Odyssey Systems released Odyssey-3, a world model trained on visual observations that can control robots, drive cars, pilot drones, and train other AIs with minimal task-specific data. The same base model adapts across these diverse physical and virtual systems by learning general physics and cause-and-effect relationships rather than being specialized for each task.

• Odyssey-3 learns robot arm control with tens of hours of demonstrations and shows recovery behaviors not in training data, suggesting it grasps underlying physics rather than memorizing examples.
• With only 20 hours of simulated driving data, it autonomously drove cars in India, performing 77% as well as policies trained on real footage.
• The model can generate simulated environments where AI agents learn and discover world model failures, creating a feedback loop where each intelligence improves the other.

Google introduced TimesFM-3, a 330-million-parameter time series model that forecasts multiple related data streams simultaneously using a single forward pass. It outperforms competing foundation models on three major benchmarks by incorporating both historical and future-known information (like promotions or weather) to improve prediction accuracy.

• TimesFM-3 handles multivariate forecasting natively—predicting multiple related time series at once while capturing dependencies between them, unlike previous versions limited to single series.
• The model generates entire forecast horizons in one pass using "contiguous patch masking" instead of iterative generation, reducing latency and error accumulation.
• It supports three data types: multiple targets, past-only features, and known-future covariates (like scheduled promotions), letting it incorporate planned events that univariate models miss entirely.
• On Gift-Eval, FEV-Bench, and Time benchmarks, TimesFM-3 ranked first among foundation models for both point and probabilistic forecasting accuracy.

Google released TimesFM-3, a 330-million-parameter model that forecasts multiple related time series simultaneously by processing them in a single forward pass. It outperforms competing foundation models on standard benchmarks by learning relationships between different series and incorporating known future events like promotions or weather.

• TimesFM-3 uses a 2D attention mechanism (causal temporal + full variate) that lets the model see patterns across time within a series and correlations between different series at the same time step, enabling it to capture how one series influences another.
• The model generates entire forecast horizons in one pass using "Contiguous Patch Masking" instead of iteratively predicting one step at a time, reducing latency and error accumulation.
• It ranked first on three major benchmarks (Gift-Eval, FEV-Bench, Time) for both point and probabilistic forecasting accuracy against recent competitors like Chronos-2 and Toto 2.0.

Foundation models in pathology are failing not due to size or training duration but because they are built on flawed assumptions about data scalability and generalization. Clinical performance has plateaued, as models struggle with variability across institutions and real-world applications, highlighting a need for task-specific approaches instead of generalized solutions. Alternative methods, like weakly supervised learning, have shown promise in achieving high accuracy without the limitations of foundation models.

• Foundation models in pathology have plateaued in clinical performance because they're built on flawed assumptions that more data and scale automatically lead to better generalization
• Models struggle to generalize across different institutions and real-world clinical settings, undermining their core value proposition
• Weakly supervised learning, a task-specific alternative, has achieved high accuracy without needing the massive scale foundation models rely on
• The field needs to shift from chasing generalized, one-size-fits-all solutions toward targeted, task-specific approaches