Tag

#LoRA

8 articles

Auditing Preference Biases and Fine-Tuning Language Models with Direct Preference Optimization on Anthropic HH-RLHF Using TRL and LoRA

This article explains Direct Preference Optimization (DPO), a method for fine-tuning language models using preference data, and how it can be implemented using TRL and LoRA tools. It also discusses the importance of auditing preference data for biases.

Aug 1919

Fine-Tuning Qwen3 with LoRA Using NVIDIA NeMo AutoModel: A Complete Single-GPU Google Colab Workflow Tutorial

This article explains how to fine-tune the Qwen3 language model using Low-Rank Adaptation (LoRA) and NVIDIA NeMo AutoModel in a single-GPU Google Colab environment, focusing on parameter-efficient training techniques and automated workflows.

Jul 1856

Mira Murati’s Thinking Machines Lab Makes The Technical Case For Human-Centered AI Built On Customizable Model Weights

This article explains how customizable AI model weights and techniques like LoRA fine-tuning can make AI systems more flexible, personalized, and human-centered.

Jul 1161

Training Gemma-3 for Structured Mathematical Reasoning with Tunix GRPO, LoRA Adapters, and GSM8K Rewards

Learn how researchers are training an AI model called Gemma-3 to solve math problems using advanced techniques like GRPO and LoRA adapters.

Jul 551

Trajectory Releases a Concurrent Multi-LoRA Training Stack for Continual Learning, Reporting a 2.81× Experiment-Throughput Gain

Learn how a new AI system enables faster and more efficient continual learning by running multiple experiments at once using LoRA adapters.

May 3095

The LoRA Assumption That Breaks in Production

LoRA, a widely used technique for fine-tuning large language models, assumes all updates are similar — a premise that fails in real-world production environments. This limitation is now prompting a reevaluation of its effectiveness in complex, diverse applications.

Apr 2692

A Coding Implementation on Microsoft’s Phi-4-Mini for Quantized Inference Reasoning Tool Use RAG and LoRA Fine-Tuning

This article explains how Microsoft's Phi-4-Mini AI model uses quantization, RAG, and LoRA techniques to create efficient, powerful language models that can answer questions and use tools.

Apr 2084

Sakana AI Introduces Doc-to-LoRA and Text-to-LoRA: Hypernetworks that Instantly Internalize Long Contexts and Adapt LLMs via Zero-Shot Natural Language

Sakana AI introduces Doc-to-LoRA and Text-to-LoRA, hypernetwork techniques that enable instant long-context internalization and zero-shot LLM adaptation via natural language instructions.

Feb 27137