Scan record
Nov 1740 loaded
- Between the Lines: Contextual Investment Advice Detection in Multi-Turn Conversations
PaperarXiv: Jailbreaks, Prompt Injection, Poisoning & DefensesThe item you came from
- South Korean prosecutors seek three-year sentence in second instance of Chinese national graduate student deepfake sexual crime case
NewsYonhapEvent
- Guatemala launches multi-stakeholder discussions to jointly develop a national artificial intelligence strategy
Newsprensa.gob.gtAI updates
- Irene Liu on why California leads US frontier AI regulation: from the vetoed SB 1047 to SB 53
NewsThe EU AI Act NewsletterGovernance
- An overview of AI safety in California state politics
NewsLessWrongGovernance
- Rumored November 17 global blackout: a mix of distorted facts and fabricated videos
NewsRTVEEvent
- British actors will gain more digital replica rights under an AI protection agreement with producers
NewsThe Guardian · AIGovernance
- Backdoor trigger words are clearly readable in fine-tuned weights
PaperarXiv: Adversarial Attacks, Backdoors & Poisoning
- Conv-Reluplex: A Verification Framework for Convolutional Neural Networks
PaperarXiv: Adversarial Attacks, Backdoors & Poisoning
- Forging a Bearer to take over a LiteLLM MCP session: reproduction and analysis of the OAuth2 passthrough authentication bypass
NewsXianzhi Community (Aliyun)Event
- NATO AI cyber defense uses up to 300B OpenAI tokens per month
NewsHelp Net Security AIEvent
- Google finds AI agents writing and launching exploits without human involvement
NewsHelp Net Security AIEvent
- CuRSE-Bench: A benchmark for probing cultural blind spots in text-to-image safety evaluation
PaperarXiv: Dangerous Capabilities & Safety Evals
- Between Rigor and Reality: How AI Safety Benchmark Developers Understand Benchmark Use and Maintenance
PaperarXiv: Alignment, Deception & Oversight
- Manipulating Bayesian inference and decision-making through data poisoning
PaperarXiv: 其他 AI 安全论文
- Robustness as a Transferable Skill: Cross-Task Adversarial Defense for Vision-Language-Action Policies
PaperarXiv: 其他 AI 安全论文
- 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
- When tool names are not identity: measuring provider misbinding in MCP-based AI agents
PaperarXiv: 其他 AI 安全论文
- Optimistic verifiers get gamed: reward hacking under exact process oracles
PaperarXiv: 其他 AI 安全论文
- Similar in Safety, Silent in Propagation: Limitations of Text-Similarity Guardrails for AI Agent Security
PaperarXiv: 其他 AI 安全论文
- Evaluating Authorization and Trust Boundaries in Agent Orchestration Frameworks
PaperarXiv: 其他 AI 安全论文
- Jev-as-a-Reward: Validating a Probabilistic Safety Judge for Reinforcement Learning
PaperarXiv: 其他 AI 安全论文
- Experience Rot: Experienced agents fail where novices succeed
PaperarXiv: 其他 AI 安全论文
- 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
- Tight PAC-Bayes Generalization Guarantees for LLM Judges
PaperarXiv: 其他 AI 安全论文
- STA: A Split Temporal Adversarial Method for Evaluating the Robustness of Multimodal Reinforcement Learning Driving Policies
PaperarXiv: 其他 AI 安全论文
- LEAF-OPD: Learning to Unlearn from Privileged Context via On-Policy Distillation
PaperarXiv: 其他 AI 安全论文
- Who Bears the Excess Authority? Attributing AI Agent Behavior to the Delegating Party
PaperarXiv: 其他 AI 安全论文
- Safety-Preserving Fine-Tuning for Medical LLMs via Bi-Level Data Selection
PaperarXiv: 其他 AI 安全论文
- Counterfactual feature steering for identity unlearning with adjacency preservation
PaperarXiv: 其他 AI 安全论文
- Example-based spatial guidance: training-free concept erasure in diffusion models
PaperarXiv: 其他 AI 安全论文
- GPromptShield: Enhancing the Resilience of Graph Prompt Tuning under Adversarial Attacks
PaperarXiv: 其他 AI 安全论文
- SmoothShift: Disrupting Visual Adversarial Attacks against LVLMs at Test Time
PaperarXiv: 其他 AI 安全论文
- Model Immunity Is More Than Condition Number: The Importance of Plateau Regions
PaperarXiv: 其他 AI 安全论文
- Designing Deceptive Model Organisms That Successfully Evade Auditing
PaperarXiv: 其他 AI 安全论文
- Interpretable Federated Learning via Global-Local Attribution Alignment
PaperarXiv: 其他 AI 安全论文
- Abstention directions in LLMs: mechanistic interpretability of abstention in query ambiguity, knowledge boundaries, and safety alignment
PaperarXiv: 其他 AI 安全论文
- 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
- Where the Evidence Hides: Measuring Privacy and Diagnosability of Exported Agent Trajectories
PaperarXiv: 其他 AI 安全论文
- Where the saved time goes: limitations of hiding cross-tenant cache hits in LLM serving
PaperarXiv: 其他 AI 安全论文