REALISTA generates semantically coherent adversarial prompts via latent-space optimization over input-dependent editing directions, achieving stronger hallucination elicitation than prior realistic attacks on open-source and reasoning LLMs.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
Develops a smoothing extension of ESQM for DC optimization over convex composite inequality constraints, proving O(ε^{-3}) iteration complexity to (ε,ε)-KKT points plus convergence in the convex case.
citing papers explorer
-
REALISTA: Realistic Latent Adversarial Attacks that Elicit LLM Hallucinations
REALISTA generates semantically coherent adversarial prompts via latent-space optimization over input-dependent editing directions, achieving stronger hallucination elicitation than prior realistic attacks on open-source and reasoning LLMs.
-
A smoothing extended sequential quadratic method for difference-of-convex optimization over a convex composite inequality constraint
Develops a smoothing extension of ESQM for DC optimization over convex composite inequality constraints, proving O(ε^{-3}) iteration complexity to (ε,ε)-KKT points plus convergence in the convex case.