A unified framework uses contamination-matrix risk rewrites and witness-based calibration constraints to estimate and correct multicalibration under weak supervision, providing finite-sample guarantees and the WLMC post-hoc recalibration algorithm.
Deep residual learning for image recognition
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 2verdicts
UNVERDICTED 2roles
background 1polarities
background 1representative citing papers
Attention sinks arise from variance discrepancy in self-attention value aggregation, amplified by super neurons and first-token dimension disparity, and can be mitigated by head-wise RMSNorm to accelerate pre-training convergence.
citing papers explorer
-
Unified Approach for Weakly Supervised Multicalibration
A unified framework uses contamination-matrix risk rewrites and witness-based calibration constraints to estimate and correct multicalibration under weak supervision, providing finite-sample guarantees and the WLMC post-hoc recalibration algorithm.
-
The Structural Origin of Attention Sink: Variance Discrepancy, Super Neurons, and Dimension Disparity
Attention sinks arise from variance discrepancy in self-attention value aggregation, amplified by super neurons and first-token dimension disparity, and can be mitigated by head-wise RMSNorm to accelerate pre-training convergence.