InfoWorld

Researchers propose a self-distillation fix for ‘catastrophic forgetting’ in LLMs

LLMs tend to lose prior skills when fine-tuned for new tasks. A new self-distillation approach aims to reduce regression and ...
InfoQ

Building Embedding Models for Large-Scale Real-World Applications

Sahil Dua discusses the critical role of embedding models in powering search and RAG applications at scale. He explains the ...
NPR

Wonder Cabinet

Wonder Cabinet is an independent podcast from Anne Strainchamps and Steve Paulson, Peabody Award-winning creators of public radio's To The Best Of Our Knowledge. For 35 years, that show brought ...
IEEE

Self-Labeling and Self-Knowledge Distillation Unsupervised Feature Selection

Abstract: This paper proposes a deep pseudo-label method for unsupervised feature selection, which learns non-linear representations to generate pseudo-labels and trains a Neural Network (NN) to ...
IEEE

Relation-Aware Contrastive Learning for Knowledge-Based Recommendation

Abstract: Knowledge Graphs (KGs) have emerged as a critical technique to enhance recommendation performance by modeling complex relationships and semantics within heterogeneous networks. However, it ...