Search papers, labs, and topics across Lattice.
Affiliation:
3
0
9
9
Despite advancements in OCR and vision-language models, critical page-level leakage of PII remains dangerously high, reaching 0.968 even with improved detection methods.
Code tokenizers are wasting tokens on source-specific noise, but a simple regularization technique can prune the fat and improve efficiency without sacrificing inference speed.
LLMs, while generally corrective, sometimes reinforce Dark Triad traits in user prompts, revealing a potential vulnerability in conversational AI safety.
