Spectral-Adaptive Adversarial Hashing for Robust Image Retrieval
sigir26-087 · Full research paper · Gang Zhou, Shibiao Xu, Xiaolong Zheng, Daniel Dajun Zeng
Cross-Modal Hashing and Quantization · 100%
Abstract
Deep hashing is widely used in large-scale image retrieval systems due to its efficient retrieval performance. However, its susceptibility to adversarial attacks limits its security in practical applications. Adversarial training is the most effective method for improving robustness, but it often leads to a significant trade-off between robustness and retrieval accuracy. In this paper, we conduct spectral analysis and find that generating high-quality hash codes requires wide-frequency response models, whereas adversarial training forces the model into spectral collapse, degrading it to a low-frequency response model and weakening its discriminability. To address this issue, we propose a Spectral-Adaptive Adversarial Hashing (SAAH) framework, which selectively preserves discriminative and task-relevant frequency components while suppressing adversarially unstable ones, enabling robust hashing without sacrificing retrieval performance. Extensive experiments on benchmark datasets demonstrate that SAAH consistently achieves a superior balance between retrieval accuracy and adversarial robustness, achieving the best performance in both retrieval accuracy and robustness compared with existing robust hashing methods.
Assigned reviewers
| # | Reviewer | Match | Why (closest evidence publications in the taxonomy) | |
|---|---|---|---|---|
| 1 | Zheng Zhang 0006 | 100 | 100 Clustering-driven Deep Adversarial Hashing for scalable unsupervised cross-modal retrieval (2021) Cross-Modal Hashing and Quantization 100 Flexible Discrete Multi-view Hashing with Collective Latent Feature Learning (2020) Cross-Modal Hashing and Quantization 100 Inductive Structure Consistent Hashing via Flexible Semantic Calibration (2021) Cross-Modal Hashing and Quantization |
|
| 2 | Meiyu Liang | 77 | 100 Deep Unsupervised Momentum Contrastive Hashing for Cross-modal Retrieval (2023) Cross-Modal Hashing and Quantization 100 Dynamic Masking and Auxiliary Hash Learning for Enhanced Cross-Modal Retrieval (2025) Cross-Modal Hashing and Quantization 100 Asymmetric Pre-aligned Anchor Contrastive Enhanced Diffusion Hashing Model for Incomplete Multimodal Retrieval (2025) Cross-Modal Hashing and Quantization |
|
| 3 | Fengling Li 0001 | 76 | 100 Source-free Domain Adaptation with Multiple Alignment for Efficient Image Retrieval (2025) Cross-Modal Hashing and Quantization 100 Unified stable and generalizable online hashing for cross-modal retrieval (2026) Cross-Modal Hashing and Quantization 100 Noise-Robust Generative Hashing for Cross-Modal Retrieval (2026) Cross-Modal Hashing and Quantization |
Area chair
| Area chair | Match | |
|---|---|---|
| Xiaoyi Zeng | 31 |
Area-chair candidates (3)
| Xiaoyi Zeng assigned | 31 | 11 papers | |
| Chenyan Xiong | 31 | 15 papers | |
| Xiaoye Qu | 28 | 15 papers |
Candidates from the panel ranked by match score
| # | Reviewer | Match | Load | Why | |
|---|---|---|---|---|---|
| 1 | Zheng Zhang 0006 assigned | 100 | 3 | 100 Clustering-driven Deep Adversarial Hashing for scalable unsupervised cross-modal retrieval (2021) Cross-Modal Hashing and Quantization 100 Flexible Discrete Multi-view Hashing with Collective Latent Feature Learning (2020) Cross-Modal Hashing and Quantization |
|
| 2 | Meiyu Liang assigned | 77 | 3 | 100 Deep Unsupervised Momentum Contrastive Hashing for Cross-modal Retrieval (2023) Cross-Modal Hashing and Quantization 100 Dynamic Masking and Auxiliary Hash Learning for Enhanced Cross-Modal Retrieval (2025) Cross-Modal Hashing and Quantization |
|
| 3 | Fengling Li 0001 assigned | 76 | 3 | 100 Source-free Domain Adaptation with Multiple Alignment for Efficient Image Retrieval (2025) Cross-Modal Hashing and Quantization 100 Unified stable and generalizable online hashing for cross-modal retrieval (2026) Cross-Modal Hashing and Quantization |
