Skip to content

Scan record

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

Nov 1739 loaded

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

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

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

  5. Counterfactual feature steering for identity unlearning with adjacency preservation

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

  10. Designing Deceptive Model Organisms That Successfully Evade Auditing

    PaperarXiv: 其他 AI 安全论文

  11. Interpretable Federated Learning via Global-Local Attribution Alignment

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

  13. 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

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

  17. 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

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

    PaperarXiv: 其他 AI 安全论文

  19. HEAMS: Holistic Evaluation of Agent Memory Systems

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

  22. HalluProbe: using VLM hallucinations for backdoor detection

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

  26. 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

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

  29. Beyond Prompts: Runtime Contracts for Coding Agents

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

  34. On-Policy Distillation as a Backdoor Propagation Channel

    PaperarXiv: 其他 AI 安全论文

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

    PaperarXiv: 其他 AI 安全论文

  36. Agent-ZTA: A Chain-Aware Zero-Trust Model for Bidirectional Security State Propagation and Runtime Containment in Multi-Agent Delegation

    PaperarXiv: 其他 AI 安全论文

  37. MemeEconomy: Will LLM agents trade ethics for survival?

    PaperarXiv: 其他 AI 安全论文

  38. Same Evidence, Different Verdicts: How Delivery of CT Images and DICOM Metadata Changes VLM Quality Control Decisions

    PaperarXiv: 其他 AI 安全论文

  39. Quantum-Inspired Orthogonal Adapters for Certifiably Robust Medical Image Classification

    PaperarXiv: 其他 AI 安全论文