Preprint, not yet peer-reviewed. Researchers at Meta and Carnegie Mellon University study reinforcement learning for tool selection by a small language model and transfer to different retrievers and tool collections. The authors report that their post-trained Qwen3-8B outperforms Llama 4 Maverick on ToolBench; this is the paper’s reported result, not an independent benchmark verification.
a16z post from October 7, 2026 announcing an investment, amount not disclosed, in Preference Model, a company that builds reinforcement-learning environments for AI labs. The post says the company is open-sourcing a framework called Karotte.
A nonprofit research institute focused on basic AI research into how minds arise from computation. Its site frames minds as computational entities capable of goal-directed adaptive behavior in open, uncertain worlds, and lists open questions such as how minds learn in real time from sensorimotor experience and how synthetic and human minds can cooperate. The institute builds on The Alberta Plan for AI Research, says all results will be shared publicly with no intellectual property restrictions, and is recruiting research fellows.
#AI research#reinforcement learning#open science#The Alberta Plan#research institute#basic research
OpenAI's free educational resource for learning deep reinforcement learning: an introduction to the field, key algorithms with implementations, exercises, and guidance for becoming an RL practitioner or researcher.
The site of Ineffable Intelligence, a London AI lab founded in late 2025 by former DeepMind reinforcement-learning lead David Silver. It aims to build a "superlearner" that learns from experience rather than human data; it raised a $1.1 billion seed round in April 2026, per CNBC. The site presents the company's mission in its own words.
A plain-language explainer on RLCD, a way of aligning language models by learning from contrasting outputs rather than human ratings alone.
From the site: RLCD is a method developed to adjust language models to human preferences without using human feedback data. This approach aims to address…
Why I recommend it: Good background reading if you want to understand how the models you use are actually steered.
A free ebook walking through reinforcement learning from the basics to RLHF, written for practitioners rather than researchers.
From the site: Reinforcement learning (RL) is transforming how reliable AI agents are trained and deployed. Discover real-world use cases, efficiency techniques like LoRA, and practical patterns you can apply today.
Why I recommend it: Free download in exchange for an email address. Solid grounding if you keep seeing "RLHF" and nodding along.
