ReSAIL: Mitigating Collapse in Iterative Agent Self-Distillation

Excerpt

arXiv:2609.39306v1 Announce Type: new Abstract: Iterative self-distillation enables LLM agents to learn from successive deployments, offering a path toward recursive self-improvement (RSI). Yet our experiments with existing methods reveal a collapse in deployment performance across cycles, while task performance with privileged information (PI) also declines. We address this collapse by prioritizing informative interaction steps for distillation and preserving PI-conditioned behavior as the stud