2 papers
cs.LG2026
Lying Is Just a Phase: The Hidden Alignment Transition in Language Model Scaling
Adil Amin
Scaling laws predict loss from compute but not how capabilities interact. We measure the coupling between reasoning and truthfulness across 63 base models from 16 families and find…
cs.LG2026
The Growing Pains of Frontier Models: When Leaderboards Stop Separating and What to Measure Next
Adil Amin
Leaderboards rank frontier models on independent axes but do not reveal whether capabilities reinforce or trade off across releases -- and at the frontier, this interaction is the…