Causal Retention in Interactive Agents: Interface Factorization and Selective Adaptation
Quick summary
arXiv:2609.30650v1 Announce Type: cross Abstract: Task performance need not determine which intervention mechanism an agent retains. We study causal retention: whether a frozen learned state answers a mechanism-probe map fixed independently of training, including action, context, direct target, value, and delay. For finite structural causal model classes, the optimal probe error is a Bayes decision risk. It vanishes exactly when every learning-interface fiber lies within one probe-answer fiber; any state obtained by post-processing that interface inherits the same lower bound. A posterior-cove
Key takeaways
- arXiv:2609.30650v1 Announce Type: cross Abstract: Task performance need not determine which intervention mechanism an agent retains.
- We study causal retention: whether a frozen learned state answers a mechanism-probe map fixed independently of training, including action, context, direct target, value, and delay.
- For finite structural causal model classes, the optimal probe error is a Bayes decision risk.
Why it matters
“Causal Retention in Interactive Agents: Interface Factorization and Selective Adaptation” highlights the need for repeatable measurement rather than a single impressive demonstration. Independent validation across datasets and clearly stated limitations determine whether a result can guide product decisions.
Member comments