Masahiro Kaneko

Stopping Computation for Converged Tokens in Masked Diffusion-LM Decoding

JailNewsBench: Multi-Lingual and Regional Benchmark for Fake News Generation under Jailbreak Attacks

ExaGPT: Example-Based Machine-Generated Text Detection for Human Interpretability

Machine Text Detectors are Membership Inference Attacks

Autoregressive Direct Preference Optimization

Bits Leaked per Query: Information-Theoretic Bounds for Adversarial Attacks on LLMs