arXiv:2609.02002v1 Announce Type: cross Abstract: Guideline-consistent semantic segmentation requires more than category recognition, as real-world…
AI News Brief Hourly Summary 2026-09-03 15h : 18 posts
18 posts published in the last hour 12:33On-Policy Distillation Meets Off-Policy GRPO: Training Compact Instruction-Following Rerankers 12:33OutageDiT: A Generative Foundation Model for Power Outage Forecasting and Scenario Simulation 12:33Bolster AI Reveals Fabricated Scale Behind Dark Web Counterfeit Market 12:32Convergence Theory…
On-Policy Distillation Meets Off-Policy GRPO: Training Compact Instruction-Following Rerankers
arXiv:2609.01947v1 Announce Type: cross Abstract: Compact instruction-following rerankers are attractive for deployment, but conventional distillation…
OutageDiT: A Generative Foundation Model for Power Outage Forecasting and Scenario Simulation
arXiv:2609.01896v1 Announce Type: cross Abstract: Power-outage planning requires scenarios before an event occurs. These scenarios must represent…
Bolster AI Reveals Fabricated Scale Behind Dark Web Counterfeit Market
Bolster AI on September 3, 2026 published research reporting that a Tor-hosted marketplace selling counterfeit currency manufactured the appearance of…
Convergence Theory of Knowledge Distillation in Asynchronous P2P Gossip Learning Network
arXiv:2609.01952v1 Announce Type: cross Abstract: Decentralized, serverless learning increasingly connects devices running different architectures, where…
5 Free Courses to Go From LLM Beginner to Practitioner
A curated, linear pipeline of high-signal free resources that takes you from backpropagation basics to deploying production-grade LLM applications.
Accurate in space, unreliable in time: how LLMs represent national cultural change
arXiv:2609.01902v1 Announce Type: cross Abstract: Assessments of cultural alignment have become an important part of the development and improvement of…
DJI Launches ROMO 2 Robot Vacuum Series Globally
DJI launched the ROMO 2 series globally on September 3, 2026, introducing two new robot vacuum and mop models, the ROMO P2 and ROMO A2. The Shenzhen-based…
Sparse Readout Prism: Explaining Logit-Lens Scores in Features Instead of Tokens
arXiv:2609.01936v1 Announce Type: cross Abstract: A language model’s prediction of its next token develops across layers, and lens methods track this…
Zeta-Lite: A Concurrent, Branchable In-Browser SQL Database for Agentic Memory
arXiv:2609.01818v1 Announce Type: cross Abstract: The browser has become a first-class database host: applications increasingly want to store, query, and…
Your Fraud Stack Was Built For Human Adversaries
For the last 30 years, CAPTCHA has done one job: make a machine fail a test that a person passes without thinking. That job is over. Researchers, testing…
Thinking effort aligns between humans and reasoning models in abductive reasoning
arXiv:2609.01867v1 Announce Type: cross Abstract: A major question in cognitive modeling concerns the behavioral alignment between large language models…
Fine-tuning a 350M Model for Better Structured Outputs in 100 GRPO Steps
This post has no text preview — click the link below to read the original article. This article has been indexed from Hugging Face – Blog Read the original article: Fine-tuning a 350M Model for Better Structured Outputs in 100…
Import What You Need: Learning When and How to Augment EHR Graphs with External Knowledge
arXiv:2609.01839v1 Announce Type: cross Abstract: Longitudinal prediction from electronic health records (EHRs) is limited by the sparsity and…
Meta closes in on the top with Muse Spark 1.3, and undercuts rivals on price
Meta has released Muse Spark 1.3, its fourth model in the series in five months. According to Artificial Analysis, the model gains the most on agentic…
Agent Memory Is a Surface for Endogenous Authorization Laundering
arXiv:2609.01836v1 Announce Type: cross Abstract: Long-running LLM agents rely on persistent memory to carry state across interactions, including…
OpenAI CEO Sam Altman warns of “unsustainable silliness” in compute buildout
Sam Altman warns of “unsustainable silliness” in the global AI data center buildout. Too many Neocloud providers are announcing massive capacity without…