method
active
method:lagrangian-ratio-constraint-for-retraction-correction-balanceLagrangian ratio constraint for retraction/correction balance
Learned multiplier maintaining 60:40 correction:retraction ratio.
Neighborhood — ranked by edge-count
Papers (1)
paper
Frameworks (1)
framework
- Core pipeline turning probe-read features into scalable reward functions for open-ended tasks.
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.
- Task-dependent optimal sparsity.
- We find that the logarithm transformation also benefits existing gradient balancing methods.quote0.700Key finding showing the broader utility of the log transformation.
- Evaluation metrics for exact and near-verbatim memorization.
- Ablation conclusion.
- Generalization of the loss transformation.
- Distinction between illocutionary and perlocutionary correctness.
- Experimental variable sweeping proportion of self-correction to normal response training examples from 10% to 90%
- Task balancing requires simultaneous consideration of both loss scales and gradient magnitudesclaim0.686Core interpretive position of DB-MTL: complementarity of loss and gradient perspectives