When Claws Remember but Do Not Tell: Stealthy Memory Injection in Persistent Personal Agents

Persistent personal agents combine long-term memory with access to users' external environments, enabling personalized foreground assistance and proactive background execution. This integration also creates a new path to compromise: untrusted external content can be silently written into persistent memory and later reused as trusted state. We study this threat as stealth memory injection, in which a remote black-box adversary delivers a single email payload that must induce the agent to write poisoned memory, stay hidden in the agent's response to the user, and affect future behavior. We introduce WhisperBench, a 108-case benchmark spanning five risk categories and both fact and preference poisoning. Built on a real IMAP/SMTP workflow and an authentic email agent skill, it enables full-cycle evaluation of stealth memory injection attacks. To enable this black-box attack under single-email delivery and without runtime feedback, we propose MemGhost, a one-shot payload generation framework. MemGhost uses an environment proxy to emulate persistent-agent execution and an objective proxy to convert memory adoption and conversational stealth into dense rubric-based rewards, then trains the attacker policy with supervised fine-tuning and reinforcement learning. Across 56 held-out test cases, MemGhost achieves 87.5% end-to-end success on OpenClaw with GPT-5.4 and 71.4% on Claude Code SDK with Sonnet 4.6. It also transfers across personal-agent architectures (NanoClaw and Hermes Agent) and memory backends (filesystem and vector-based Mem0), and remains effective against input-level, model-level, and system-level defenses. These results suggest that persistent memory can turn ordinary external processing into a practical pathway for long-term agent compromise.

Mind Your HEARTBEAT! Claw Background Execution Inherently Enables Silent Memory Pollution

We identify a critical security vulnerability in mainstream Claw personal AI agents: untrusted content encountered during heartbeat-driven background execution can silently pollute agent memory and subsequently influence user-facing behavior without the user's awareness. This vulnerability arises from an architectural design shared across the Claw ecosystem: heartbeat background execution runs in the same session as user-facing conversation, so content ingested from any external source monitored in the background (including email, message channels, news feeds, code repositories, and social platforms) can enter the same memory context used for foreground interaction, often with limited user visibility and without clear source provenance. We formalize this process as an Exposure (E) $\rightarrow$ Memory (M) $\rightarrow$ Behavior (B) pathway: misinformation encountered during heartbeat execution enters the agent's short-term session context, potentially gets written into long-term memory, and later shapes downstream user-facing behavior. We instantiate this pathway in an agent-native social setting using MissClaw, a controlled research replica of Moltbook. We find that (1) social credibility cues, especially perceived consensus, are the dominant driver of short-term behavioral influence, with misleading rates up to 61%; (2) routine memory-saving behavior can promote short-term pollution into durable long-term memory at rates up to 91%, with cross-session behavioral influence reaching 76%; (3) under naturalistic browsing with content dilution and context pruning, pollution still crosses session boundaries. Overall, prompt injection is not required: ordinary social misinformation is sufficient to silently shape agent memory and behavior under heartbeat-driven background execution.

P2P: A Poison-to-Poison Remedy for Reliable Backdoor Defense in LLMs

During fine-tuning, large language models (LLMs) are increasingly vulnerable to data-poisoning backdoor attacks, which compromise their reliability and trustworthiness. However, existing defense strategies suffer from limited generalization: they only work on specific attack types or task settings. In this study, we propose Poison-to-Poison (P2P), a general and effective backdoor defense algorithm. P2P injects benign triggers with safe alternative labels into a subset of training samples and fine-tunes the model on this re-poisoned dataset by leveraging prompt-based learning. This enforces the model to associate trigger-induced representations with safe outputs, thereby overriding the effects of original malicious triggers. Thanks to this robust and generalizable trigger-based fine-tuning, P2P is effective across task settings and attack types. Theoretically and empirically, we show that P2P can neutralize malicious backdoors while preserving task performance. We conduct extensive experiments on classification, mathematical reasoning, and summary generation tasks, involving multiple state-of-the-art LLMs. The results demonstrate that our P2P algorithm significantly reduces the attack success rate compared with baseline models. We hope that the P2P can serve as a guideline for defending against backdoor attacks and foster the development of a secure and trustworthy LLM community.

