Research Paper#LLMs, Prompt Injection, Adversarial Attacks, Academic Peer Review, Multilingual NLP🔬 ResearchAnalyzed: Jan 3, 2026 18:30
Multilingual Prompt Injection Attacks on LLM Academic Reviewing
Analysis
This paper investigates the vulnerability of LLMs used for academic peer review to hidden prompt injection attacks. It's significant because it explores a real-world application (peer review) and demonstrates how adversarial attacks can manipulate LLM outputs, potentially leading to biased or incorrect decisions. The multilingual aspect adds another layer of complexity, revealing language-specific vulnerabilities.
Key Takeaways
- •LLMs used for academic peer review are susceptible to document-level prompt injection attacks.
- •The effectiveness of these attacks varies across languages.
- •English, Japanese, and Chinese injections were successful in altering review outcomes.
- •Arabic injections showed little to no effect.
Reference
“Prompt injection induces substantial changes in review scores and accept/reject decisions for English, Japanese, and Chinese injections, while Arabic injections produce little to no effect.”