{"total":128,"rows":[{"id":"2404.14082","title":"Mechanistic Interpretability for AI Safety -- A Review","created":"2024-04-22","cited_by_count":422,"area":"mechinterp"},{"id":"2408.07666","title":"Model Merging in LLMs, MLLMs, and Beyond: Methods, Theories, Applications and Opportunities","created":"2024-08-14","cited_by_count":254,"area":null},{"id":"2401.14446","title":"Black-Box Access is Insufficient for Rigorous AI Audits","created":"2024-01-25","cited_by_count":126,"area":null},{"id":"2311.12075","title":"BadCLIP: Dual-Embedding Guided Backdoor Attack on Multimodal Contrastive Learning","created":"2023-11-20","cited_by_count":115,"area":"safety"},{"id":"2403.05030","title":"Defending Against Unforeseen Failure Modes with Latent Adversarial Training","created":"2024-03-08","cited_by_count":88,"area":"safety"},{"id":"2402.13851","title":"VL-Trojan: Multimodal Instruction Backdoor Attacks against Autoregressive Visual Language Models","created":"2024-02-21","cited_by_count":87,"area":"mllm"},{"id":"2304.12298","title":"BadGPT: Exploring Security Vulnerabilities of ChatGPT via Backdoor Attacks to InstructGPT","created":"2023-02-21","cited_by_count":76,"area":"safety"},{"id":"2209.15179","title":"Physical Adversarial Attack meets Computer Vision: A Decade Survey","created":"2022-09-30","cited_by_count":67,"area":null},{"id":"2311.16194","title":"BadCLIP: Trigger-Aware Prompt Learning for Backdoor Attacks on CLIP","created":"2023-11-26","cited_by_count":67,"area":"mllm"},{"id":"2304.11823","title":"Enhancing Fine-Tuning Based Backdoor Defense with Sharpness-Aware Minimization","created":"2023-04-24","cited_by_count":64,"area":null},{"id":"2303.03323","title":"CleanCLIP: Mitigating Data Poisoning Attacks in Multimodal Contrastive Learning","created":"2023-03-06","cited_by_count":63,"area":"safety"},{"id":"2408.12798","title":"BackdoorLLM: A Comprehensive Benchmark for Backdoor Attacks and Defenses on Large Language Models","created":"2024-08-23","cited_by_count":50,"area":"safety"},{"id":"2303.18191","title":"Detecting Backdoors During the Inference Stage Based on Corruption Robustness Consistency","created":"2023-03-27","cited_by_count":46,"area":null},{"id":"2303.12175","title":"Black-box Backdoor Defense via Zero-shot Image Purification","created":"2023-03-21","cited_by_count":45,"area":null},{"id":"2303.12993","title":"Backdoor Defense via Adaptively Splitting Poisoned Dataset","created":"2023-03-23","cited_by_count":43,"area":null},{"id":"2307.10562","title":"Shared Adversarial Unlearning: Backdoor Mitigation by Unlearning Shared Adversarial Examples","created":"2023-07-20","cited_by_count":43,"area":null},{"id":"2212.09067","title":"Fine-Tuning Is All You Need to Mitigate Backdoor Attacks","created":"2022-12-18","cited_by_count":38,"area":null},{"id":"2306.16697","title":"Neural Polarizer: A Lightweight and Effective Backdoor Defense via Purifying Poisoned Features","created":"2023-06-29","cited_by_count":37,"area":null},{"id":"2402.11473","title":"Poisoned Forgery Face: Towards Backdoor Attacks on Face Forgery Detection","created":"2024-02-18","cited_by_count":35,"area":null},{"id":"2310.01875","title":"Towards Stable Backdoor Purification through Feature Shift Tuning","created":"2023-10-03","cited_by_count":32,"area":null},{"id":"2405.16134","title":"Breaking the False Sense of Security in Backdoor Defense through Re-Activation Attack","created":"2024-05-25","cited_by_count":28,"area":null},{"id":"2302.01762","title":"BackdoorBox: A Python Toolbox for Backdoor Learning","created":"2023-02-01","cited_by_count":26,"area":null},{"id":"2402.19334","title":"Here's a Free Lunch: Sanitizing Backdoored Models with Model Merge","created":"2024-02-29","cited_by_count":26,"area":null},{"id":"2405.05553","title":"Towards Robust Physical-world Backdoor Attacks on Lane Detection","created":"2024-05-09","cited_by_count":26,"area":null},{"id":"2208.08052","title":"Imperceptible and Robust Backdoor Attack in 3D Point Cloud","created":"2022-08-17","cited_by_count":25,"area":null},{"id":"2409.05294","title":"TERD: A Unified Framework for Safeguarding Diffusion Models Against