Machine Learning Research - The Batch | DeepLearning.AI (Page 18)

Machine Learning Research

Efficient Subject Consistency For Stable Diffusion

Published in mid-2022, DreamBooth enables Stable Diffusion to depict variations on a given subject; say, a particular dog and the same dog with angel wings or wearing a chef’s hat.

Machine Learning Research

Like LoRA, But for Pretraining: GaLore, a memory-saving method for pretraining and fine-tuning LLMs

Low-rank adaptation (LoRA) reduces memory requirements when fine-tuning large language models, but it isn’t as conducive to pretraining.

Machine Learning Research

Model Merging Evolves: Researchers developed automated system for efficient model merging

The technique of model merging combines separate models into a single, more capable model without further training, but it requires expertise and manual effort. Researchers automated the process.

Machine Learning Research

24 Hours on an Old Consumer GPU: Optimizing LLMs for low-resource hardware

BERT, a large language model released in 2018 and built upon the then-new transformer architecture, marked a paradigm shift in AI.

Machine Learning Research

Benchmarks for Agentic Behaviors: New LLM benchmarks for Tool Use and Planning in workplace tasks

Tool use and planning are key behaviors in agentic workflows that enable large language models (LLMs) to execute complex sequences of steps. New benchmarks measure these capabilities in common workplace tasks.

Machine Learning Research

For Faster Diffusion, Think a GAN: Adversarial Diffusion Distillation, a method to accelerate diffusion models

Generative adversarial networks (GANs) produce images quickly, but they’re of relatively low quality. Diffusion image generators typically take more time, but they produce higher-quality output. Researchers aimed to achieve the best of both worlds.

Different results from new text-to-image models from Nvidia, Alibaba, and Stability AI

Jun 19, 2024

More New Open Models: New models from Nvidia, Alibaba, and Stability AI expand open options

A trio of powerful open and semi-open models give developers new options for both text and image generation.

Machine Learning Research

The LLM Will See You Now: AMIE, a chatbot that outperforms doctors in diagnostic conversations

A critical step in diagnosing illnesses is a conversation between doctor and patient to assemble a medical history, discuss approaches to managing symptoms, and so on.

Tech & Society

Audio Generation Clear of Copyrights: Stability AI releases enhanced text-to-audio generator Stable Audio Open

Sonically minded developers gained a high-profile text-to-audio generator.

Machine Learning Research

Better Teachers Make Better Students: Microsoft‘s Orca 2 strengthens the native reasoning abilities of smaller models

A relatively small student LLM that learns to mimic a larger teacher model can perform nearly as well as the teacher while using much less computation. It can come even closer if the teacher also teaches reasoning techniques.

Machine Learning Research

Richer Context for RAG: RAPTOR, a recursive summarizer, captures more relevant context for LLM inputs

Text excerpts used in retrieval augmented generation (RAG) tend to be short. Researchers used summarization to pack more relevant context into the same amount of text.

Machine Learning Research

Interpreting Image Edit Instructions: Meta’s Emu Edit improves text-to-image generation with task classification.

The latest text-to-image generators can alter images in response to a text prompt, but their outputs often don’t accurately reflect the text. They do better if, in addition to a prompt, they’re told the general type of alteration they’re expected to make.