Sequential recommendation systems aim to predict users’ next interactions by analyzing historical sequences of their behaviors. However, these systems are vulnerable to adversarial attacks that disrupt input sequences through the injection of noisy or irrelevant interactions. Accordingly, this paper introduces the D ual cascad E d i N formati O n fil TER ing ( DENOTER ) algorithm designed to mitigate the impact of adversarial attacks. Unlike existing recommendation methods, which often overlook the varying importance of different components ( i.e ., items and their associated features) within behavioral sequences, DENOTER employs a two-tier adaptive filtering mechanism. Specifically, at the feature level, a dimension-adjustable, forward-only stochastic model is introduced to dynamically select and preserve essential features, thereby effectively filtering out malicious perturbations. At the item level, a Gumbel-driven sampling strategy is leveraged to selectively retain salient items. Through multi-layer refinement with alternating feature- and item-level filtering, DENOTER progressively reduces the influence of adversarially perturbed items and features within input sequences, enabling robust recommendation without specialized data augmentation or model retraining. Extensive empirical evaluations on five datasets and four encoders demonstrate that DENOTER consistently outperforms existing baselines under diverse adversarial conditions, achieving superior robustness and accuracy even in low-resource and out-of-domain settings.

In this paper, we introduce the IMPortance-awaReness maskIng NeTwork (IMPRINT), a novel approach to enhance the robustness of document retrieval systems against query variations, particularly those containing misspellings. Unlike previous models that treat all query components (words/features) equally, IMPRINT prioritizes the most important components while masking out less relevant ones. Specifically, we propose a Mutual Information-based measure to quantify component importance and integrate it into a dynamic masking mechanism that adjusts the retention probability of each component. Our method is evaluated on a combination of three benchmark datasets and three types of query variations. The experimental results show substantial performance gains compared to state-of-the-art models, achieving an average improvement of 1.2 absolute MRR@10 points in retrieval accuracy.