Refusal Lives Downstream of Persona in Chat Models
arXiv:2606.26161v1 Announce Type: new
Abstract: Linear directions in activation space have been identified for both refusal and persona traits in instruction-tuned chat models, but the two have been studied as separate mechanisms. We show they interact: a compliant persona gates refusal. In Qwen2.5...
AlgoEvolve: LLM-driven Meta-evolution of Algorithmic Trading Programs
arXiv:2606.26173v1 Announce Type: new
Abstract: Recent work shows that Large Language Models (LLMs) can act as semantic mutation operators for the evolutionary discovery of programs and proofs. Most current applications focus on static coding benchmarks. We extend this paradigm to algorithmic tradi...
Agentic Analysis for Agentic Infrastructure: An LLM-Powered Pipeline for Comparative Governance of DAO and Corporate AI Protocols
arXiv:2606.26203v1 Announce Type: new
Abstract: As AI agent protocols proliferate, the governance structures shaping their interoperability standards remain empirically underexamined. We introduce an LLM-powered comparative pipeline for large-scale governance discourse analysis, integrating automat...
The White House is asking OpenAI to slow roll the release of its new model over safety concerns
penAI reportedly plans to share its newest model, GPT 5.6, with a select group of partners instead of to the broader public. The reason: the Trump administration told it to.
Vector RAG Isn’t Enough — I Built a Context Graph Layer for Multi-Agent Memory
I benchmarked raw chat history, vector-only RAG, and a context graph on the same multi-agent conversations. The results exposed a surprising weakness in relational retrieval.
The post Vector RAG Isn’t Enough — I Built a Context Graph Layer for Multi-Agent Memory appeared first on Towards Data Scienc...
The Hot Path Belongs to GBDTs, Agents Own the Cold Path: A Payment-Fraud Benchmark
A reproducible benchmark on latency, cost, and reproducibility, and where agents actually earn their keep.
The post The Hot Path Belongs to GBDTs, Agents Own the Cold Path: A Payment-Fraud Benchmark appeared first on Towards Data Science.
DeepReinforce Releases Ornith-1.0: An Open-Source Coding Model Family That Learns Its Own RL Scaffolds
DeepReinforce released Ornith-1.0, an open-source coding model family built on Gemma 4 and Qwen 3.5. Instead of a fixed harness, the model learns its own scaffold during reinforcement learning. The 397B flagship reports 82.4 on SWE-Bench Verified, with all weights under the MIT license.
The post Dee...
General Intuition’s $2.3B bet that video games can train AI agents for the real world
General Intuition has raised $320 million to scale AI trained on millions of hours of gameplay, betting action data can help AI develop something closer to human intuition.
In this tutorial, we will show you how to use Gemini to create Google Sheets, build a useful table, generate formulas, analyze data, and improve the spreadsheet with follow-up prompts.
From Fortnite to robots: General Intuition raises $2.3B on bet that video games can train AI agents for the real world
General Intuition has raised $320 million to scale AI trained on millions of hours of gameplay, betting action data can help AI develop something closer to human intuition.
Understanding the brain with AI-driven explanations and experiments
Researchers introduce generative causal testing, which translates black box models into clear hypotheses and verifies them in the scanner, revealing what specific brain regions respond to in language.
The post Understanding the brain with AI-driven explanations and experiments appeared first on Micr...
3 Agents. 3 LLMs. 1 Aging GPU: Engineering Parallel Inference on Bare Metal
Beat the 8GB VRAM limit. Learn how to run three different LLMs on a single 8GB GPU using C++ layer multiplexing and admission control.
The post 3 Agents. 3 LLMs. 1 Aging GPU: Engineering Parallel Inference on Bare Metal appeared first on Towards Data Science.
Artificial intelligence is rapidly reshaping retail, but not in the ways consumers might immediately notice. The biggest transformation may not be flashy virtual try-ons or chatbot shopping assistants, but in how decisions are made behind the scenes: how products surface in search results, how inven...
5 Open Source Omni AI Models That Handle Text, Images, Audio, and Video
Take a practical look at multimodal, any-to-any systems for vision-language reasoning, speech interaction, document intelligence, real-time assistants, local deployment.
Letting an LLM Pick the Right RAG Page: The Arbiter Pattern at the End of Retrieval
Enterprise Document Intelligence [Vol.1 #7C] - One LLM call ranks the candidates with reasons. The output is one typed object your auditor can defend
The post Letting an LLM Pick the Right RAG Page: The Arbiter Pattern at the End of Retrieval appeared first on Towards Data Science.
I got a really interesting question last week from Mike Loukides, my editor at Radar, after he read the third part of this trilogy on context management. “Another issue I’ve read about,” Mike asked, “is the tendency for a model to ignore the middle of the context. I’ve seen that particularly for the...
The Self-Improving Loop in AI Agents: Architecture, Benefits, and How it Outperforms Traditional Agent Workflows
Most AI agents today follow fixed instructions and never get smarter on their own. They finish a task, forget what happened, and repeat the same mistakes tomorrow. A new design called the self-improving loop changes this. It lets agents learn from every result and improve over time. This guide expla...