achieves an Sα score more than 0.12 higher than that of the state-of-the-art quantization method without retraining. The code is available at https://github.com/MCG-NKU/nku-model-compre.