llm

Purified OPSD: Fixing Self-Distillation for Long Chain-of-Thought Models

The MLAS Matrix: Why Self-Evolving LLM Agents Are a Security Nightmare

Scientific Amnesia in Continual DPO: Why Repeated Post-Training Isn't Self-Improvement

Agents-K1: Towards Agent-Native Knowledge Orchestration

Context-CoT: Teaching LLMs to Actually Learn from Context

Qwen3.7-Max: The Agent Frontier — 35 Hours of Autonomous Coding, 1,158 Tool Calls, 10x Kernel Speedup

NanoResearch: Co-Evolving Skills, Memory, and Policy for Personalized Research Automation

A Comprehensive Survey on Agent Skills: Taxonomy, Techniques, and Applications

Agentic Memory Is a Memo, Not True Memory

The Future of Secure: AI, Quantum, and Zero Trust

Quantum Knowledge Graphs: Context-Dependent Triplet Validity

LARQL: The Model as a Queryable Graph Database

Self-Distillation: The Model as Its Own Teacher

AI Attacks: How Hackers Weaponize Artificial Intelligence

Microsoft BitNet 1.58: The Era of 1-Bit Large Language Models

DeepSeek V4: Compressed Sparse Attention and the Million-Token Context

DSPy Generative Feedback Loops: Compiling LM Programs That Improve Themselves

Learning to Reason with Insight for Informal Theorem Proving

LiteLLM: Unified API for 100+ LLM Providers

Trace2Skill: Distilling Trajectory-Local Lessons into Transferable Agent Skills

Learning to Self-Evolve: Training LLMs to Improve Their Own Contexts

Compiling Intelligence: How DSPy Optimizes Agent Pipelines

SGLang: Structured Generation Language for Efficient LLM Serving

vLLM: High-Throughput LLM Inference at Scale

Ollama: Run Local LLMs on Your Own Hardware

LoRA: The Efficiency Revolution in Language Model Fine-Tuning

OpenRouter: Unified API Gateway for 200+ LLM Models

Large Language Models: The Engine Behind Modern AI Agents

GraphRAG: Microsoft's Graph-Based Retrieval Augmented Generation