Tag
TerminalTraj is a scalable pipeline that generates high-quality terminal agent trajectories using Dockerized environments, achieving up to 20% improvement on terminal task benchmarks with models like TerminalTraj-32B.
This paper investigates whether agentic mechanisms such as reflection and memory lead to controllable improvements over fixed LLM workflows for information extraction from scholarly PDFs, using conference-paper dataset extraction as a testbed.
TACO introduces a novel credit optimization method for code-tool agents that uses a differential reward probe and outcome-gated advantage routing to distinguish useful from redundant or misleading tool calls, improving multimodal agent performance.
This paper introduces OpenThoughts-Agent, an open-source data curation pipeline for training agentic language models, achieving a 44.8% average accuracy across seven benchmarks and outperforming prior open datasets through systematic experiments.
OpenBioRQ is a new benchmark of 12,553 unsolved biomedical research questions that tests agentic models' ability to verify sources and avoid false citations. It reveals that current models often link to wrong papers and suffer from agentic collapse on hard questions.
Nex-N2-Pro, a new open-source LLM from NexEcosystem, reportedly outperforms leading models like Kimi-K2.6 and DeepSeek-V4-Pro on benchmarks, featuring adaptive and coherent thinking.
Magma is an open-source repository from Microsoft Research for building multimodal AI agents that integrate vision, language, and action, providing model links, inference examples, training instructions, and demos.
Microsoft introduces the Fara1.5 family of small browser agents (4B, 9B, 27B) that achieve state-of-the-art performance on computer use benchmarks, scoring 63% on Online-Mind2Web and beating larger models like Operator and Gemini.
Andrej Karpathy discusses how the OpenClaw moment resonated with non-technical audiences because it was their first exposure to advanced agentic AI models beyond ChatGPT as a consumer product.