Pre-training - Edgentiq

MISA: Optimizing Large Language Models with Module-wise Importance Sampling

Evaluating Agentic Potential in LLMs During Pre-Training with APTBench

Unlocking Language Model Success: The Coverage Principle Explains Pre-training’s Power

Empowerment-Based Pre-Training Boosts Reinforcement Learning Adaptability

Finding the Right Dose: A Scaling Law for Knowledge Infusion in LLMs

spot_img

Recently Added

Advancing Autonomous Driving Perception with Query-based Splatting

Read more

Enhancing AI Learning from Databases with Task-Aware Pre-training

Read more

Balancing Speed, Loss, and Performance in LLM Training: An Optimizer Comparison

Read more

Real-TabPFN: Boosting Tabular AI Models with Real-World Data

Read more

Unlocking Performance in Remote Sensing Change Detection: A Focus on Fundamental Design Choices

Read more

FastDINOv2: A New Approach for Efficient and Robust Vision Model Pre-training

Read more

Gen AI News and Updates

spot_img

- Advertisement -

MISA: Optimizing Large Language Models with Module-wise Importance Sampling

November 4, 2025

Evaluating Agentic Potential in LLMs During Pre-Training with APTBench

October 29, 2025

Unlocking Language Model Success: The Coverage Principle Explains Pre-training’s Power

October 20, 2025