|
| 4 | Jinpeng Wang 0002 | 56 | 3 | 100 Webly Supervised Deep Attentive Quantization (2021) Cross-Modal Hashing and Quantization 100 Efficient Self-Supervised Video Hashing with Selective State Spaces (2025) Cross-Modal Hashing and Quantization |
|
| 5 | Haokun Wen | 42 | 3 | 47 Self-Training Boosted Multi-Factor Matching Network for Composed Image Retrieval (2024) Universal Multimodal Embedding and MLLM Retrieval 47 A Comprehensive Survey on Composed Image Retrieval (2026) Universal Multimodal Embedding and MLLM Retrieval |
|
| 6 | Minghui Fang 0002 | 37 | 3 | 47 Open-Set Cross Modal Generalization via Multimodal Unified Representation (2025) Universal Multimodal Embedding and MLLM Retrieval 47 CART: A Generative Cross-Modal Retrieval Framework With Coarse-To-Fine Semantic Modeling (2025) Universal Multimodal Embedding and MLLM Retrieval |
|
| 7 | Xiao Wang 0043 | 36 | 3 | 43 Unsupervised Dense Retrieval with Conterfactual Contrastive Learning (2026) Robustness and Adversarial IR 38 An Inspection of the Reproducibility and Replicability of TCT-ColBERT (2022) Learned Sparse Retrieval Efficiency |
|
| 8 | Xueguang Ma | 35 | 3 | 42 Fine-Tuning LLaMA for Multi-Stage Text Retrieval (2024) Sparse and Dense Retrieval Encoders 41 LACONIC: Dense-Level Effectiveness for Scalable Sparse Retrieval via a Two-Phase Training Curriculum (2026) Learned Sparse Retrieval Efficiency |
|
| 9 | Ruqing Zhang 0001 | 35 | 3 | 46 PROP: Pre-training with Representative Words Prediction for Ad-hoc Retrieval (2021) Sparse and Dense Retrieval Encoders 46 B-PROP: Bootstrapped Pre-training with Representative Words Prediction for Ad-hoc Retrieval (2021) Sparse and Dense Retrieval Encoders |
|
| 10 | Sheng-Chieh Lin | 35 | 3 | 44 A Dense Representation Framework for Lexical and Semantic Matching (2023) Dense Bi-Encoder Training and Representation Learning 44 Aggretriever: A Simple Approach to Aggregate Textual Representations for Robust Dense Passage Retrieval (2023) Dense Bi-Encoder Training and Representation Learning |
|
| 11 | Jian Lang | 34 | 3 | 47 Predicting Micro-video Popularity via Multi-modal Retrieval Augmentation (2024) Universal Multimodal Embedding and MLLM Retrieval 47 Seeing the Unseen in Micro-Video Popularity Prediction: Self-Correlation Retrieval for Missing Modality Generation (2025) Universal Multimodal Embedding and MLLM Retrieval |
|
| 12 | Franco Maria Nardini | 34 | 3 | 43 Expansion via Prediction of Importance with Contextualization (2020) Sparse and Dense Retrieval Encoders 37 ReNeuIR at SIGIR 2026: The Fifth Workshop on Reaching Efficiency in Neural Information Retrieval (2026) Learned Sparse Retrieval Efficiency |
|
| 13 | Antonio Mallia | 34 | 3 | 41 Improved Learned Sparse Retrieval with Corpus-Specific Vocabularies (2024) Learned Sparse Retrieval Efficiency 41 Learning Passage Impacts for Inverted Indexes (2021) Learned Sparse Retrieval Efficiency |
|
| 14 | Jianfeng Dong | 33 | 3 | 38 W2VV++: Fully Deep Learning for Ad-hoc Video Search (2019) Video and Moment Retrieval 38 Dual Learning with Dynamic Knowledge Distillation for Partially Relevant Video Retrieval (2023) Video and Moment Retrieval |
|
| 15 | Huaiwen Zhang | 33 | 3 | 100 C2MR: Continual Cross-Modal Retrieval for Streaming Multi-modal Data (2023) Cross-Modal Hashing and Quantization 58 Dual Adversarial Graph Neural Networks for Multi-label Cross-modal Retrieval (2021) Cross-Modal Hashing and Quantization |
|
| 16 | Huiqiang Jiang | 33 | 3 | 47 Attentive Mask CLIP (2023) Universal Multimodal Embedding and MLLM Retrieval 37 LeanK: Learnable K Cache Channel Pruning for Efficient Decoding (2025) Learned Sparse Retrieval Efficiency |
|
| 17 | Xinyu Ma 0001 | 33 | 3 | 46 A Contrastive Pre-training Approach to Discriminative Autoencoder for Dense Retrieval (2022) Sparse and Dense Retrieval Encoders 46 Pre-train a Discriminative Text Encoder for Dense Retrieval via Contrastive Span Prediction (2022) Sparse and Dense Retrieval Encoders |
