Hugging Face's logo
Eugene Siow's picture
15 3 233

Eugene Siow

eugenesiow
MaggieAutumn's profile picture brxerq's profile picture simonholm's profile picture
·
https://eugenesiow.com

AI & ML interests

None yet

Recent Activity

liked a model 5 days ago
Datou1111/shou_xin
upvoted an article 12 days ago
🐺🐦‍⬛ LLM Comparison/Test: 25 SOTA LLMs (including QwQ) through 59 MMLU-Pro CS benchmark runs
Reacted to m-ric's post with 🔥 12 days ago
𝗦𝗵𝗼𝘄𝗨𝗜: 𝗮 𝘀𝗺𝗮𝗹𝗹 𝗲𝗻𝗱-𝘁𝗼-𝗲𝗻𝗱 𝗮𝗴𝗲𝗻𝘁 𝘁𝗵𝗮𝘁 𝗰𝗮𝗻 𝗻𝗮𝘃𝗶𝗴𝗮𝘁𝗲 𝗮𝗻𝘆 𝗨𝗜 𝗮𝗻𝗱 𝗼𝘂𝘁𝗽𝗲𝗿𝗳𝗼𝗿𝗺𝘀 𝗺𝘂𝗰𝗵 𝗯𝗶𝗴𝗴𝗲𝗿 𝘀𝘆𝘀𝘁𝗲𝗺𝘀! 📲 A team from NUS and Microsoft just released an agent that can act on any UI (Desktop, Android, Web) without needing additional text information. It works extremely well : they applied their method on a tiny Qwen2-VL-2B, and they managed to beat methods that use either much more powerful vision models (like GPT-4V) without using any additional info (e.g. leveraging the DOM of a webpage) like previous methods did ! 👏👏 They started from the idea that most existing methods rely heavily on text, which makes them less generalizable, while letting aside rich UI structure that user actually rely on when navigating this interfaces. ⚙️ They put several good ideas to work: 💡 Simplify screenshots to the max: They prune a lot the heavy visual content of UI screenshots, by removing cloned image patches (like any vast patch of the same color will be reduced to a small patch, while maintaining positional embeddings), then group patches from the same GUI elements together to simplify even further 💡 Build a truly generalist dataset: To train a general UI agent, you need trajectories from each possible UI, and express them in a common language. Authors merge datasets like OmniAct for Desktop, Mind2Web for websites, AMEX for Android trajectories to create a high-quality and diverse dataset. ➡️ Nice results ensued: They fine-tune a tiny Qwen-2-VL-2B on their method, and it reaches SOTA on several task (element identification, web navigation), even beating methods that either use additional info from the DOM or use much bigger VLMS like GPT-4v! 🏆 And performance could certainly jump with a slightly bigger vision model. Let's hope the community builds this soon! 🚀 Paper added to my "Agents" collection 👉 https://huggingface.co/collections/m-ric/agents-65ba776fbd9e29f771c07d4e
View all activity

Organizations

Spaces-explorers's profile picture DSO National Laboratories's profile picture AI Singapore's profile picture DSO Intern's profile picture Top Contributors: Model Downloads's profile picture open/ acc's profile picture