Y Hacker News new | ask | show | jobs

user: codelion created: 2008-02-23 karma: 3058 submissions: A 1B humanizer that matches human writing on an AI detector 3 points | 0 comments 0 points | 0 comments 0 points | 0 comments Zero-Dependency Programming 2 points | 0 comments 0 points | 0 comments Scaling Pedagogical Pre-Training: From Optimal Mixing to 10B Tokens 2 points | 0 comments From HashHop to Memory-Augmented Language Models 2 points | 0 comments The Optimal Architecture for Small Language Models 1 points | 0 comments Enhancing LLMs with LoRA – Standardized Recipes for Capability Enhancement 3 points | 0 comments OpenEvolve: Teaching LLMs to Discover Algorithms Through Evolution 53 points | 9 comments Ellora: Enhancing LLMs with LoRA Standardized Recipes for Capability Enhancement 1 points | 0 comments The 1B Token Challenge: Finding the Perfect Pre-Training Mix 6 points | 0 comments The 1B Token Challenge: Finding the Perfect Pre-Training Mix 3 points | 0 comments 0 points | 0 comments 0 points | 0 comments Pivotal Token Search (PTS): Targeting Critical Decision Points in LLM Training 2 points | 1 comments 0 points | 0 comments Internal Coherence Maximization(ICM): Label-Free Unsupervised Training Framework 1 points | 0 comments Unsupervised Model Improvement via Internal Coherence Maximization 1 points | 0 comments Show HN: PTS Library – Analyze LLM reasoning through "thought anchors" 2 points | 0 comments Automated Discovery of High-Performance GPU Kernels with OpenEvolve 4 points | 0 comments Adaptive Classifier: Dynamic Text Classification with Continuous Learning 1 points | 0 comments Show HN: DeepThink Plugin – Bring Gemini 2.5's parallel reasoning to open models 4 points | 0 comments Eliciting Fine-Tuned Transformer Capabilities via Inference-Time Techniques 1 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments