Tag
45 articles
Researchers at IIT Bombay and Adobe Research have developed a method to reverse-engineer LLM prompts from outputs with near-perfect accuracy, raising serious security concerns.
AllenAI's Open Instruct framework offers a comprehensive post-training pipeline for LLMs using SFT, DPO, and GRPO, optimized for 16GB hardware.
This article explains the technical concept behind OpenClaw AI agents and their ability to autonomously interact with digital systems, using a gym hacking example to illustrate the implications for AI safety and system security.
This explainer article explains what AI agent frameworks are, how they work, and why different frameworks vary in cost and speed. It's designed for beginners with no technical background.
Learn how to set up and run inference with AMD's Instella-MoE-16B-A3B, a 16B parameter Mixture-of-Experts language model that activates only 2.8B parameters per token.
Tencent open-sources AngelSpec, a torch-native framework for speculative decoding that introduces DFly, a block-diffusion drafter, and achieves significant speedups in LLM inference.
A comparison of four leading open-source LLM fine-tuning frameworks—Unsloth, Axolotl, TRL, and LLaMA-Factory—reveals their distinct approaches to speed, VRAM usage, and multi-GPU support.
NVIDIA's srt-slurm framework simplifies distributed LLM serving benchmarking using SLURM, enabling reproducible workflows and advanced performance analysis.
A roundup of 10 open-source no-code AI platforms that simplify the creation of LLM apps, RAG systems, and AI agents using visual and plain-English tools.
Sakana AI has launched Sakana Fugu, an orchestration model that dynamically routes tasks across a swappable pool of frontier LLMs, achieving strong performance on coding, reasoning, and agentic benchmarks.
Learn to build a simple multi-LLM orchestration system that dynamically selects the best AI model for each question, similar to Sakana AI's Fugu system.
This explainer introduces the Open Knowledge Format (OKF), a vendor-neutral specification for structuring knowledge for AI agents. It explains how OKF uses Markdown and YAML to create curated knowledge bundles, distinguishing it from traditional RAG systems.
