concept
active
concept:l-retain-loss-termL_retain Loss Term
Regularization component of the composite loss that penalizes deviation from baseline model behavior on Alpaca instructions
Neighborhood — ranked by edge-count
Methods (1)
method
- Optimization procedure that learns orthonormal basis vectors satisfying causal truth and retention constraints via composite loss
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Novel variant of CL loss introduced in this paper targeting only causal subspace dimensions to improve OOD performance
- In machine learning, a function measuring the distance between current and desired output; analogous to stress.
- Auxiliary training objective from Grant (2025) that constrains intervened representations to remain near natural distribution
- Addressing disparity in loss magnitudes across tasks at the loss level
- The deadening effect of modern processes that prevent people from acting according to their feeling for the whole, damaging the global whole.
- Auxiliary objective combining L2 and cosine losses against pre-recorded CL vectors to improve causal relevance when one model is causally inaccessible.
- Additional term in RL SOO loss preserving agent capabilities analogous to KL term in RLHF
- One of two contrastive objectives analyzed; shown to be minimized by PMI kernel representation up to scaling