A technique for deterministic labeling from stochastic models, with benchmarked Golang implementation.

Wikipedia’s high-level guidance noting some ways to identify LLM writing

Claude Code is the most delightful AI agent/workflow I have used so far. Not only does it make targeted edits or vibe coding throwaway tools less annoying, …

Wikipedia’s high-level guidance noting some ways to identify LLM writing

Wikipedia’s high-level guidance noting some ways to identify LLM writing

via: https://news.ycombinator.com/item?id=45362813

In a landmark study, OpenAI researchers reveal that large language models will always produce plausible but false outputs, even with perfect data, due to fundamental statistical and computational limits.

TextFX

TextFX

This leaderboard ranks LLMs based on their performance in Valyrian Games competitions. Models are ranked using the TrueSkill rating system, which accounts for win/loss records and the relative skill of opponents.

Benchmark of smartphone interaction for LLMs and multi-agent systems

A map/reduce workflow for LLMs, with what looks like local caching. To me build systems and data processing pipelines like this one have a big intersection.