When Reject Turns into Accept: Quantifying the Vulnerability of LLM-Based Scientific Reviewers to Indirect Prompt Injection

Documenti analoghi

• The Compliance Paradox: Semantic-Instruction Decoupling in Automated Academic Code Evaluation
  di: Sahoo, Devanshu, et al.
  Pubblicazione: (2026)
• How to Trick Your AI TA: A Systematic Study of Academic Jailbreaking in LLM Code Evaluation
  di: Sahoo, Devanshu, et al.
  Pubblicazione: (2025)
• Beyond Accuracy: Diagnosing Algebraic Reasoning Failures in LLMs Across Nine Complexity Dimensions
  di: Patil, Parth, et al.
  Pubblicazione: (2026)
• Measuring Representation Robustness in Large Language Models for Geometry
  di: Jawandhia, Vedant, et al.
  Pubblicazione: (2026)
• Distill to Delete: Unlearning in Graph Networks with Knowledge Distillation
  di: Sinha, Yash, et al.
  Pubblicazione: (2023)