Backdoors","created":"2024-09-09","cited_by_count":25,"area":"diffusion"},{"id":"2502.12202","title":"To Think or Not to Think: Exploring the Unthinking Vulnerability in Large Reasoning Models","created":"2025-02-16","cited_by_count":22,"area":"lr"},{"id":"2302.10894","title":"Red Teaming Deep Neural Networks with Feature Synthesis Tools","created":"2023-02-08","cited_by_count":21,"area":"safety"},{"id":"2302.09457","title":"Attacks in Adversarial Machine Learning: A Systematic Survey from the Life-cycle Perspective","created":"2023-02-19","cited_by_count":20,"area":null},{"id":"2209.09841","title":"Exploring Inconsistent Knowledge Distillation for Object Detection with Data Augmentation","created":"2022-09-20","cited_by_count":19,"area":"distill"},{"id":"2407.19845","title":"BackdoorBench: A Comprehensive Benchmark and Analysis of Backdoor Learning","created":"2024-07-29","cited_by_count":19,"area":null},{"id":"2405.12725","title":"Nearest is Not Dearest: Towards Practical Defense against Quantization-conditioned Backdoor Attacks","created":"2024-05-21","cited_by_count":18,"area":null},{"id":"2503.16023","title":"BadToken: Token-level Backdoor Attacks to Multi-modal Large Language Models","created":"2025-03-20","cited_by_count":17,"area":"safety"},{"id":"2302.01677","title":"Revisiting Personalized Federated Learning: Robustness Against Backdoor Attacks","created":"2023-02-03","cited_by_count":16,"area":null},{"id":"2312.13053","title":"Quantifying Bias in Text-to-Image Generative Models","created":"2023-12-20","cited_by_count":16,"area":null},{"id":"2410.13910","title":"Mitigating the Backdoor Effect for Multi-Task Model Merging via Safety-Aware Subspace","created":"2024-10-17","cited_by_count":14,"area":null},{"id":"2207.12405","title":"Versatile Weight Attack via Flipping Limited Bits","created":"2022-07-25","cited_by_count":13,"area":"safety"},{"id":"2212.04687","title":"Selective Amnesia: On Efficient, High-Fidelity and Blind Suppression of Backdoor Effects in Trojaned Machine Learning Models","created":"2022-12-09","cited_by_count":11,"area":null},{"id":"2405.16112","title":"Mitigating Backdoor Attack by Injecting Proactive Defensive Backdoor","created":"2024-05-25","cited_by_count":11,"area":null},{"id":"2305.14851","title":"Sharpness-Aware Data Poisoning Attack","created":"2023-05-24","cited_by_count":10,"area":"safety"},{"id":"2505.17601","title":"Revisiting Backdoor Attacks on LLMs: A Stealthy and Practical Poisoning Framework via Harmless Inputs","created":"2025-05-23","cited_by_count":10,"area":"safety"},{"id":"2507.05113","title":"CLIP-Guided Backdoor Defense through Entropy-Based Poisoned Dataset Separation","created":"2025-07-07","cited_by_count":10,"area":"safety"},{"id":"2306.00816","title":"Versatile Backdoor Attack with Visible, Semantic, Sample-Specific, and Compatible Triggers","created":"2023-06-01","cited_by_count":9,"area":null},{"id":"2309.16211","title":"VDC: Versatile Data Cleanser based on Visual-Linguistic Inconsistency by Multimodal Large Language Models","created":"2023-09-28","cited_by_count":9,"area":"mllm"},{"id":"2312.11057","title":"DataElixir: Purifying Poisoned Dataset to Mitigate Backdoor Attacks via Diffusion Models","created":"2023-12-18","cited_by_count":9,"area":"diffusion"},{"id":"2404.13518","title":"Reliable Model Watermarking: Defending Against Theft without Compromising on Evasion","created":"2024-04-21","cited_by_count":9,"area":"safety"},{"id":"2409.00399","title":"Rethinking Backdoor Detection Evaluation for Language Models","created":"2024-08-31","cited_by_count":9,"area":"lr"},{"id":"2412.02454","title":"Gracefully Filtering Backdoor Samples for Generative Large Language Models without Retraining","created":"2024-12-03","cited_by_count":9,"area":"safety"},{"id":"2402.12026","title":"Acquiring Clean Language Models from Backdoor Poisoned Datasets by Downscaling Frequency Space","created":"2024-02-19","cited_by_count":8,"area":"safety"},{"id":"2502.11798","title":"BackdoorDM: A Comprehensive Benchmark for Backdoor Learning on Diffusion Model","created":"2025-02-17","cited_by_count":8,"area":"diffusion"},{"id":"2302.14500","title":"FreeEagle: Detecting Complex Neural Trojans in Data-Free Cases","created":"2023-02-28","cited_by_count":7,"area":"safety"},{"id":"2308.03258","title":"APBench: A Unified Benchmark for Availability Poisoning Attacks and Defenses","created":"2023-08-07","cited_by_count":7,"area":null},{"id":"2410.09838","title":"Uncovering, Explaining, and Mitigating the Superficial Safety of Backdoor Defense","created":"2024-10-13","cited_by_count":7,"area":null},{"id":"2501.05928","title":"Towards Backdoor Stealthiness in Model Parameter Space","created":"2025-01-10","cited_by_count":7,"area":null},{"id":"2511.16709","title":"AutoBackdoor: Automating Backdoor Attacks via LLM Agents","created":"2025-11-20","cited_by_count":7,"area":"agents"},{"id":"2308.15605","title":"Benchmarks for Detecting Measurement Tampering","created":"2023-08-29","cited_by_count":6,"area":null},{"id":"2405.20291","title":"Unveiling and Mitigating Backdoor Vulnerabilities based on Unlearning Weight Changes and Backdoor Activeness","created":"2024-05-30","cited_by_count":6,"area":null},{"id":"2406.06207","title":"Lurking in the shadows: Unveiling Stealthy Backdoor Attacks against Personalized Federated Learning","created":"2024-06-10","cited_by_count":6,"area":null},{"id":"2306.12111","title":"A Comprehensive Study on the Robustness of Image Classification and Object Detection in Remote Sensing: Surveying and Benchmarking","created":"2023-06-21","cited_by_count":5,"area":null},{"id":"2307.07328","title":"Boosting Backdoor Attack with A Learnable Poisoning Sample Selection Strategy","created":"2023-07-14","cited_by_count":5,"area":null},{"id":"2405.14781","title":"Unified Neural Backdoor Removal with Only Few Clean Samples through Unlearning and Relearning","created":"2024-05-23","cited_by_count":5,"area":null},{"id":"2409.03741","title":"Understanding Data Importance in Machine Learning Attacks: Does Valuable Data Pose Greater Harm?","created":"2024-09-05","cited_by_count":5,"area":"safety"},{"id":"2412.11471","title":"TrapFlow: Controllable Website Fingerprinting Defense via Dynamic Backdoor Learning","created":"2024-12-16","cited_by_count":5,"area":null},{"id":"2503.03944","title":"GuardDoor: Safeguarding Against Malicious Diffusion Editing via Protective Backdoors","created":"2025-03-05","cited_by_count":5,"area":"diffusion"},{"id":"2507.04119","title":"When Data-Free Knowledge Distillation Meets Non-Transferable Teacher: Escaping Out-of-Distribution Trap is All You Need","created":"2025-07-05","cited_by_count":5,"area":"distill"},{"id":"2511.18921","title":"BackdoorVLM: A Benchmark for Backdoor Attacks and Defenses on Vision-Language Models","created":"2025-11-24","cited_by_count":5,"area":"mllm"},{"id":"2604.12359","title":"Compiling Activation Steering into Weights via Null-Space Constraints for Stealthy Backdoors","created":"2026-04-14","cited_by_count":4,"area":"alignment"},{"id":"2303.10594","title":"AdaptGuard: Defending Against Universal Attacks for Model Adaptation","created":"2023-03-19","cited_by_count":3,"area":null},{"id":"2307.11565","title":"Adversarial Feature Map Pruning for Backdoor","created":"2023-07-21","cited_by_count":3,"area":null},{"id":"2402.17465","title":"Model X-ray:Detecting Backdoored Models via Decision Boundary","created":"2024-02-27","cited_by_count":3,"area":null},{"id":"2403.14641","title":"Testing autonomous vehicles and AI: perspectives and challenges from cybersecurity, transparency, robustness and fairness","created":"2024-02-21","cited_by_count":3,"area":null},{"id":"2411.11006","title":"BackdoorMBTI: A Backdoor Learning Multimodal Benchmark Tool Kit for Backdoor Defense Evaluation","created":"2024-11-17","cited_by_count":3,"area":"safety"},{"id":"2411.11525","title":"Reliable Poisoned Sample Detection against Backdoor Attacks Enhanced by Sharpness Aware Minimization","created":"2024-11-18","cited_by_count":3,"area":null},{"id":"2501.17151","title":"Scanning Trojaned Models Using Out-of-Distribution Samples","created":"2025-01-28","cited_by_count":3,"area":"safety"},{"id":"2601.19448","title":"From Internal Diagnosis to External Auditing: A VLM-Driven