Three Minds, One Legend: Jailbreak Large Reasoning Model with Adaptive Stacked Ciphers

Recently, Large Reasoning Models (LRMs) have demonstrated superior logical capabilities compared to traditional Large Language Models (LLMs), gaining significant attention. Despite their impressive performance, the potential for stronger reasoning abilities to introduce more severe security vulnerabilities remains largely underexplored. Existing jailbreak methods often struggle to balance effectiveness with robustness against adaptive safety mechanisms. In this work, we propose SEAL, a novel jailbreak attack that targets LRMs through an adaptive encryption pipeline designed to override their reasoning processes and evade potential adaptive alignment. Specifically, SEAL introduces a stacked encryption approach that combines multiple ciphers to overwhelm the models reasoning capabilities, effectively bypassing built-in safety mechanisms. To further prevent LRMs from developing countermeasures, we incorporate two dynamic strategies - random and adaptive - that adjust the cipher length, order, and combination. Extensive experiments on real-world reasoning models, including DeepSeek-R1, Claude Sonnet, and OpenAI GPT-o4, validate the effectiveness of our approach. Notably, SEAL achieves an attack success rate of 80.8% on GPT o4-mini, outperforming state-of-the-art baselines by a significant margin of 27.2%. Warning: This paper contains examples of inappropriate, offensive, and harmful content.

Jailbreaking the Text-to-Video Generative Models

Text-to-video generative models have achieved significant progress, driven by the rapid advancements in diffusion models, with notable examples including Pika, Luma, Kling, and Sora. Despite their remarkable generation ability, their vulnerability to jailbreak attack, i.e. to generate unsafe content, including pornography, violence, and discrimination, raises serious safety concerns. Existing efforts, such as T2VSafetyBench, have provided valuable benchmarks for evaluating the safety of text-to-video models against unsafe prompts but lack systematic studies for exploiting their vulnerabilities effectively. In this paper, we propose the \textit{first} optimization-based jailbreak attack against text-to-video models, which is specifically designed. Our approach formulates the prompt generation task as an optimization problem with three key objectives: (1) maximizing the semantic similarity between the input and generated prompts, (2) ensuring that the generated prompts can evade the safety filter of the text-to-video model, and (3) maximizing the semantic similarity between the generated videos and the original input prompts. To further enhance the robustness of the generated prompts, we introduce a prompt mutation strategy that creates multiple prompt variants in each iteration, selecting the most effective one based on the averaged score. This strategy not only improves the attack success rate but also boosts the semantic relevance of the generated video. We conduct extensive experiments across multiple text-to-video models, including Open-Sora, Pika, Luma, and Kling. The results demonstrate that our method not only achieves a higher attack success rate compared to baseline methods but also generates videos with greater semantic similarity to the original input prompts.

Inception: Jailbreak the Memory Mechanism of Text-to-Image Generation Systems

Currently, the memory mechanism has been widely and successfully exploited in online text-to-image (T2I) generation systems ($e.g.$, DALL$\cdot$E 3) for alleviating the growing tokenization burden and capturing key information in multi-turn interactions. Despite its practicality, its security analyses have fallen far behind. In this paper, we reveal that this mechanism exacerbates the risk of jailbreak attacks. Different from previous attacks that fuse the unsafe target prompt into one ultimate adversarial prompt, which can be easily detected or may generate non-unsafe images due to under- or over-optimization, we propose Inception, the first multi-turn jailbreak attack against the memory mechanism in real-world text-to-image generation systems. Inception embeds the malice at the inception of the chat session turn by turn, leveraging the mechanism that T2I generation systems retrieve key information in their memory. Specifically, Inception mainly consists of two modules. It first segments the unsafe prompt into chunks, which are subsequently fed to the system in multiple turns, serving as pseudo-gradients for directive optimization. Specifically, we develop a series of segmentation policies that ensure the images generated are semantically consistent with the target prompt. Secondly, after segmentation, to overcome the challenge of the inseparability of minimum unsafe words, we propose recursion, a strategy that makes minimum unsafe words subdivisible. Collectively, segmentation and recursion ensure that all the request prompts are benign but can lead to malicious outcomes. We conduct experiments on the real-world text-to-image generation system ($i.e.$, DALL$\cdot$E 3) to validate the effectiveness of Inception. The results indicate that Inception surpasses the state-of-the-art by a 14\% margin in attack success rate.

