Showing cs.CLShow all
2 papers · 1 filter
cs.CL2026
Hard Negatives Reveal What Easy Negatives Hide: Cross-Lingual Harmfulness Representations Degrade with Resource Tier Under Hard Negatives
Paras Balani, Subhrakanta Panda
Safety alignment in large language models is trained primarily in English, and recent work reports that the underlying harmfulness representation survives translation: English-trai…
cs.CL2026
Self-Referential Induction Increases Response Instability Relative to Unresolvable and Verifiable Questions in Large Language Models
Paras Balani, Subhrakanta Panda
Self-referential prompting has been shown to reliably induce large language models to produce first-person reports resembling subjective experience, but no prior work measures how…