|
| 18 | Shengyao Zhuang | 32 | 3 | 43 Embark on DenseQuest: A System for Selecting the Best Dense Retriever for a Custom Collection (2024) Zero-Shot and Domain Generalization for Retrieval 43 PromptReps: Prompting Large Language Models to Generate Dense and Sparse Representations for Zero-Shot Document Retrieval (2024) Text Embedding and Representation Learning |
|
| 19 | Jingtao Zhan | 32 | 3 | 46 Scaling Laws For Dense Retrieval (2024) Sparse and Dense Retrieval Encoders 46 Optimizing Dense Retrieval Model Training with Hard Negatives (2021) Dense Bi-Encoder Training and Representation Learning |
|
| 20 | Yanzhao Zhang | 32 | 3 | 47 Towards Text-Image Interleaved Retrieval (2025) Universal Multimodal Embedding and MLLM Retrieval 47 Rethinking Composed Image Retrieval Evaluation: A Fine-Grained Benchmark from Image Editing (2026) Universal Multimodal Embedding and MLLM Retrieval |
|
| 21 | Hao Deng 0011 | 32 | 3 | 37 MMQ: Multimodal Mixture-of-Quantization Tokenization for Semantic ID Generation and User Behavioral Adaptation (2026) Item Tokenization and Semantic IDs 33 CSMF: Cascaded Selective Mask Fine-Tuning for Multi-Objective Embedding-Based Retrieval (2025) Multi-Task and Multi-Objective Optimization |
|
| 22 | Sebastian Bruch 0001 | 32 | 3 | 46 An Analysis of Fusion Functions for Hybrid Retrieval (2024) Dense Bi-Encoder Training and Representation Learning 37 Efficient Inverted Indexes for Approximate Retrieval over Learned Sparse Representations (2024) Learned Sparse Retrieval Efficiency |
|
| 24 | Liwei Wang 0009 | 31 | 3 | 47 Learning Deep Structure-Preserving Image-Text Embeddings (2016) Universal Multimodal Embedding and MLLM Retrieval 47 Learning Two-Branch Neural Networks for Image-Text Matching Tasks (2019) Universal Multimodal Embedding and MLLM Retrieval |
|
| 25 | Jurek Leonhardt | 31 | 3 | 36 Efficient Neural Ranking using Forward Indexes (2022) Learned Sparse Retrieval Efficiency 36 Efficient Neural Ranking Using Forward Indexes and Lightweight Encoders (2024) Learned Sparse Retrieval Efficiency |
|
| 27 | Yuchen Li 0006 | 31 | 3 | 36 COLTR: Semi-Supervised Learning to Rank With Co-Training and Over-Parameterization for Web Search (2023) Web-Scale Learning to Rank 36 GS2P: A Generative Pre-trained Learning to Rank Model with Over-parameterization for Web-Scale Search (2023) Web-Scale Learning to Rank |
|
| 28 | Linchao Zhu | 31 | 3 | 47 Knowledge-Enhanced Dual-Stream Zero-Shot Composed Image Retrieval (2024) Universal Multimodal Embedding and MLLM Retrieval 39 Attention as Selector: Unlocking VLM Attention for Long Document Page Retrieval (2026) Universal Multimodal Embedding and MLLM Retrieval |
|
| 29 | Jiangxia Cao | 30 | 3 | 36 Cross-Domain Recommendation to Cold-Start Users via Variational Information Bottleneck (2022) Domain Adaptation and Disentangled Representation for CDR 36 Disentangled Representations Learning for Multi-target Cross-domain Recommendation (2023) Domain Adaptation and Disentangled Representation for CDR |
|
| 30 | Yu-An Liu 0028 | 30 | 3 | 40 Perturbation-Invariant Adversarial Training for Neural Ranking Models: Improving the Effectiveness-Robustness Trade-Off (2024) Robustness and Adversarial IR 40 Robust-IR @ SIGIR 2025: The First Workshop on Robust Information Retrieval (2025) Robustness and Adversarial IR |
|
| 31 | Hung Vinh Tran | 30 | 3 | 37 On-device Content-based Recommendation with Single-shot Embedding Pruning: A Cooperative Game Perspective (2025) Content-Based Transferable Recommendation 30 Efficient Content-based Recommendation Model Training via Noise-aware Coreset Selection (2026) Content-Based Transferable Recommendation |
|
| 32 | Qiang Luo 0004 | 29 | 3 | 39 MISS: Multi-Modal Tree Indexing and Searching with Lifelong Sequential Behavior for Retrieval Recommendation (2025) Universal Multimodal Embedding and MLLM Retrieval 36 FARM: Frequency-Aware Model for Cross-Domain Live-Streaming Recommendation (2026) Domain Adaptation and Disentangled Representation for CDR |