Papers
Topics
Authors
Recent
Search
2000 character limit reached

NetConv: Convolutional Model for Encrypted Traffic

Updated 7 July 2026
  • The paper introduces NetConv, a fully convolutional and pre-trained model that replaces self-attention with traffic convolution layers and a Continuous Byte Masking objective.
  • NetConv leverages window-wise byte scoring and sequence-wise byte gating to capture both local and global byte-patterns while maintaining linear runtime.
  • Empirical results show NetConv improves average Macro-F1 by 6.88% and increases inference throughput up to 7.41× compared to Transformer-based models.

1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 OR \131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 OR \131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 OR \131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 OR \131313 OR \131313 1313 1313 1313 1313 1313 OR \131313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 OR \131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 OR \131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 OR \131313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 OR \131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 OR \131313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 OR \131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 OR \131313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 OR \131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 OR \131313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 OR \131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 OR \131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 OR \131313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 OR \131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 OR \131313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 OR \131313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 OR \131313 1313 1313 OR \131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 OR \131313 OR \131313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 OR \131313 1313 1313 1313 1313 1313 1313 1313 1313 OR \131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 OR \131313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 OR \131313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 (Lin et al., 4 Aug 2025) NetConv Convolutions are Competitive with Transformers for Encrypted Traffic Classification with Pre-training131313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313 1313

Definition Search Book Streamline Icon: https://streamlinehq.com
References (1)

Topic to Video (Beta)

No one has generated a video about this topic yet.

Whiteboard

No one has generated a whiteboard explanation for this topic yet.

Follow Topic

Get notified by email when new papers are published related to NetConv.