Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
Which Rollout Taught It That? BehaviorTrace and the Limits of Training-Data Attribution in Online RL
Amit Nautiyal
When reinforcement learning teaches a language model a new behavior, can we find the training rollouts that taught it? And when an attribution method says it can, how do we know th…
cs.LG2026
Traceprop: Training Data Attribution in a Single Pass
Amit Nautiyal
Data attribution tools used in practice, TRAK, LoGRA/LogIX, and EK-FAC, run after training: they recompute per-sample gradients in a separate pass over the training set, and curvat…