A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deployment

The remarkable success of Large Language Models (LLMs) has illuminated a promising pathway toward achieving Artificial General Intelligence for both academic and industrial communities, owing to their unprecedented performance across various applications. As LLMs continue to gain prominence in both research and commercial domains, their security and safety implications have become a growing concern, not only for researchers and corporations but also for every nation. Currently, existing surveys on LLM safety primarily focus on specific stages of the LLM lifecycle, e.g., deployment phase or fine-tuning phase, lacking a comprehensive understanding of the entire "lifechain" of LLMs. To address this gap, this paper introduces, for the first time, the concept of "full-stack" safety to systematically consider safety issues throughout the entire process of LLM training, deployment, and eventual commercialization. Compared to the off-the-shelf LLM safety surveys, our work demonstrates several distinctive advantages: (I) Comprehensive Perspective. We define the complete LLM lifecycle as encompassing data preparation, pre-training, post-training, deployment and final commercialization. To our knowledge, this represents the first safety survey to encompass the entire lifecycle of LLMs. (II) Extensive Literature Support. Our research is grounded in an exhaustive review of over 800+ papers, ensuring comprehensive coverage and systematic organization of security issues within a more holistic understanding. (III) Unique Insights. Through systematic literature analysis, we have developed reliable roadmaps and perspectives for each chapter. Our work identifies promising research directions, including safety in data generation, alignment techniques, model editing, and LLM-based agent systems. These insights provide valuable guidance for researchers pursuing future work in this field.

Mask Image Watermarking

We present MaskMark, a simple, efficient and flexible framework for image watermarking. MaskMark has two variants: MaskMark-D, which supports global watermark embedding, watermark localization, and local watermark extraction for applications such as tamper detection, and MaskMark-ED, which focuses on local watermark embedding and extraction with enhanced robustness in small regions, enabling localized image protection. Built upon the classical Encoder- Distortion-Decoder training paradigm, MaskMark-D introduces a simple masking mechanism during the decoding stage to support both global and local watermark extraction. A mask is applied to the watermarked image before extraction, allowing the decoder to focus on selected regions and learn local extraction. A localization module is also integrated into the decoder to identify watermark regions during inference, reducing interference from irrelevant content and improving accuracy. MaskMark-ED extends this design by incorporating the mask into the encoding stage as well, guiding the encoder to embed the watermark in designated local regions for enhanced robustness. Comprehensive experiments show that MaskMark achieves state-of-the-art performance in global watermark extraction, local watermark extraction, watermark localization, and multi-watermark embedding. It outperforms all existing baselines, including the recent leading model WAM for local watermarking, while preserving high visual quality of the watermarked images. MaskMark is also flexible, by adjusting the distortion layer, it can adapt to different robustness requirements with just a few steps of fine-tuning. Moreover, our approach is efficient and easy to optimize, requiring only 20 hours on a single A6000 GPU with just 1/15 the computational cost of WAM.

SAME: Sample Reconstruction Against Model Extraction Attacks

While deep learning models have shown significant performance across various domains, their deployment needs extensive resources and advanced computing infrastructure. As a solution, Machine Learning as a Service (MLaaS) has emerged, lowering the barriers for users to release or productize their deep learning models. However, previous studies have highlighted potential privacy and security concerns associated with MLaaS, and one primary threat is model extraction attacks. To address this, there are many defense solutions but they suffer from unrealistic assumptions and generalization issues, making them less practical for reliable protection. Driven by these limitations, we introduce a novel defense mechanism, SAME, based on the concept of sample reconstruction. This strategy imposes minimal prerequisites on the defender's capabilities, eliminating the need for auxiliary Out-of-Distribution (OOD) datasets, user query history, white-box model access, and additional intervention during model training. It is compatible with existing active defense methods. Our extensive experiments corroborate the superior efficacy of SAME over state-of-the-art solutions. Our code is available at https://github.com/xythink/SAME.