LEAD uses online adaptive mechanisms including Potential-Scaled Instability and symmetric efficiency rewards based on correct rollouts to achieve higher accuracy-efficiency scores with substantially shorter reasoning outputs than base models on math benchmarks.
Multi-task learning using uncertainty to weigh losses for scene geometry and semantics
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
background 1
baseline 1
method 1
citation-polarity summary
years
2026 5verdicts
UNVERDICTED 5representative citing papers
Excess risk decomposes into independent alignment (trace of inverse average Hessian times gradient covariance) and curvature terms, so both flatness and gradient alignment are required; SAGE achieves this and sets new SOTA on DomainBed.
Proposes meta-learning attack with priority-aware gradient alignment for sample-wise targeted attacks on TTA that maintain label distribution consistency with no-attack baseline.
A CTM-GNN model with EnSRF assimilation and flow-weighted transition matrix fuses floating car data and camera observations to deliver physically consistent, network-wide traffic volume estimates and forecasts, demonstrated with improved accuracy in Manhattan.
A knowledge-guided loss enforces saliency map consistency between a primary and final model in multi-task learning to boost accuracy, AP, and interpretability in defect detection.
citing papers explorer
-
LEAD: Length-Efficient Adaptive and Dynamic Reasoning for Large Language Models
LEAD uses online adaptive mechanisms including Potential-Scaled Instability and symmetric efficiency rewards based on correct rollouts to achieve higher accuracy-efficiency scores with substantially shorter reasoning outputs than base models on math benchmarks.