DoNews reports that Princeton University introduced SSoN, a benchmark for evaluating spatial intelligence in large models. The cited results show 83.3% accuracy for humans, while the highest accuracy reported among large models was 39.7%.
Princeton AI Lab introduced Same Space or Not (SSoN), a benchmark that asks models to identify which of four static indoor views contradicts the others and explain the evidence. In results reported by 36氪 and AIbase, humans scored 83.3% on 1,000 questions, while the best of six tested VLMs, GPT-6 Astra Pro, reached 39.7%. Targeted reasoning-grounding supervision improved Qwen2.5-VL-7B from 21.89% to 41.79% on 201 questions from unseen rooms.
What problem it solves
Creating and revising presentations often consumes substantial time on layout, styling, animation, and other presentation-specific details. The project addresses the difficulty of adapting AI-generated content in fixed-format files by providing editable HTML slides, visual editing, import support, and export options.
Key features
AI-assisted slide creation — Users can provid…
What problem it solves
Learners need a practical path from AI engineering fundamentals to tested systems rather than isolated API examples. The curriculum is structured to make model internals, retrieval pipelines, agent runtimes, evaluation, and failure inspection understandable through runnable code and retained evidence.
Key features
Structured AI engineering curriculum — The curriculum spans m…
What problem it solves
Coding agents commonly pause for human approval before risky steps, which makes unattended repository work difficult to run in CI, daemons, and scheduled jobs. Teams also need consistent policies, tools, and supervision across different execution environments without maintaining separate framework integrations.
Key features
Repository tooling through MCP — Give an LLM tools …
What problem it solves
Learners need tutoring that can adapt across long-term study, use their own learning materials, and connect answers to sources. They also need a flexible environment for reading, practice, research, visualization, and other agent-assisted workflows without separating each activity from their learning history.
Key features
Personalized agent collaboration — The tutoring envir…
What problem it solves
AI coding agents need structured, reusable workflows to produce polished design, media, visual documentation, and knowledge-based results. The collection packages these workflows as Skills that can be installed and used across several SKILL.md-compatible agent environments.
Key features
Polished front-end design workflow — The web-design-engineer Skill guides agents through …
What problem it solves
Creating engaging lessons from a topic or set of materials typically requires assembling slides, quizzes, simulations, and other learning activities. OpenMAIC addresses this by using AI teachers and classmates to generate and facilitate interactive classroom experiences.
Key features
One-click lesson generation — Describe a topic or attach materials to generate a complete in…
oh-my-ppt• Pinned
