Teaching LLMs to Ask: Self-Querying Category-Theoretic Planning for Under-Specified Reasoning
arXiv:2601.20014v1 Announce Type: new
Abstract: Inference-time planning with large language models frequently breaks under partial observability: when task-critical preconditions are not specified at query time, models tend to hallucinate missing facts or produce plans that violate hard constraints...
Insight Agents: An LLM-Based Multi-Agent System for Data Insights
arXiv:2601.20048v1 Announce Type: new
Abstract: Today, E-commerce sellers face several key challenges, including difficulties in discovering and effectively utilizing available programs and tools, and struggling to understand and utilize rich data from various tools. We therefore aim to develop Ins...
Should I Have Expressed a Different Intent? Counterfactual Generation for LLM-Based Autonomous Control
arXiv:2601.20090v1 Announce Type: new
Abstract: Large language model (LLM)-powered agents can translate high-level user intents into plans and actions in an environment. Yet after observing an outcome, users may wonder: What if I had phrased my intent differently? We introduce a framework that enab...
Alibaba Introduces Qwen3-Max-Thinking, a Test Time Scaled Reasoning Model with Native Tool Use Powering Agentic Workloads
Qwen3-Max-Thinking is Alibaba’s new flagship reasoning model. It does not only scale parameters, it also changes how inference is done, with explicit control over thinking depth and built in tools for search, memory, and code execution. Model scale, data, and deployment Qwen3-Max-Thinking is a trill...
Weaviate in 2025: Reliable Foundations for Agentic Systems
2025 was a defining year for us at Weaviate. Instead of chasing shiny features, we focused on an overarching goal - upgrading our infrastructure and technology in order to better support AI systems.
Retiring GPT-4o, GPT-4.1, GPT-4.1 mini, and OpenAI o4-mini in ChatGPT
On February 13, 2026, alongside the previously announced retirement of GPT‑5 (Instant, Thinking, and Pro), we will retire GPT‑4o, GPT‑4.1, GPT‑4.1 mini, and OpenAI o4-mini from ChatGPT. In the API, there are no changes at this time.
MBZUAI Releases K2 Think V2: A Fully Sovereign 70B Reasoning Model For Math, Code, And Science
Can a fully sovereign open reasoning model match state of the art systems when every part of its training pipeline is transparent. Researchers from Mohamed bin Zayed University of Artificial Intelligence (MBZUAI) release K2 Think V2, a fully sovereign reasoning model designed to test how far open an...
Accelerating Science: A Blueprint for a Renewed National Quantum Initiative
Quantum technologies are rapidly emerging as foundational capabilities for economic competitiveness, national security and scientific leadership in the 21st century. Sustained U.S. leadership in quantum information science is critical to ensuring that breakthroughs in computing, sensing, networking ...
40 companies shaping Silicon Valley’s AI landscape in 2026
Silicon Valley still sits at the center of the AI conversation, not because it has a monopoly on ideas, but because so many of the forces shaping AI’s future collide here.
With Apple’s new Creator Studio Pro, AI is a tool to aid creation, not replace it
Apple’s Creator Studio Pro leverages AI to help creators with tedious tasks—like finding clips or building slides—without trying to do the work for them
Federated Learning, Part 2: Implementation with the Flower Framework 🌼
Implementing cross-silo federated learning step by step
The post Federated Learning, Part 2: Implementation with the Flower Framework 🌼 appeared first on Towards Data Science.
By Chester Curme and Mason DaughertyAs the addressable task length of AI agents continues to grow, effective context management becomes critical to prevent context rot and to manage LLMs’ finite memory constraints.The Deep Agents SDK is LangChain’s open source, batteries-included agent harness. It p...
By Chester Curme and Mason DaughertyAs the addressable task length of AI agents continues to grow, effective context management becomes critical to prevent context rot and to manage LLMs’ finite memory constraints.The Deep Agents SDK is LangChain’s open source, batteries-included agent harness. It p...
What AI “remembers” about you is privacy’s next frontier
The ability to remember you and your preferences is rapidly becoming a big selling point for AI chatbots and agents. Earlier this month, Google announced Personal Intelligence, a new way for people to interact with the company’s Gemini chatbot that draws on their Gmail, photos, search, and YouTube ...
From the Gemini Calendar prompt-injection attack of 2026 to the September 2025 state-sponsored hack using Anthropic’s Claude code as an automated intrusion engine, the coercion of human-in-the-loop agentic actions and fully autonomous agentic workflows are the new attack vector for hackers. In the A...
This Is How Successful Data Teams Are Using AI (Sponsored)
Successful data teams aren’t using more AI; they’re using AI differently. They embed it into workflows and decisions, employing ownership models that many SMBs haven’t adopted.