類似資料

• Monitoring the Internal Monologue: Probe Trajectories Reveal Reasoning Dynamics
  著者:: Chrabąszcz, Maciej, 等
  出版事項: (2026)
• Backdoor Vectors: a Task Arithmetic View on Backdoor Attacks and Defenses
  著者:: Pawlak, Stanisław, 等
  出版事項: (2025)
• Conditional misalignment: common interventions can hide emergent misalignment behind contextual triggers
  著者:: Dubiński, Jan, 等
  出版事項: (2026)
• REALISTA: Realistic Latent Adversarial Attacks that Elicit LLM Hallucinations
  著者:: Liang, Buyun, 等
  出版事項: (2026)
• FLAME: Flexible LLM-Assisted Moderation Engine
  著者:: Bakulin, Ivan, 等
  出版事項: (2025)