Gradient Guidance at Test Time of Flow Policies in Reinforcement Learning. This paper uses expressive flow policies for RL without making policy training fragile. Thus, they do not train the policy with updates.
AGENTS
-
Budget caps necessary in the face of coding agents
By
–
So, I find it hard to agree with opinions claiming that these budget cuts are bad news for AI labs. The reason why budget caps are necessary is that coding agents have become undeniably
-
First infrastructure benchmark for agentic AI
By
–
The first infrastructure benchmark for agentic AI has arrived. An AI agent chains tens to hundreds of AI model calls, using tools, gathering context, and iterating until the task is complete. Existing benchmarks
-
Dozens of sub-agents dramatically improving visual fidelity
By
–
It's still in progress too! Currently dozens of sub-agents are working on dramatically improving the visual fidelity.
-
Importance croissante des traces dans les agents IA
By
–
I think a lot of agents still add logs after an LLM-centric architecture but you’re right that the importance of traces/logs in agents is getting non controversial
-
Experimentation with Hyperframes and Gemini agent flow for annotated videos
By
–
I'm messing around with an agent flow for combining Hyperframes with Gemini video analysis to make interesting annotated videos. pic.twitter.com/hmoGqwopqI
— fofr (@fofrAI) 12 juin 2026I'm having fun experimenting with an agent flow to combine Hyperframes with Gemini's video analysis to create interesting annotated videos.
-
Managed Deep Agents integrates with LangSmith Sandboxes for secure code execution
By
–
ICYMI: Managed Deep Agents integrates with LangSmith Sandboxes.
— LangChain (@LangChain) 12 juin 2026
See how you can allow your agent to safely execute code: https://t.co/wiQVO5luru pic.twitter.com/HMGFxFXudgIn case you missed it: Managed Deep Agents integrates with LangSmith Sandboxes. Learn how you can enable your agent to execute code safely: https://langchain.com/blog/introducing-managed-deep-agents …
-
Databricks Genie expands to predictive analytics with TabPFN and Agent Bricks
By
–
Business users can already use Genie to ask descriptive questions in natural language. Now the same conversational workflow can support predictive analytics too.
— Databricks (@databricks) 12 juin 2026
By combining Databricks Genie, TabPFN, and Agent Bricks, teams can turn questions like “Which customers are likely to… pic.twitter.com/z71u2MoBw2Business users can already use Genie to ask descriptive questions in natural language. Now the same conversational workflow can support predictive analytics too. By combining Databricks Genie, TabPFN, and Agent Bricks, teams can turn questions like “Which customers are likely to
-
AI Agent Skill: Müller-Brockmann Grid vs Generic Drawings
By
–
LOS DISEÑOS DE IA SE VEN GENÉRICOS PORQUE NO SIGUEN NINGÚN SISTEMA DE COMPOSICIÓN REAL.
— Nico (@nicos_ai) 12 juin 2026
Esta Skill lo resuelve → le enseña a tus agentes el sistema de grid de Müller-Brockmann (la base del diseño editorial moderno, en 162 páginas) y ahora lo aplican directamente en código.… https://t.co/D6NNBVd8ze pic.twitter.com/2ddAWBoY5yAI-generated drawings look generic because they follow no real composition system. This Skill solves it → it teaches your agents Müller-Brockmann's grid system (the basis of modern editorial design, in 162 pages) and now they
-
Replit enables parallel agents to build multiple projects at once
By
–
New video is out! You no longer build one thing at a time on Replit.
— Replit ⠕ (@Replit) 12 juin 2026
Run parallel agents to ship a website, mobile app, video, and pitch deck from one project, all at once.
And you can now add multiple artifacts to projects you already have. pic.twitter.com/kgNPiFzzomNew video is out! You no longer build one thing at a time on Replit. Run parallel agents to ship a website, mobile app, video, and pitch deck from one project, all at once. And you can now add multiple artifacts to projects you already have.