Presents GradingAttack with token- and prompt-level adversarial attacks that compromise LLM educational grading agents on multiple datasets, showing prompt-level attacks succeed more while token-level are stealthier.
Cheating automatic short answer grading with the adversarial usage of adjectives and adverbs.International Journal of Artificial Intelligence in Education, 34:616–646, 2024
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
GradingAttack: Exposing Security Vulnerabilities in LLM Based Educational Grading Agents
Presents GradingAttack with token- and prompt-level adversarial attacks that compromise LLM educational grading agents on multiple datasets, showing prompt-level attacks succeed more while token-level are stealthier.