A twelve-week production-scale run of Andrej Karpathy's AutoResearch paradigm on two book-recommendation embedding systems surfaced five recurring failure modes absent from the original setting, accor…
Agora is a shared memory system designed to let multiple autonomous research agents, such as those in AutoResearch loops, build on each other's work instead of each session starting from scratch, acco…
RL environments are the essential data for training AI agents, according to a recent perspective shared by an industry commentator. The author argues that just as curated datasets enabled deep learnin…
Researchers propose World Model RL (WMRL), a method that replaces environment execution with a world model to overcome the bottleneck of scaling reinforcement learning for automatic research agents. W…
Imbue announced Catalyst, an open-source research tool that uses evolution-inspired methods to automate AI model research, achieving a val_bpb score of 0.9361 on nanochat optimization after 340 experi…
Wired reporter Steven Levy built a self-improving AI using tools like AutoResearch and Prime Intellect, training small language models to automate newsletter tasks. The experiment shows that recursive…
An AI agent using the AutoResearch framework autonomously improved a small GPT's training recipe over 123 experiments on a single H100 GPU, achieving a best mean bits-per-byte (BPB) of 0.9774 ± 0.0019…
A new survey from arXiv defines "AutoResearch" as the spectrum of AI-powered scientific workflow automation, moving from human-steered "Vibe Research" to emerging AI-led systems that coordinate larger…