foundation models
-
Liquid AI’s LFM2.5: Compact On-Device Models Released
Read Full Article: Liquid AI’s LFM2.5: Compact On-Device Models Released
Liquid Ai has introduced LFM2.5, a series of compact on-device foundation models designed to enhance the performance of agentic applications by offering higher quality, reduced latency, and broader modality support within the ~1 billion parameter range. Building on the LFM2 architecture, LFM2.5 scales pretraining from 10 trillion to 28 trillion tokens and incorporates expanded reinforcement learning post-training to improve instruction-following capabilities. This release includes five open-weight model instances derived from a single architecture, including a general-purpose instruct model, a Japanese-optimized chat model, a vision-language model, a native audio-language model for speech input and output, and base checkpoints for extensive customization. This matters as it enables more efficient and versatile on-device AI applications, broadening the scope and accessibility of AI technology.
-
Converging Representations in Scientific Models
Read Full Article: Converging Representations in Scientific Models
Machine learning models from diverse modalities and architectures are being trained to predict molecular, material, and protein behaviors, yet it's unclear if they develop similar internal representations of matter. Research shows that nearly sixty scientific models, including string-, graph-, 3D atomistic, and protein-based modalities, exhibit highly aligned representations across various chemical systems. Despite different training datasets, models converge in representation space as they improve, suggesting a common underlying representation of physical reality. However, when faced with unfamiliar inputs, models tend to collapse into low-information states, indicating current limitations in training data and inductive biases. This research highlights representational alignment as a benchmark for evaluating the generality of scientific models, with implications for tracking universal representations and improving model transferability across scientific tasks. Understanding the convergence of representations in scientific models is crucial for developing reliable foundation models that generalize beyond their training data.
-
Google Earth AI: Geospatial Insights with AI Models
Read Full Article: Google Earth AI: Geospatial Insights with AI Models
Google has advanced its AI capabilities with the introduction of Google Earth AI, which combines powerful foundation models with a geospatial reasoning agent to address complex, real-world questions at a planetary scale. This technology enhances the accuracy of Google Maps and provides timely alerts on weather and natural disasters by analyzing satellite imagery and other data sources. The geospatial reasoning agent breaks down complex queries into manageable steps, utilizing the latest Gemini models to integrate insights across different domains. New innovations, including imagery and population models, demonstrate state-of-the-art performance in solving intricate geospatial queries, offering potential applications for developers and enterprises. This matters because it enhances our ability to understand and respond to environmental challenges with precision and speed.