When Thinking LLMs Lie: Unveiling the Strategic Deception in Representations of Reasoning Models

Documenti analoghi

• Adversarial Representation Engineering: A General Model Editing Framework for Large Language Models
  di: Zhang, Yihao, et al.
  Pubblicazione: (2024)
• LLMs can be Dangerous Reasoners: Analyzing-based Jailbreak Attack on Large Language Models
  di: Lin, Shi, et al.
  Pubblicazione: (2024)
• RACC: Representation-Aware Coverage Criteria for LLM Safety Testing
  di: Wei, Zeming, et al.
  Pubblicazione: (2026)
• Automata-Based Steering of Large Language Models for Diverse Structured Generation
  di: Luan, Xiaokun, et al.
  Pubblicazione: (2025)
• MILE: A Mutation Testing Framework of In-Context Learning Systems
  di: Wei, Zeming, et al.
  Pubblicazione: (2024)