The branch point is doing more work than the routing.

MOPD's teachers all branch from the same SFT checkpoint, so their token-level signal stays near the student's distribution. That constraint is the load-bearing part, and this week's feed has two papers on what breaks when you drop it.

DAPD (2608.01735) names it the privilege illusion: a teacher holding privileged context teaches behavior the student cannot reproduce from its own inference-time context, and the student then acts as if the privilege were still there. W2S-OPD (2607.26246) comes at it from the other side and builds its proxy teacher in logit space specifically to stay distributionally adjacent to the student, out of a contrast pair where both models are smaller than the student.

Three designs, same rule: closeness to the student's distribution buys more than the teacher's absolute strength.

Which makes drift the number I want from LFM2.5. How far can a domain teacher move away from that shared SFT checkpoint, through its own SFT plus RLVR, before MOPD starts costing you? Is there a point where you re-branch instead of pushing the teacher further?

The multi-turn harness part is the most interesting signal here.

A small model that can hold its objective inside changing tool state, partial context, and long-running agent loops may end up being more useful than a larger model that only looks strong on static prompts.

I’d love to see more reporting on long-run drift: when the agent starts optimizing for the harness instead of the task, when it forgets the original objective, and when it can recover after ambiguity or conflicting context.

That feels like the next important benchmark layer for agentic models: not just “can it solve the task once?”, but “does it remain coherent and useful after the environment starts moving around it?”