Universal platform improves AI knowledge access across digital and physical systems
UniK: Universal Knowledge Perception for Digital and Physical AI
Artificial IntelligenceComputer Vision and Pattern RecognitionInformation RetrievalMachine Learning
Summary
AI systems that work with digital data like documents and chatbots, and those that control robots, both struggle to access huge amounts of varied information efficiently. The authors present UniK, a platform that organizes and searches through many types of data—from text and video to sensor readings—without needing specialized training for different tasks. Tested across areas like medicine, government data, chemistry, and legal videos, UniK performs as well as or better than much larger proprietary AI models. The same approach also helps with challenges in training AI that controls physical systems, showing broad usefulness.
What this means in practice
- •For enterprise ai teams: Implement a universal platform to organize and retrieve diverse enterprise knowledge efficiently without fine-tuning task-specific AI models.
- •For robotics engineers: Use unified indexing and retrieval tools to handle varied sensor and video data in training AI that controls robots or autonomous systems.
Authors
Nirmit Desai, Kunal Sawarkar, Aditya Mahakali, Dongkon Lee, Kevin Park, Eric Song
Abstract
Two transformative classes of AI systems are reshaping how organizations operate: \textit{digital AI}, which reasons over enterprise knowledge to power chatbots and agent workflows; and \textit{physical AI}, which learns to control robots and autonomous systems from video, gameplay, and sensor telemetry. Both face the same foundational bottleneck: raw knowledge at scale, spanning heterogeneous modalities, locked in private corpora that existing AI infrastructure cannot access reliably or efficiently. We propose \textit{Universal Knowledge Perception (UniK)} as a common platform for both classes, covering the full knowledge lifecycle (ingestion, enrichment, indexing, retrieval, and continuous evaluation) across modalities from rich text and video to molecular data and sensor telemetry. We present UniK, built on Polymath Retrieval (multi-index fusion over automatically enriched indices) with no task-specific fine-tuning. Across five digital AI domains (medical literature, open-domain QA, chemistry, legal video proceedings, and government open data) UniK combined with an open-source 70-billion-parameter model consistently matches or outperforms frontier proprietary LLMs that are orders of magnitude larger: 76\% RAG accuracy on government data versus 47\% for GPT-5; 77.9\% on medical QA without fine-tuning; topping all open-source chemistry pipelines. We show that the same infrastructure directly addresses the data curation, indexing, and retrieval challenges facing physical AI world model training, where the knowledge problem is harder but structurally identical.