Skip to content

Scan record

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

Nov 1740 loaded

  1. Evaluating Authorization and Trust Boundaries in Agent Orchestration Frameworks

    PaperarXiv: 其他 AI 安全论文The item you came from

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

    PaperarXiv: 其他 AI 安全论文

  3. Experience Rot: Experienced agents fail where novices succeed

    PaperarXiv: 其他 AI 安全论文

  4. 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

  5. Tight PAC-Bayes Generalization Guarantees for LLM Judges

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

  10. Counterfactual feature steering for identity unlearning with adjacency preservation

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

  15. Designing Deceptive Model Organisms That Successfully Evade Auditing

    PaperarXiv: 其他 AI 安全论文

  16. Interpretable Federated Learning via Global-Local Attribution Alignment

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

  18. 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

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

  22. 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

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

    PaperarXiv: 其他 AI 安全论文

  24. HEAMS: Holistic Evaluation of Agent Memory Systems

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

  27. HalluProbe: using VLM hallucinations for backdoor detection

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

  29. Fine-Grained RLAIF for Visual Safety: Learning to Magnify Hazards for Multimodal Safety Alignment

    PaperarXiv: 其他 AI 安全论文

  30. When backdoors can be switched off: suppression-aware backdoor attacks on vision-language-action models

    PaperarXiv: 其他 AI 安全论文

  31. Building enterprise AI infrastructure: the evolution of the AI governance committee and the chief ethics officer role

    PaperarXiv: 其他 AI 安全论文

  32. BadGCL: Persistent Backdoor Attacks Against Graph Continual Learning Systems

    PaperarXiv: 其他 AI 安全论文

  33. Protecting the system while leaving users stranded: the limits of multi-agent authorization

    PaperarXiv: 其他 AI 安全论文

  34. Beyond Prompts: Runtime Contracts for Coding Agents

    PaperarXiv: 其他 AI 安全论文

  35. From facts to personas: interpretable persona unlearning in LLMs based on mixture of experts

    PaperarXiv: 其他 AI 安全论文

  36. Hardening agent network populations via multi-criteria competitive co-evolution: insights from wargame strategy optimization

    PaperarXiv: 其他 AI 安全论文

  37. Auditable Authorization for LLM Agents: Empirical Evidence Across Model, Harness, and Workflow Layers

    PaperarXiv: 其他 AI 安全论文

  38. Operating points of frozen guardrail probes fail under base model updates

    PaperarXiv: 其他 AI 安全论文

  39. On-Policy Distillation as a Backdoor Propagation Channel

    PaperarXiv: 其他 AI 安全论文

  40. A Bag of Concepts? Testing Role-Filler Bindings in the Verbalizable Workspace

    PaperarXiv: 其他 AI 安全论文