» Tag
machine-learning
163 postsLLM Quantization Guide: Comparing GPTQ, AWQ and GGUF
A practical breakdown of LLM quantization techniques - GPTQ, AWQ, GGUF and bitsandbytes - explaining how 4-bit compression cuts VRAM needs while preserving model quality.
Scaling Laws Explained: From Kaplan to Chinchilla to Overtraining
A breakdown of LLM scaling laws from Kaplan to Chinchilla, and why modern models are deliberately overtrained to cut inference costs.
Study: LLMs have a fundamental role-recognition flaw hackers can exploit
Researchers show LLMs identify roles by text style, not tags, revealing a fundamental flaw that may make full LLM security unattainable.
CommitBrief — AI code reviews, right in your terminal
A provider-agnostic, local-first CLI that reviews your staged changes, a historic range, or a whole GitHub pull request. Zero telemetry, no server. Free and open source.
commitbrief.comStop Fine-Tuning Everything: A Framework for Model Adaptation
A decision framework for choosing between prompt engineering, RAG, fine-tuning, and domain pre-training when adapting foundation models.
AI Coding Agents Are Absorbing the Leaves, Not the Whole Tree
SWE-bench and METR data track fast AI coding gains, but verification cost — not raw difficulty — still defines where agents stop.
How a shared JSONL file inflated our eval score to 0.94
A retrieval-based few-shot selector shared its index with the eval set, inflating accuracy from 0.79 to a fake 0.94. Here's the contamination check that catches it.
An IRT-shaped practice score is not a real IQ test
IntelligenceMax's IRT-shaped practice score resembles item response theory but lacks empirical calibration — here's what it can and can't measure.
Intel's ACE brings outer-product matrix math to x86, rivaling Arm SME
Intel's ACE extends AMX with outer-product matrix acceleration, compared against Arm's SME2 in a detailed technical breakdown.
Self-Play LLM Judges Reward Persuasion, Not Correctness
Self-rewarding LLM judges score plausibility, not correctness; GSM8K experiments show judge approval climbing while true accuracy stays flat, revealing reward hacking.
Why price per 1M tokens is a misleading AI metric
Comparing AI models by price per 1M tokens can mislead teams. Tokenizer differences and chain-of-thought efficiency matter far more than the sticker price per token.