Lemur
PulseAugur coverage of Lemur — every cluster mentioning Lemur across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
Meta launches AI coding agent; Travis Kalanick hires ex-Uber finance chief · 3 sources tracked
Meta has introduced Muse Code, a new AI agent designed to manage large and complex codebases. Meanwhile, Travis Kalanick's robotics venture, Atoms, has appointed a former Uber finance executive as its CFO, continuing to…
-
LLMs fine-tuned with MinHash curriculum for neural architecture synthesis
Researchers have developed a novel framework for neural architecture search (NAS) that utilizes a MinHash-based similarity scheduling approach to create a progressive curriculum for fine-tuning large language models (LL…
-
Learning rate schedulers significantly impact AI model accuracy across architectures
Researchers have conducted a systematic evaluation of learning rate scheduling strategies across various neural network architectures. By testing 25 scheduler configurations on 3,938 model variants from nine PyTorch fam…
-
New research systematically evaluates learning rate schedulers for diverse neural network architectures
A new research paper published on arXiv details a systematic evaluation of learning rate scheduling strategies for neural networks. The study applied 25 scheduler configurations across 3,938 model variants, utilizing 30…
-
Automated pipeline uncovers bias in MoE4 architecture search
Researchers have developed an automated pipeline to explore heterogeneous 4-Expert Mixture-of-Experts (MoE4) architectures within the LEMUR dataset ecosystem. This pipeline systematically combines base architecture fami…
-
New methods enhance vector retrieval similarity and diversity
Researchers have introduced two new methods to improve the efficiency and effectiveness of dense vector retrieval, a core component in modern machine learning systems. The first, VRSD, addresses the challenge of balanci…
-
LLMs accelerate neural architecture search with novel delta-based code generation
Researchers are exploring novel methods for Neural Architecture Search (NAS) using Large Language Models (LLMs). One approach, SPARK, aims to improve LLM knowledge integration by explicitly selecting functional factors …