Developer intelligence
What changed, why it matters, and what you can do now.
Your stack
Add your stack to personalize.
OpenAI introduces Dots, proactive agents powered by GPT-6 Astra, that can orchestrate tasks, integrate with user tools, and operate independently of the user's desktop. Developers on Pro plans get one Dot at no extra token cost.
Researchers introduce a new method (UNM-DPO) for direct preference optimization in language models by incorporating prompt-dependent uncertainty and preference-strength normalization. ULNM-DPO-WR, a variant with length normalization, outperforms prior DPO methods on several evaluation benchmarks.
JARQ is a plug-in refinement method for group-wise post-training quantizers that improves accuracy and perplexity on quantized large language models, without additional inference cost.
MILO is a new framework that automates the discovery and optimization of agent harnesses by evolving both harnesses and their search strategies. It significantly outperforms existing state-of-the-art methods across multiple benchmarks, reducing human effort and resource usage.
A new paper presents evalstats, an open-source package addressing inflated false positives in AI evaluation based on LLM judge scores. Developers gain access to robust, calibrated statistical methods and actionable guidelines for small-sample studies.
A new framework called Less Uniform Diffusion (LUDI) addresses key limitations in uniform diffusion language models (UDLMs) by introducing a less uniform loss and per-token time embeddings, resulting in faster and more effective language generation.
