Perisic and Icardi winning league titles and still in the UCL while Conte & Inter........

How robust are the instructions in your instruction-tuned model? In our most recent work (w/ Chantal and byron wallace), we show that there is a considerable dip in performance on in-domain tasks when you slightly vary the instruction. arxiv.org/abs/2306.11270

What if we combine PPO with Monte-Carlo Tree Search – the secret sauce for AlphaGo to reach superhuman performance? Spoiler: MAGIC!! Our inference-time decoding method, PPO-MCTS, achieves impressive results across many text generation tasks. 📜 arxiv.org/abs/2309.15028 🧵(1/n)

This is a joint work co-led with my awesome mentees Bowen Zhao and Zander Brumbaugh. Please reach out to them if you have job or collaboration opportunities for them! Finally, always many thanks to our fantastic advisors Hanna Hajishirzi and Noah A. Smith ❤️❤️❤️

Super excited to share this paper! Temporal chaos in pretraining corpora greatly impacts language model’s capabilities in answering time-sensitive questions. We will publish our codes for dataset generation and temporal alignment later. Stay tuned!

Importance of Demand Generation ✍️ 🌟 The number of GPUs we own and the robustness of our L1 integration are crucial, but they're not the sole determinants of our long-term success. It's all about creating demand for our product. 🥅 Our ultimate goal is to stimulate demand for

Our Robotics Engineer, Tim Huff, gives a demo of Groundlight AI's Visual Inspections 🔍 and Anomaly Detection solution, compatible with Universal Robots 🤖 cobots:

🚀 Exciting News from Groundlight AI! 🚀 Introducing, the Groundlight Hub, your easy access to computer vision. youtu.be/u3xPnOnvTcE

#CVPR2024 down the street gets our science team out of the office! 10,000+ computer vision researchers in Seattle figuring out what's next in the field from multimodal foundation models to the most creative applications.

Just dropped ⚡ - our latest youtube tutorial, running Groundlight's computer vision on a Raspberry Pi! Watch it here: youtube.com/watch?v=YpNKHj… Tim Huff, our Robotics Engineer, walks you step-by-step how to create an application to get notified if someone steals your parking

It’s my pleasure making some minor contribution to this great project! This attention rollout video is my favorite “visualization of the year” so far. Try out our demo if you are interested and stay tuned for our future research!

We just released an open-source framework that makes it easy to build visual reasoning agents (with GRPO). github.com/groundlight/r1…

We've open-sourced a MCP that allows big models to use huggingface computer vision models as tools. This allows Claude to act as a "visual agent", using other task specific models to help it solve problems. Below, is an example of Claude using an open vocab object detector to

Excited to share some new work: we show how to efficiently train small vision-language models to use a zoom tool with GRPO. We identify key points for doing small-scale RL for tool use in VLMs generally, and share what works (and what doesn’t) under significant resource