Teaching students to think and learn through questions
Celebrate Teachers’ Day with AI for Educators in Brazilian Portuguese and see how Study Mode can help teachers support active, step-by-step learning.
Author:
Plan or Retrieve First? A Scale Study of Tool Discovery Architectures
Tool-augmented LLM agents discover relevant APIs from large registries before planning execution, yet finite context windows prohibit injecting all tool definitions into a single prompt. The dominant approach, which we term Retrieve-then-Plan (RtP), embeds the user query and retrieves similar tools via approximate nearest-neighbor search. We study an alternative Plan-then-Retrieve (PtR) paradigm in which the…
Selection-Based Structured Reasoning: Toward Efficient Multimodal Search Agents
Multimodal agents commonly generate freeform reasoning before each action. For small models, limited model capacity can result in lengthy reasoning that provides little useful guidance for action generation while incurring substantial inference cost. To address this challenge, we introduce Selection-based Structured Reasoning (SSR), a framework that reformulates reasoning as selection instead of open-ended generation. SSR…
Local and Distributed Large LLM Inference with MLX and RDMA on Mac Studio
How Does Starshield Work and Who Controls It
SpaceX’s secure network for governments comes in many forms, including some services vital to the Ukrainian war effort.
Tesla’s Energy Boss Leaves the Company
Mike Snyder, the vice president of energy and charging, stepped down as Tesla pivots to AI.
Opinion | The Market’s Rude Reminder on AI
Sooner or later investors will demand returns, and there will be losers.
Opinion | History Raises Its Eyebrows at the Week in AI
A Google executive warned on Tuesday of serious incidents starting in 2027. That’s 11 weeks away.
NVIDIA Nemotron 3.5 Super VL Powers Specialized Agents with Multimodal Reasoning
Always-on agents are taking on longer, more complex enterprise workflows. A single request can involve reading a document, checking information across sources,…
Scale Agentic Coding Workflows with Meta’s Muse Spark Open on NVIDIA GB300 NVL72
Meta launches Muse Spark Open, a coding-optimized 816B mixture-of-experts (MoE) model with 42B active parameters and a 1M token context window. Trained on…