TISD: On-Policy Self-Distillation with Trajectory Intervention
Quick summary
arXiv:2609.30878v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) provides dense teacher targets, but evaluates them only along student-sampled rollouts. When the privileged teacher favors an alternative action at a visited prefix, OPSD can provide a target for the branch decision but cannot supervise the successor contexts induced by that action unless the student samples it. This creates a training-time data-collection bottleneck and suggests a different role for teacher-student disagreement: proposing a trajectory branch rather than identifying a sufficient local repair. Ou
Key takeaways
- arXiv:2609.30878v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) provides dense teacher targets, but evaluates them only along student-sampled rollouts.
- When the privileged teacher favors an alternative action at a visited prefix, OPSD can provide a target for the branch decision but cannot supervise the successor contexts induced by that action unless the student samples it.
- This creates a training-time data-collection bottleneck and suggests a different role for teacher-student disagreement: proposing a trajectory branch rather than identifying a sufficient local repair.
Why it matters
The significance is not only the legal text but how it changes product design. Decisions around “TISD: On-Policy Self-Distillation with Trajectory Intervention” may reshape data collection, model training, output accountability and market access.
Member comments