How controllers from industrial machinery can coordinate multitask machine learning
Instead of compromising among parameter updates dictated by different training objectives, ControlG allocates computational capacity to objectives sequentially and dynamically.
A new benchmark for evaluating patient-facing health AI agents
PatientAgentBench generates a synthetic patient health record, a realistic clinical vignette, and a patient agent that converses with the AI system under evaluation, to capture wh…
Amazon is investing in the Lean Focused Research Organization
As AI agents take on higher-stakes decisions, Lean programming language makes it possible to mathematically prove they will behave safely.
AI research papers and lab updates — page 3
Current feed archive. Entries come from the loaded publisher feeds and change as those feeds update. Search and saved-story filters apply to this page.
Browse the news
Headlines grouped by publisher. Search this page by source or headline; use the archive for older entries.
Spacing
Order
Echoverse: Deep, evolving environments for computer-use agents
Computer-use AI agents struggle with multi-step workflows like email and customer support. Echoverse trains agents in realistic environments rather than simply providing more trai…
EvoLib: Turning experience into evolving knowledge
LLMs do not get smarter just by remembering more. EvoLib turns experience into evolving knowledge, taking reusable skills and insights that help models learn and adapt across task…
Cloud-Native Evaluation-as-a-Service: A Microservices Architecture for Scalable AI Monitoring with Conformal Guarantees
Abstract page for arXiv paper 2607.21623: Cloud-Native Evaluation-as-a-Service: A Microservices Architecture for Scalable AI Monitoring with Conformal Guarantees
On the Depth Scalability of Logic Gate Networks
Abstract page for arXiv paper 2607.21633: On the Depth Scalability of Logic Gate Networks
MotifRole-Diff: Risk-Optimal Role-Aware Corruption for Masked Molecular Graph Diffusion
Abstract page for arXiv paper 2607.21634: MotifRole-Diff: Risk-Optimal Role-Aware Corruption for Masked Molecular Graph Diffusion
Toward User-Conditioned Evaluation of Personal LLM Agents under Temporal Interventions
Abstract page for arXiv paper 2607.21635: Toward User-Conditioned Evaluation of Personal LLM Agents under Temporal Interventions
Measuring the Dependency Gap: Diagnosing Inter-Column Fidelity in Tabular Generative Models
Abstract page for arXiv paper 2607.21636: Measuring the Dependency Gap: Diagnosing Inter-Column Fidelity in Tabular Generative Models
Quasi-Monte Carlo Initialization for Meta-Reinforcement Learning
Abstract page for arXiv paper 2607.21637: Quasi-Monte Carlo Initialization for Meta-Reinforcement Learning
Toward Goal-Agnostic Joint-Embedding Predictive Control of Partial Differential Equations
Abstract page for arXiv paper 2607.21644: Toward Goal-Agnostic Joint-Embedding Predictive Control of Partial Differential Equations
Multi-Horizon Consistency as Geometry: When Latent Dynamics Contract, and When They Do Not
Abstract page for arXiv paper 2607.21645: Multi-Horizon Consistency as Geometry: When Latent Dynamics Contract, and When They Do Not
Adjustment Speed as a Safety Constraint for Nonstationary Reinforcement Learning
Abstract page for arXiv paper 2607.21646: Adjustment Speed as a Safety Constraint for Nonstationary Reinforcement Learning
A Drift Stable Quantum Federated Learning for Intelligent Services
Abstract page for arXiv paper 2607.21647: A Drift Stable Quantum Federated Learning for Intelligent Services
Prior laundering: learned priors with inherited, undetectable overconfidence
Abstract page for arXiv paper 2607.21721: Prior laundering: learned priors with inherited, undetectable overconfidence
Simulation-Based Empirical Bayes
Abstract page for arXiv paper 2607.21843: Simulation-Based Empirical Bayes
Efficient Online LLM Watermark Detection via Rao-Blackwellized E-Processes
Abstract page for arXiv paper 2607.21958: Efficient Online LLM Watermark Detection via Rao-Blackwellized E-Processes
Convergence analysis of a family of Zermelo-type iterations for the Bradley--Terry model
Abstract page for arXiv paper 2607.22221: Convergence analysis of a family of Zermelo-type iterations for the Bradley--Terry model
Variational Low-rank Tensor Decomposition for Multisubject Spatiotemporal Data Analysis
Abstract page for arXiv paper 2607.22262: Variational Low-rank Tensor Decomposition for Multisubject Spatiotemporal Data Analysis
General Value Functions for Remaining Useful Life and Failure-Mode Prediction
Abstract page for arXiv paper 2607.22268: General Value Functions for Remaining Useful Life and Failure-Mode Prediction
Hopformer: Homogeneity-Pursuit Transformer for Time Series Forecasting
Abstract page for arXiv paper 2607.22299: Hopformer: Homogeneity-Pursuit Transformer for Time Series Forecasting
Learning Bidirectional Causal Interactions with Heteroscedastic Neural Networks
Abstract page for arXiv paper 2607.22313: Learning Bidirectional Causal Interactions with Heteroscedastic Neural Networks
Learning Ergodic Dynamical Systems from a Finite Trajectory
Abstract page for arXiv paper 2607.22399: Learning Ergodic Dynamical Systems from a Finite Trajectory
Graph-Based Correlation Matrix Generation: A Convex Optimization Approach
Abstract page for arXiv paper 2607.22436: Graph-Based Correlation Matrix Generation: A Convex Optimization Approach
No matching sources found.