Technology
SPARKLE: A Unified Single-Loop Primal-Dual Framework for Decentralized Bilevel Optimization
This paper studies decentralized bilevel optimization, in which multiple agents collaborate to solve problems involving nested optimization structures with neighborhood communications.
Multimodal Clinical Benchmark for Emergency Care (MC-BEC): A Comprehensive Benchmark for Evaluating Foundation Models in Emergency Medicine
However, developing effective foundation models for electronic health record (EHR) data in Emergency Medicine requires addressing several challenges.
Pre-RMSNorm and Pre-CRMSNorm Transformers: Equivalent and Efficient Pre-LN Transformers Zixuan Jiang, Jiaqi Gu, Hanqing Zhu, David Z. Pan Chandra Department of Electrical and Computer Engineering
Transformers have achieved great success in machine learning applications.
SelfCodeAlign: Self-Alignment for Code Generation Y uxiang Wei
SelfCodeAlign employs the same base model for inference throughout the data generation process. It first extracts diverse coding concepts from high-quality seed snippets to generate new tasks.
Robust Distributed Learning: Tight Error Bounds and Breakdown Point under Data Heterogeneity
A vast majority of prior work on robustness however assumes data homogeneity, i.e., local datasets are generated from the same distribution.
STREAMER: Streaming Representation Learning and Event Segmentation in a Hierarchical Manner
We present a novel self-supervised approach for hierarchical representation learning and segmentation of perceptual inputs in a streaming fashion.