Self-supervised and multimodal vision encoders learn strong visual representations that are widely adopted in downstream vision tasks and large vision-language models (LVLMs). However, downstream users often rely on third-party pretrained encoders with uncertain provenance, exposing them to backdoor attacks. In this work, we propose BackdoorIDS, a simple yet effective zero-shot, inference-time backdoor samples detection method for pretrained vision encoders. BackdoorIDS is motivated by two observations: Attention Hijacking and Restoration. Under progressive input masking, a backdoored image initially concentrates attention on malicious trigger features. Once the masking ratio exceeds the trigger's robustness threshold, the trigger is deactivated, and attention rapidly shifts to benign content. This transition induces a pronounced change in the image embedding, whereas embeddings of clean images evolve more smoothly across masking progress. BackdoorIDS operationalizes this signal by extracting an embedding sequence along the masking trajectory and applying density-based clustering such as DBSCAN. An input is flagged as backdoored if its embedding sequence forms more than one cluster. Extensive experiments show that BackdoorIDS consistently outperforms existing defenses across diverse attack types, datasets, and model families. Notably, it is a plug-and-play approach that requires no retraining and operates fully zero-shot at inference time, making it compatible with a wide range of encoder architectures, including CNNs, ViTs, CLIP, and LLaVA-1.5.

Large language models (LLMs) remain vulnerable to jailbreak attacks that bypass safety guardrails to elicit harmful outputs. Existing approaches overwhelmingly operate within the prompt-optimization paradigm: whether through traditional algorithmic search or recent agent-based workflows, the resulting prompts typically retain malicious semantic signals that modern guardrails are primed to detect. In contrast, we identify a deeper, largely overlooked vulnerability stemming from the highly interconnected nature of an LLM's internal knowledge. This structure allows harmful objectives to be realized by weaving together sequences of benign sub-queries, each of which individually evades detection. To exploit this loophole, we introduce the Correlated Knowledge Attack Agent (CKA-Agent), a dynamic framework that reframes jailbreaking as an adaptive, tree-structured exploration of the target model's knowledge base. The CKA-Agent issues locally innocuous queries, uses model responses to guide exploration across multiple paths, and ultimately assembles the aggregated information to achieve the original harmful objective. Evaluated across state-of-the-art commercial LLMs (Gemini2.5-Flash/Pro, GPT-oss-120B, Claude-Haiku-4.5), CKA-Agent consistently achieves over 95% success rates even against strong guardrails, underscoring the severity of this vulnerability and the urgent need for defenses against such knowledge-decomposition attacks. Our codes are available at https://github.com/Graph-COM/CKA-Agent.

Fine-tuned Large Language Models (LLMs) are vulnerable to backdoor attacks through data poisoning, yet the internal mechanisms governing these attacks remain a black box. Previous research on interpretability for LLM safety tends to focus on alignment, jailbreak, and hallucination, but overlooks backdoor mechanisms, making it difficult to understand and fully eliminate the backdoor threat. In this paper, aiming to bridge this gap, we explore the interpretable mechanisms of LLM backdoors through Backdoor Attribution (BkdAttr), a tripartite causal analysis framework. We first introduce the Backdoor Probe that proves the existence of learnable backdoor features encoded within the representations. Building on this insight, we further develop Backdoor Attention Head Attribution (BAHA), efficiently pinpointing the specific attention heads responsible for processing these features. Our primary experiments reveals these heads are relatively sparse; ablating a minimal \textbf{$\sim$ 3%} of total heads is sufficient to reduce the Attack Success Rate (ASR) by \textbf{over 90%}. More importantly, we further employ these findings to construct the Backdoor Vector derived from these attributed heads as a master controller for the backdoor. Through only \textbf{1-point} intervention on \textbf{single} representation, the vector can either boost ASR up to \textbf{$\sim$ 100% ($\uparrow$)} on clean inputs, or completely neutralize backdoor, suppressing ASR down to \textbf{$\sim$ 0% ($\downarrow$)} on triggered inputs. In conclusion, our work pioneers the exploration of mechanistic interpretability in LLM backdoors, demonstrating a powerful method for backdoor control and revealing actionable insights for the community.

Split Learning (SL) is a distributed learning approach that enables resource-constrained clients to collaboratively train deep neural networks (DNNs) by offloading most layers to a central server while keeping in- and output layers on the client-side. This setup enables SL to leverage server computation capacities without sharing data, making it highly effective in resource-constrained environments dealing with sensitive data. However, the distributed nature enables malicious clients to manipulate the training process. By sending poisoned intermediate gradients, they can inject backdoors into the shared DNN. Existing defenses are limited by often focusing on server-side protection and introducing additional overhead for the server. A significant challenge for client-side defenses is enforcing malicious clients to correctly execute the defense algorithm. We present ZORRO, a private, verifiable, and robust SL defense scheme. Through our novel design and application of interactive zero-knowledge proofs (ZKPs), clients prove their correct execution of a client-located defense algorithm, resulting in proofs of computational integrity attesting to the benign nature of locally trained DNN portions. Leveraging the frequency representation of model partitions enables ZORRO to conduct an in-depth inspection of the locally trained models in an untrusted environment, ensuring that each client forwards a benign checkpoint to its succeeding client. In our extensive evaluation, covering different model architectures as well as various attack strategies and data scenarios, we show ZORRO's effectiveness, as it reduces the attack success rate to less than 6\% while causing even for models storing \numprint{1000000} parameters on the client-side an overhead of less than 10 seconds.

With the proliferation of LLM-driven multi-agent systems (MAS), the security of Web links has become a critical concern. Once MAS is induced to trust a malicious link, attackers can use it as a springboard to expand the attack surface. In this paper, we propose Web Fraud Attacks, a novel type of attack manipulating unique structures of web links to deceive MAS. We design 12 representative attack variants that encompass various methods, such as homoglyph deception, sub-directory nesting, and parameter obfuscation. Through extensive experiments on these attack vectors, we demonstrate that Web fraud attacks not only exhibit significant destructive potential across different MAS architectures but also possess a distinct advantage in evasion: they circumvent the need for complex input design, lowering the threshold for attacks significantly. These results underscore the importance of addressing Web fraud attacks, providing new insights into MAS safety. Our code is available at https://github.com/JiangYingEr/Web-Fraud-Attack-in-MAS.

Minimizing computational overhead in time-series classification, particularly in deep learning models, presents a significant challenge due to the high complexity of model architectures and the large volume of sequential data that must be processed in real time. This challenge is further compounded by adversarial attacks, emphasizing the need for resilient methods that ensure robust performance and efficient model selection. To address this challenge, we propose ReLATE+, a comprehensive framework that detects and classifies adversarial attacks, adaptively selects deep learning models based on dataset-level similarity, and thus substantially reduces retraining costs relative to conventional methods that do not leverage prior knowledge, while maintaining strong performance. ReLATE+ first checks whether the incoming data is adversarial and, if so, classifies the attack type, using this insight to identify a similar dataset from a repository and enable the reuse of the best-performing associated model. This approach ensures strong performance while reducing the need for retraining, and it generalizes well across different domains with varying data distributions and feature spaces. Experiments show that ReLATE+ reduces computational overhead by an average of 77.68%, enhancing adversarial resilience and streamlining robust model selection, all without sacrificing performance, within 2.02% of Oracle.