Pith. sign in

REVIEW 11 cited by

QDrop: Randomly Dropping Quantization for Extremely Low-bit Post-Training Quantization

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2203.05740 v2 pith:KSZCFVQP submitted 2022-03-11 cs.CV cs.AI

Xiuying Wei , Ruihao Gong , Yuhang Li , Xianglong Liu , Fengwei Yu This is my paper
classification cs.CVcs.AI
keywords qdropquantizationlow-bitaccuracyactivationclassificationextremelygithub
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Recently, post-training quantization (PTQ) has driven much attention to produce efficient neural networks without long-time retraining. Despite its low cost, current PTQ works tend to fail under the extremely low-bit setting. In this study, we pioneeringly confirm that properly incorporating activation quantization into the PTQ reconstruction benefits the final accuracy. To deeply understand the inherent reason, a theoretical framework is established, indicating that the flatness of the optimized low-bit model on calibration and test data is crucial. Based on the conclusion, a simple yet effective approach dubbed as QDROP is proposed, which randomly drops the quantization of activations during PTQ. Extensive experiments on various tasks including computer vision (image classification, object detection) and natural language processing (text classification and question answering) prove its superiority. With QDROP, the limit of PTQ is pushed to the 2-bit activation for the first time and the accuracy boost can be up to 51.49%. Without bells and whistles, QDROP establishes a new state of the art for PTQ. Our code is available at https://github.com/wimh966/QDrop and has been integrated into MQBench (https://github.com/ModelTC/MQBench)

Discussion (0). Sign in to comment.

Forward citations

Cited by 11 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

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

    cs.CV 2026-04 unverdicted novelty 7.0 of 10

    Token-group activation scaling with dual constraints on step-to-dispersion and zero-bin mass restores W4A4 COD accuracy from catastrophic collapse to near-FP32 on CFRN and ESCNet.

  2. Enhancing Generalization in Data-free Quantization via Mixup-class Prompting

    cs.CV 2025-07 conditional novelty 6.0 of 10

    Using two class labels in text prompts to generate synthetic calibration images improves data-free post-training quantization accuracy, especially in low-bit settings.

  3. Task-Specific Zero-shot Quantization-Aware Training for Object Detection

    cs.CV 2025-07 conditional novelty 6.0 of 10

    A zero-shot quantization-aware training method for object detectors that synthesizes task-specific images with bounding-box labels via adaptive label sampling, then distills task-specific knowledge into the quantized network.

  4. MPQ-DMv2: Flexible Residual Mixed Precision Quantization for Low-Bit Diffusion Models with Temporal Distillation

    cs.CV 2025-07 conditional novelty 6.0 of 10

    MPQ-DMv2 adds binary residual quantization, temporal relation distillation, and SVD-initialized LoRA to mixed-precision quantization, improving low-bit diffusion model generation quality.

  5. Flexible Mixed Precision Quantization for Learned Image Compression

    eess.IV 2025-06 conditional novelty 6.0 of 10

    A rate-distortion sensitivity criterion assigns per-layer bit-widths, yielding about 1 to 2 percent BD-Rate improvement over 8-bit fixed-precision quantization at matched model size for learned image compression.

  6. Compress Any Segment Anything Model (SAM)

    cs.CV 2025-07 conditional novelty 5.0 of 10

    Birkhoff compresses 18 SAM variants to about one-fifth their size with less than 1% accuracy loss, data-free, using a trajectory-based codebook and a fused GPU operator.

  7. PQCAD-DM: Progressive Quantization and Calibration-Assisted Distillation for Extremely Efficient Diffusion Model

    cs.CV 2025-06 conditional novelty 5.0 of 10

    PQCAD-DM combines progressive two-stage quantization with calibration-assisted distillation to halve the sampling steps of a quantized diffusion model while keeping FID scores close to baseline.

  8. Post-Training Quantization for Video Matting

    cs.CV 2025-06 conditional novelty 5.0 of 10

    PTQ4VM applies block-wise, globally affine-calibrated, optical-flow-guided post-training quantization to video matting, reaching near full-precision accuracy at 4-bit weights and activations.

  9. MixFrag: Fragility-Guided Mixed-Precision Post-Training Quantization for Vision Transformers

    cs.CV 2026-07 conditional novelty 4.5 of 10

    KL-isolation fragility plus MCKP bit allocation yields mixed-precision ViT PTQ that lags recent ImageNet PTQ but reports large COCO AP gains at MP3/MP3.

  10. Structured Pruning and Quantization for Learned Image Compression

    eess.IV 2025-06 conditional novelty 4.0 of 10

    A rate-distortion-aware neural architecture search plus joint pruning and 8-bit quantization reduces learned image compression model size by up to about 4.9x with small rate-distortion penalty.

  11. NQKV: A KV Cache Quantization Scheme Based on Normal Distribution Characteristics

    cs.LG 2025-05 conditional novelty 4.0 of 10

    NQKV is a block-wise 4-bit KV cache quantizer that uses a normal-distribution lookup table to let OPT models run with larger batches or longer contexts at negligible measured accuracy cost.

Pith tools