most citedfrom Benign import Toxic: Jailbreaking the Language Model via Adversarial Metaphors

3 citations · 3 across the 2 of their papers we have counts for

collaborators
Showing cs.LGShow all

1 paper · 1 filter

cs.LG2026

TACIT-Switch: Cost-Aware Model Escalation for LLM Agents from Censored Supervision

Ji'an Lei, Jian Huang

Agents with smaller language-model backbones are less expensive but can drift into persistent failure modes, whereas those with larger backbones are generally more reliable but mor…