AI Briefing#foundation-model
The latest AI and developer news about #foundation-model, with the original source and a short summary.
The latest AI and developer news about #foundation-model, with the original source and a short summary.
IBM has unveiled Granite Timeseries Ensemble R1, which combines four time series models.
The pre-trained foundation model requires no feature engineering or table flattening, using up to 10,000 historical examples for single-pass inference on databases with over 500 billion rows.
Netflix experimented with embeddings, subgraph integration, and fine-tuning to balance latency, cost, and model power in production.
The new model improves conversion rates by 18.9% and click-through rates by 12% across Meta's advertising surfaces.
Kumo Tabular achieves state-of-the-art accuracy on TabArena with ELO 1950 while being 17x faster than LimiX-2, trained entirely on synthetic data.
Prior Labs has released TabPFN-3.5, a foundation model that significantly improves performance in tabular data processing.
Skild AI has released 'S1', a foundation model that leverages NVIDIA infrastructure to enable robots to perform new long-horizon tasks using only a single video.
IBM has released Granite PatchTST-FM-r2, a time series foundation model licensed under Apache 2.0.
LG AI Research has released EXAONE Tabular, a lightweight tabular foundation model pre-trained without real data.
LG AI Research has released 'EXAONE Tabular', a foundation model for tabular data pre-trained exclusively on synthetic data without using any real-world data.
Google Research has released TimesFM-3, a 330-million-parameter time-series foundation model that predicts multivariate data and covariates in a single forward pass.
The Ministry of Science and ICT (MSIT) has selected the Naver Cloud consortium as the final executing agency for the project to develop a cybersecurity-specialized AI foundation model.
The Qwen team released Qwen-Drive-1.0, a vision-language foundation model dedicated to autonomous driving that integrates 3D perception and motion planning.
IBM and Confluent have released the IBM Granite time-series foundation model in Early Access, operating within real-time streaming environments based on Apache Flink.
Google has released TimesFM-3, a foundation model for multivariate time series forecasting based on 330 billion parameters.
Ornith-1.5 achieves open-source SOTA through self-improvement via self-generated tasks and reinforcement learning
Proposes the Mobius-v0 architecture, which improves training efficiency and inference speed by separating knowledge storage and reasoning processes.
This is the August 2026 risk report released by Anthropic concerning AI model misalignment and the risks of automated R&D.
OlmoEarth Studio now supports on-demand generation and export of Earth observation data embeddings.
Experimental results suggest that LLMs can partially detect and regulate their internal states.
Dyna-2 demonstrated scaling laws for both humans and robots using 1 million hours of human video.
Xiaomi validated VLA scaling using 100,000 hours of manipulation data collected without robots.
DeepSeek V4 Flash 0731 achieved 89.0% on ARC-AGI-1.
Multimodal models developed more effectively when modalities were trained together from the beginning.