AI RESEARCH
Prediction Bottlenecks Don't Discover Causal Structure (But Here's What They Actually Do)
arXiv CS.AI
•
ArXi:2605.09169v1 Announce Type: cross A Mamba state-space model trained only for next-step prediction appears to recover Granger-causal structure through a simple readout $S = |W_{out} W_{in}|$, with early experiments suggesting the phenomenon generalized across architectures and benefited from interventional data at $p < 10^{-5