Today's trail

AI Benchmark for time-sensitive safety awareness in conversational assistants.

AI Context-aware autoregressive image generation for improved reasoning.

We introduce Micius-Flash, a 27-billion-parameter model for engineering with native multimodal input, deep reasoning and a 256K-token context window. We design the Micius data pipeline as a knowledge-tree-driven data flywheel comprising three stages: knowledge tree construction…

AI Forecasts LLM activations before first answer token using context activations.