Forgetting in LLM continual post-training is a geometry conflict between task-induced covariance structures and the evolving model state, controlled by gating Wasserstein barycenter merging on measured conflict.
Null-space filtering for data-free continual model merging: Preserving transparency, promoting fidelity
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
baseline 1
citation-polarity summary
fields
cs.LG 2years
2026 2verdicts
UNVERDICTED 2roles
baseline 1polarities
baseline 1representative citing papers
MADE-IT outperforms baselines in continual model merging accuracy and robustness by using manifold-aware expert evolution with adaptive thresholds and implicit subspace-based routing while pruning redundant experts.
citing papers explorer
-
Geometry Conflict: Explaining and Controlling Forgetting in LLM Continual Post-Training
Forgetting in LLM continual post-training is a geometry conflict between task-induced covariance structures and the evolving model state, controlled by gating Wasserstein barycenter merging on measured conflict.