Skip to main content
Search

# Data Attribution

1 related articles

Subscribe to updates
  1. arXiv

    One token-level interaction decomposition links data selection, collision-versus-erosion forgetting, and plasticity loss into a single learning-dynamics account

    The work derives a token- and layer-wise decomposition of how learning from one token changes another prediction, splitting it into a direct force-alignment channel (CH1) and a diffuse channel mediated by shared readout geometry (CH2), with a forward-computable approximation using only logits, hidden states, and the readout; following this interaction over time, the authors report that positive interaction supports data selection, negative interaction splits into concentrated collision and accumulated erosion with distinct controls, and that declining task-conditioned readout transmission during long-horizon training tracks declining future learnability, with readout restoration partially recovering plasticity.

You’re all caught up