Skip to content

Scan record

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

Nov 1639 loaded

  1. Adversarial attacks on symbolic vocabulary vulnerabilities in LLM planners

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

  2. Cracks in the Embedding: Robustness and Stability of JEPA-Based Chest X-ray Representations

    PaperarXiv: 其他 AI 安全论文

  3. Soteria: Formally Verified Planning and Runtime Enforcement for Safe LLM Agents

    PaperarXiv: 其他 AI 安全论文

  4. Anti-distillation is cheap: lightweight distillation resistance via style disruption

    PaperarXiv: 其他 AI 安全论文

  5. Auditing Chatbots' Encroachment into the Professional Legal Practice Domain

    PaperarXiv: 其他 AI 安全论文

  6. Safety side effects of preventive personality steering: an empirical audit

    PaperarXiv: 其他 AI 安全论文

  7. System security is only as strong as the weakest agent: an attack that occurs only in multi-agent systems

    PaperarXiv: 其他 AI 安全论文

  8. Pushing Down the Mountain: The Effect of Framing Effects and Internal State Interventions on Collusive Behavior of Long-Horizon LLM Agents

    PaperarXiv: 其他 AI 安全论文

  9. Adversarial tracking: attacks on image sensors for autonomous vehicle object tracking

    PaperarXiv: 其他 AI 安全论文

  10. Unfinished-Bench: Measuring Rule-Violating Behavior of Persistent Agents in Real Work Environments

    PaperarXiv: 其他 AI 安全论文

  11. Physical Attacks on End-to-End Autoencoder-Based Communications via Evolutionary Transfer Learning

    PaperarXiv: 其他 AI 安全论文

  12. Google Project Zero will shorten vulnerability disclosure deadlines in stages: 60 days starting in 2027, 30 days starting in 2028

    NewsiThome (Taiwan)Governance

  13. New York Times: Lutnick family business's ties to Malaysian data centers raise conflict-of-interest questions

    NewsSeoul Economic DailyGovernance

  14. Cheating in China's college entrance exam can carry a maximum sentence of 7 years in prison; AI question-solving features suspended during exams

    NewsSeoul Economic DailyEvent

  15. Top scorer in South Korea's tax accountant exam admits to cheating with a miniature camera and generative AI and turns himself in

    NewsFinancial News KoreaEvent

  16. AI cheating suspicions spread, calls emerge in multiple South Korean national exams to verify high scorers

    NewsChosunBizEvent

  17. South Korean prime minister asks for study on shutting down Wi-Fi at CSAT exam venues to prevent AI cheating

    NewsYonhapEvent

  18. Indian officials say India is less concerned about AI threats than the United States

    NewsSemaforAI updates

  19. Taiwan's deputy digital minister on the Jensen Huang scandal: "We feel sad"

    NewsSemaforGovernance

  20. A generalizable safety-aware robot behavior planning method based on an explicit constraint space

    PaperarXiv

  21. Adaptable Watermarking: Transforming Image Watermarking Models via Lightweight Adapters

    PaperarXiv

  22. VLAN: Vision-Language Accessible Navigation

    PaperarXiv

  23. Securing Multi-Agent Systems by Design

    PaperarXiv

  24. An operational definition of opaque serial depth

    PaperarXiv

  25. Beyond Attack-Based Accuracy: Rethinking Adversarial Robustness Evaluation via Structural Gradient Alignment

    PaperarXiv

  26. Decoupling Forgetting and Utility: Lightweight LLM Unlearning via Orthogonal Null-Space Projection

    PaperarXiv: Jailbreaks, Prompt Injection, Poisoning & Defenses

  27. Has Abliteration Become Harder? Not Consistently: Direction-Based Ablation and ARA for Refusal Removal in Qwen3.8, Gemma 4, Olmo 3.1 and Mistral Small 3.2 LLMs and Their Predecessors in Heretic

    PaperarXiv: Jailbreaks, Prompt Injection, Poisoning & Defenses

  28. Constructing a steering vector via input optimization

    PaperarXiv: Jailbreaks, Prompt Injection, Poisoning & Defenses

  29. Alignment via Reflect-Reasoning: Learning How to Follow Principles

    PaperarXiv: Jailbreaks, Prompt Injection, Poisoning & Defenses

  30. FT Lex commentary: publishers are capable of a tough contest with big tech companies

    NewsFinancial Times · AIAI updates

  31. Former Nvidia chief Jensen Huang steps down in disgrace, dealing a blow to "open" models, with Anthropic and OpenAI benefiting

    NewsFinancial Times · AIAI updates

  32. Japan plans to regulate unauthorized commercial use of AI to replicate celebrities' voices and likenesses

    Newskmkj.krGovernance

  33. OpenAI Foundation board member says AI incidents will become the most likely cause of human death, warns of AI losing control

    NewsSegye IlboOpinion

  34. US Commerce Secretary Lutnick reportedly clears AI chips and advocates lax regulation, possibly involving family business interests in Southeast Asia

    Newsglobal.hk01.comGovernance

  35. Korean Association of Tax Accountants says practical training for trainee tax accountants will introduce assessments without AI assistance

    NewsFinancial News KoreaEvent

  36. Online post claims to have used AI to cheat into medical school; anxiety rises ahead of Korea's Suneung

    NewsSBSEvent

  37. Korean police seize devices including a micro camera in the AI cheating case involving the top scorer in the tax accountant exam, and are expected to summon him soon

    NewsSBSEvent

  38. Korean government to form interagency emergency team to prevent AI cheating on the College Scholastic Ability Test

    NewsKorea JoongAng DailyEvent

  39. Anonymous internet user posts claiming to have cheated with AI on Korea's Suneung as early as 2024, prompting calls to postpone the exam

    NewsThe Korea HeraldEvent