Infrastructure & deployment
TAP: Efficient Long-Horizon Agent Pruning via Trajectory-Anchored Recovery
TAP treats agent pruning as an adaptive recovery problem: it anchors training to teacher trajectories while letting the pruned student generate responses and update channel choices.
Why it matters: For long-horizon agents, pruning must preserve task success while reducing the cost of repeated inference. TAP’s reported GPU time per successful task makes that tradeoff explicit, and its adaptive scoring targets a failure that one-shot pruning misses.