The inferred user preference descriptions are used to define prompts for generating responses in the future.

Chile's Torres Del Paine is famous for its stunning landscapes, but it's also home to a fierce predator: the puma. These majestic creatures feed primarily on guanacos, although the hunting success rate is not very high, especially for female pumas. The photographer followed this female and her two cubs for several days, before witnessing her hunting. Breakthroughs, discoveries, and DIY tips sent six days a week. In Chile's famous Torres Del Paine National Park, a mother puma with her two cubs in tow attacks a guanaco.

The current paradigm of evaluating Large Language Models (LLMs) through static benchmarks comes with significant limitations, such as vulnerability to data contamination and a lack of adaptability to the evolving capabilities of LLMs.

Unsupervised Environment Design (UED) formalizes the problem of autocur-ricula through interactive training between a teacher agent and a student agent. The teacher generates new training environments with high learning potential, curating an adaptive curriculum that strengthens the student's ability to handle unseen scenarios. Existing UED methods mainly rely on regret, a metric that measures the difference between the agent's optimal and actual performance, to

Vision Language Models (VLMs) have demonstrated remarkable performance in 2D vision and language tasks. However, their ability to reason about spatial arrangements remains limited. In this work, we introduce Spatial Region GPT (SpatialRGPT) to enhance VLMs' spatial perception and reasoning capabilities.