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.
A survey on multi-task learning.IEEE Transactions on Knowledge and Data Engineering, 34(12):5586–5609
3 Pith papers cite this work. Polarity classification is still indexing.
citation-role summary
citation-polarity summary
verdicts
UNVERDICTED 3roles
background 3representative citing papers
Introduces TTP with time windows, creates new benchmarks from existing TTP instances, and shows a new heuristic outperforms adapted TSP and TTP methods on many instances.
This survey paper identifies opportunities for LLMs in low-resource language humanities research along with challenges in data accessibility, model adaptability, and cultural sensitivity.
citing papers explorer
-
Flatness and Gradient Alignment Are Both Necessary: Spectral-Aware Gradient-Aligned Exploration for Multi-Distribution Learning
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.
-
The Traveling Thief Problem with Time Windows: Benchmarks and Heuristics
Introduces TTP with time windows, creates new benchmarks from existing TTP instances, and shows a new heuristic outperforms adapted TSP and TTP methods on many instances.
-
Opportunities and Challenges of Large Language Models for Low-Resource Languages in Humanities Research
This survey paper identifies opportunities for LLMs in low-resource language humanities research along with challenges in data accessibility, model adaptability, and cultural sensitivity.