Breaking the Solver Bottleneck: Training Task Generators at the Learnable Frontier
arXiv:2606.18284v1 Announce Type: new
Abstract: The limiting resource for training agents via reinforcement learning (RL) is increasingly frontier task supply: valid, solvable tasks just difficult enough to train the current model. As reasoning and agentic models improve, fixed task distributions s...
CODEBLOCK: Learning to Supervise Code at the Right Granularity
arXiv:2606.18286v1 Announce Type: new
Abstract: Supervised fine-tuning of code LLMs typically applies uniform cross-entropy loss to all response tokens, implicitly assuming that every token provides equally useful learning signal. Recent token-level selection methods challenge this assumption in na...
A Link between Shock-wave Theory and Symmetry-reduced Stochastic Gradient Descent for Artificial Neural Networks
arXiv:2606.18303v1 Announce Type: new
Abstract: We develop a mathematically explicit link between shock-wave theory and the symmetry-quotiented learning dynamics of stochastic gradient descent, drawing on differential geometry, Lie group theory, and fluid mechanics. Specifically, after quotienting ...
NAVI-Orbital: First In-Orbit Demonstration of a Zero-Shot Vision-Language Model for Autonomous Earth Observation
arXiv:2606.18271v1 Announce Type: new
Abstract: As Earth Observation data generation outpaces downlink bandwidth and human-in-the-loop processing, a widening gap has emerged between onboard collection and actionable ground intelligence. This paper presents NAVI-Orbital, a software system deployed o...
CaVe-VLM-CoT: An Interpretable Vision-Language Model Framework
arXiv:2606.18385v1 Announce Type: new
Abstract: Vision-Language Models (VLMs) remain prone to hallucinations, producing fluent but visually unfaithful outputs. Existing chain-of-thought and retrieval-augmented methods only partially address this, as they neither enforce step-level citation groundin...
Searching for Synergy in Shared Workspace Human-AI Collaboration
arXiv:2606.18413v2 Announce Type: new
Abstract: Automated AI agents are increasingly capable, yet many scientific and professional tasks require human judgment and contextual expertise. We study shared-workspace human-AI teams, where AI agents and human collaborators must coordinate responsibilitie...
arXiv:2606.18543v1 Announce Type: new
Abstract: Language model agents are becoming proficient executors at isolated, short-horizon tasks such as software engineering and customer service. Yet real-world challenges require a combination of sophisticated skills that remain largely untested in agents:...
DeFAb: A Verifiable Benchmark for Defeasible Abduction in Foundation Models
arXiv:2606.18557v1 Announce Type: new
Abstract: A rule-based logic solver resolves every instance in our benchmark in under 50 microseconds with 100% accuracy; the best frontier language model reaches 65% at best and drops to 23.5% under rendering-robust evaluation (worst case over four surface ren...
OpenAI Releases LifeSciBench, a 750-Task Benchmark Grading AI Models on Real Life-Science Research With Expert-Written Rubric
OpenAI's LifeSciBench evaluates whether frontier AI can handle real life-science research across 750 expert-authored tasks, seven workflows, and seven biological domains. Built by 173 PhD scientists with 19,020 rubric criteria, it grades reasoning and decisions, not just recall. The best model, GPT-...
Social media’s next evolution: User-controlled algorithms
Social media feeds are becoming more customizable as platforms like Threads, Instagram, and TikTok introduce tools that let users directly influence the algorithms powering their recommendations.
NEA’s Tiffany Luck on AI IPOs, personal agents, and the ROI reckoning
Tokenmaxxing was the hottest trend in Silicon Valley earlier this year, with CEOs encouraging employees to push AI usage as far as it would go. Then the bill came due. Uber reportedly blew through its annual AI budget in a few months, some companies cut Claude licenses for parts of their org, and Me...
Vercel Releases Eve: An Open-Source AI Agent Framework Where Each Agent is a Directory of Files Mapped to Capabilities
Vercel has open-sourced eve, an Apache-2.0 agent framework now in public preview. An agent is a directory of files, with durable execution, sandboxes, approvals, connections, channels, and evals built in. Scaffold with npx eve@latest init and deploy unchanged via vercel deploy.
The post Vercel Relea...
Google bets on Gemini to reinvent the smart home speaker
Google is betting generative AI can breathe new life into the smart speaker. The company's new $99.99 Google Home Speaker replaces the rigid commands of the Google Assistant era with more conversational Gemini interactions.
Is your most capable AI agent also your biggest data leak?
A Microsoft and Huazhong University benchmark tested GPT-4o, GPT-5, Grok-3, and others on realistic enterprise data scenarios. Privacy violation rates hit 50.9%. More capable models made it worse, and the fix has nothing to do with model selection...
The Secret to Reproducible and Portable Optimization: ORPilot’s Intermediate Representation (IR)
Why production-level AI optimization modeling agent needs reproducibility and portability, and how IR helps achieve them
The post The Secret to Reproducible and Portable Optimization: ORPilot’s Intermediate Representation (IR) appeared first on Towards Data Science.
You know the meeting. The board wants an AI agent strategy by end of quarter. Someone on the leadership team has read a McKinsey report. You’ve been voluntold to build the platform. The slide deck says “AI-native.” The acceptance criteria are vague. Somebody mentions LangGraph, and somebody else say...
Most LLM applications need a clear workflow, not an autonomous agent. Here's how to build one in plain Python.
The post You Probably Don’t Need an Agent Framework appeared first on Towards Data Science.