The paper "Efficient Noise-Decoupling for Multi-Behavior Sequential Recommendation" addresses the challenges of mining user behavior sequences and reducing noise signals in multi-behavior sequential recommendation systems. The authors propose an Efficient Behavior Sequence Miner (EBM) to efficiently capture complex user patterns while maintaining low computational complexity and parameter count. EBM leverages fast Fourier transforms to achieve $O(N \log N)$ time complexity, incorporating techniques like frequency-aware fusion, chunked diagonal mechanism, and compactness regularization. The paper also introduces a Behavior-Aware Denoising module, including a Hard Noise Eliminator for discrete noise and a Soft Noise Filter for continuous noise, along with a Noise-Decoupling Contrastive Learning approach and a guided training strategy to effectively remove noise. Experiments on real-world datasets demonstrate the effectiveness and efficiency of the proposed method, END4Rec, in handling multi-behavior sequential recommendation tasks.The paper "Efficient Noise-Decoupling for Multi-Behavior Sequential Recommendation" addresses the challenges of mining user behavior sequences and reducing noise signals in multi-behavior sequential recommendation systems. The authors propose an Efficient Behavior Sequence Miner (EBM) to efficiently capture complex user patterns while maintaining low computational complexity and parameter count. EBM leverages fast Fourier transforms to achieve $O(N \log N)$ time complexity, incorporating techniques like frequency-aware fusion, chunked diagonal mechanism, and compactness regularization. The paper also introduces a Behavior-Aware Denoising module, including a Hard Noise Eliminator for discrete noise and a Soft Noise Filter for continuous noise, along with a Noise-Decoupling Contrastive Learning approach and a guided training strategy to effectively remove noise. Experiments on real-world datasets demonstrate the effectiveness and efficiency of the proposed method, END4Rec, in handling multi-behavior sequential recommendation tasks.