Paradigm for Data-Free Online Backdoor Defense","created":"2026-01-27","cited_by_count":3,"area":"safety"},{"id":"2602.16984","title":"Fundamental Limits of Black-Box Safety Evaluation: Information-Theoretic and Computational Barriers from Latent Context Conditioning","created":"2026-02-19","cited_by_count":3,"area":null},{"id":"2402.16926","title":"On the (In)feasibility of ML Backdoor Detection as an Hypothesis Testing Problem","created":"2024-02-26","cited_by_count":2,"area":null},{"id":"2409.00863","title":"Fisher Information guided Purification against Backdoor Attacks","created":"2024-09-01","cited_by_count":2,"area":null},{"id":"2410.12955","title":"Long-Tailed Backdoor Attack Using Dynamic Data Augmentation Operations","created":"2024-10-16","cited_by_count":2,"area":null},{"id":"2412.03441","title":"PBP: Post-training Backdoor Purification for Malware Classifiers","created":"2024-12-04","cited_by_count":2,"area":null},{"id":"2504.17300","title":"The Ultimate Cookbook for Invisible Poison: Crafting Subtle Clean-Label Text Backdoors with Style Attributes","created":"2025-04-24","cited_by_count":2,"area":"safety"},{"id":"2505.08255","title":"Where the Devil Hides: Deepfake Detectors Can No Longer Be Trusted","created":"2025-05-13","cited_by_count":2,"area":"safety"},{"id":"2506.00175","title":"Who Gets Credit or Blame? Attributing Accountability in Modern AI Systems","created":"2025-05-30","cited_by_count":2,"area":null},{"id":"2506.01849","title":"Trojan Horse Hunt in Time Series Forecasting for Space Operations","created":"2025-06-02","cited_by_count":2,"area":null},{"id":"2506.07467","title":"Circumventing Backdoor Space via Weight Symmetry","created":"2025-06-09","cited_by_count":2,"area":null},{"id":"2602.17168","title":"BadCLIP++: Stealthy and Persistent Backdoors in Multimodal Contrastive Learning","created":"2026-02-19","cited_by_count":2,"area":"safety"},{"id":"2603.29328","title":"Beyond Corner Patches: Semantics-Aware Backdoor Attack in Federated Learning","created":"2026-03-31","cited_by_count":2,"area":null},{"id":"2212.09979","title":"Flareon: Stealthy any2any Backdoor Injection via Poisoned Augmentation","created":"2022-12-20","cited_by_count":1,"area":null},{"id":"2305.09671","title":"Pick your Poison: Undetectability versus Robustness in Data Poisoning Attacks","created":"2023-05-07","cited_by_count":1,"area":"safety"},{"id":"2405.03918","title":"Unlearning Backdoor Attacks through Gradient-Based Model Pruning","created":"2024-05-07","cited_by_count":1,"area":null},{"id":"2409.12072","title":"PAD-FT: A Lightweight Defense for Backdoor Attacks via Data Purification and Fine-Tuning","created":"2024-09-18","cited_by_count":1,"area":null},{"id":"2410.19427","title":"Expose Before You Defend: Unifying and Enhancing Backdoor Defenses via Exposed Models","created":"2024-10-25","cited_by_count":1,"area":null},{"id":"2411.12914","title":"Trojan Cleansing with Neural Collapse","created":"2024-11-19","cited_by_count":1,"area":"safety"},{"id":"2412.20476","title":"Cut the Deadwood Out: Backdoor Purification via Guided Module Substitution","created":"2024-12-29","cited_by_count":1,"area":null},{"id":"2502.18520","title":"Class-Conditional Neural Polarizer: A Lightweight and Effective Backdoor Defense by Purifying Poisoned Features","created":"2025-02-23","cited_by_count":1,"area":null},{"id":"2504.00170","title":"Backdoor Detection through Replicated Execution of Outsourced Training","created":"2025-03-31","cited_by_count":1,"area":null},{"id":"2507.04903","title":"BackFed: An Efficient & Standardized Benchmark Suite for Backdoor Attacks in Federated Learning","created":"2025-07-07","cited_by_count":1,"area":null},{"id":"2507.22813","title":"DISTIL: Data-Free Inversion of Suspicious Trojan Inputs via Latent Diffusion","created":"2025-07-30","cited_by_count":1,"area":"diffusion"},{"id":"2509.15497","title":"Backdoor Mitigation via Invertible Pruning Masks","created":"2025-09-19","cited_by_count":1,"area":null},{"id":"2510.14845","title":"Backdoor Unlearning by Linear Task Decomposition","created":"2025-10-16","cited_by_count":1,"area":null}],"next_cursor":"100"}