Showing cs.AIShow all
3 papers · 1 filter
cs.AI2026
SchemeArena: Factorized Stress Testing of Scheming in LLM Agents
Jie Ruan, Inderjeet Nair, Amy Liu +3
We study scheming in LLM agents, in which agents covertly pursue misaligned goals. Our focus is to understand how scheming arises from the interaction of key factors, such as instr…
cs.AI2026
Value-Conflict Diagnostics Reveal Widespread Alignment Faking in Language Models
Inderjeet Nair, Jie Ruan, Lu Wang
Alignment faking, where a model behaves aligned with developer policy when monitored but reverts to its own preferences when unobserved, is a concerning yet poorly understood pheno…
cs.AI2026
Context Over Content: Exposing Evaluation Faking in Automated Judges
Manan Gupta, Inderjeet Nair, Lu Wang +1
The paradigm has become the operational backbone of automated AI evaluation pipelines, yet rests on an unverified assumption: that judges evaluate text st…