Skip to content

Scan record

Everything the site collected, newest first by collection time (Beijing time), back seven days.

Nov 1740 loaded

  1. Conv-Reluplex: A Verification Framework for Convolutional Neural Networks

    PaperarXiv: Adversarial Attacks, Backdoors & PoisoningThe item you came from

  2. Forging a Bearer to take over a LiteLLM MCP session: reproduction and analysis of the OAuth2 passthrough authentication bypass

    NewsXianzhi Community (Aliyun)Event

  3. NATO AI cyber defense uses up to 300B OpenAI tokens per month

    NewsHelp Net Security AIEvent

  4. Google finds AI agents writing and launching exploits without human involvement

    NewsHelp Net Security AIEvent

  5. CuRSE-Bench: A benchmark for probing cultural blind spots in text-to-image safety evaluation

    PaperarXiv: Dangerous Capabilities & Safety Evals

  6. Between Rigor and Reality: How AI Safety Benchmark Developers Understand Benchmark Use and Maintenance

    PaperarXiv: Alignment, Deception & Oversight

  7. Manipulating Bayesian inference and decision-making through data poisoning

    PaperarXiv: 其他 AI 安全论文

  8. Robustness as a Transferable Skill: Cross-Task Adversarial Defense for Vision-Language-Action Policies

    PaperarXiv: 其他 AI 安全论文

  9. The Paradox of Provenance and Privacy: Can Private Cloud Compute (PCC) Prove That Apple Reference Images Are Not Used for Learning?

    PaperarXiv: 其他 AI 安全论文

  10. When tool names are not identity: measuring provider misbinding in MCP-based AI agents

    PaperarXiv: 其他 AI 安全论文

  11. Optimistic verifiers get gamed: reward hacking under exact process oracles

    PaperarXiv: 其他 AI 安全论文

  12. Similar in Safety, Silent in Propagation: Limitations of Text-Similarity Guardrails for AI Agent Security

    PaperarXiv: 其他 AI 安全论文

  13. Evaluating Authorization and Trust Boundaries in Agent Orchestration Frameworks

    PaperarXiv: 其他 AI 安全论文

  14. Jev-as-a-Reward: Validating a Probabilistic Safety Judge for Reinforcement Learning

    PaperarXiv: 其他 AI 安全论文

  15. Experience Rot: Experienced agents fail where novices succeed

    PaperarXiv: 其他 AI 安全论文

  16. Which AI Use Cases Are Worth Piloting? A Risk and Value Framework for Funding, Scoping and Scaling AI

    PaperarXiv: 其他 AI 安全论文

  17. Tight PAC-Bayes Generalization Guarantees for LLM Judges

    PaperarXiv: 其他 AI 安全论文

  18. STA: A Split Temporal Adversarial Method for Evaluating the Robustness of Multimodal Reinforcement Learning Driving Policies

    PaperarXiv: 其他 AI 安全论文

  19. LEAF-OPD: Learning to Unlearn from Privileged Context via On-Policy Distillation

    PaperarXiv: 其他 AI 安全论文

  20. Who Bears the Excess Authority? Attributing AI Agent Behavior to the Delegating Party

    PaperarXiv: 其他 AI 安全论文

  21. Safety-Preserving Fine-Tuning for Medical LLMs via Bi-Level Data Selection

    PaperarXiv: 其他 AI 安全论文

  22. Counterfactual feature steering for identity unlearning with adjacency preservation

    PaperarXiv: 其他 AI 安全论文

  23. Example-based spatial guidance: training-free concept erasure in diffusion models

    PaperarXiv: 其他 AI 安全论文

  24. GPromptShield: Enhancing the Resilience of Graph Prompt Tuning under Adversarial Attacks

    PaperarXiv: 其他 AI 安全论文

  25. SmoothShift: Disrupting Visual Adversarial Attacks against LVLMs at Test Time

    PaperarXiv: 其他 AI 安全论文

  26. Model Immunity Is More Than Condition Number: The Importance of Plateau Regions

    PaperarXiv: 其他 AI 安全论文

  27. Designing Deceptive Model Organisms That Successfully Evade Auditing

    PaperarXiv: 其他 AI 安全论文

  28. Interpretable Federated Learning via Global-Local Attribution Alignment

    PaperarXiv: 其他 AI 安全论文

  29. Abstention directions in LLMs: mechanistic interpretability of abstention in query ambiguity, knowledge boundaries, and safety alignment

    PaperarXiv: 其他 AI 安全论文

  30. Refusal Register Can Be Causally Manipulated: Decomposing the Refusal Direction of Gemma-2-9B-IT into Category-Aligned Components

    PaperarXiv: 其他 AI 安全论文

  31. Where the Evidence Hides: Measuring Privacy and Diagnosability of Exported Agent Trajectories

    PaperarXiv: 其他 AI 安全论文

  32. Where the saved time goes: limitations of hiding cross-tenant cache hits in LLM serving

    PaperarXiv: 其他 AI 安全论文

  33. Protecting safety in reasoning training with low-rank Fisher regularization

    PaperarXiv: 其他 AI 安全论文

  34. Who Can You Hide Behind? Author-Conditioned Style Transfer and Evasion of a Commercial AI-Text Detector

    PaperarXiv: 其他 AI 安全论文

  35. Risk-Certified Prompt Optimization for LLM-as-a-Judge Guardrails

    PaperarXiv: 其他 AI 安全论文

  36. HEAMS: Holistic Evaluation of Agent Memory Systems

    PaperarXiv: 其他 AI 安全论文

  37. Can AI judge you? An empirical audit of the deployment readiness of LLM-assisted hiring systems

    PaperarXiv: 其他 AI 安全论文

  38. "Don't leave the task unfinished": persistence-driven violations under silent supervision

    PaperarXiv: 其他 AI 安全论文

  39. HalluProbe: using VLM hallucinations for backdoor detection

    PaperarXiv: 其他 AI 安全论文

  40. Where in multimodal LM-derived models can Rank-1 DiM refusal be intervened on?

    PaperarXiv: 其他 AI 安全论文