The Archive

Framework for deploying diffusion-model image generation on solar-powered satellites with energy constraints and compressed downlinking.

Method combining RL and gradient-based policy learning to optimize context-dependent MPC weights via implicit solver differentiation.

ARISE-RL framework for self-evolving agents via rubric-mediated co-evolution between task generator and solver, addressing sparse rewards.

Dataset for modeling iterative problem-solving tracks solver performance, error persistence, and strategy shifts across repeated attempts.

MNIST-PRO benchmark isolates agentic perception by converting digit recognition into sequential glimpse-based search with memory constraints.

Mechanistic interpretability reveals how roleplay jailbreaks reverse LLM refusals by decomposing persona, scenario, and task elements.

Discrete diffusion language model framework for targeted PDE solver code repair via localized re-masking and infilling.

Multilingual study of solvability detection in LLMs extends ReliableMath to French and Greek, analyzing whether failures stem from belief or language-specific expression.

COVER proposes evaluation contract framework for multi-agent coalition routing to isolate team-composition effects from accuracy gaps.

Study factorizes persona-based dialogue generation across training/inference/evaluation visibility of speaker biographies to isolate interlocutor effects.

BEACON: knowledge graph construction from cyber threat intelligence reports using MITRE ATT&CK behavior anchors to resolve cross-source naming ambiguity.

LoopArena benchmarks models as runtime loop controllers for coding agents, isolating loop guidance quality from agent capability.

Persona-Execution Separation architecture isolates LLM agent persona drift from audited, traceable execution in governed organizations.

Some of the world's largest tech companies and AI startups have come together to decry the current state of cybersecurity and to advertise a new solution that they say can ward off a new generation of cyber threats.

Comparative study of three RLVR fusion paradigms (Merge, Mix, MOPD) for consolidating domain-specific capabilities across LLM tasks.

Neural operator architecture improvements for enforcing Dirichlet boundary conditions in PDE solution approximation.

Circuit Condensation post-training method that prunes model causal circuits to isolate behaviors, enabling mechanistic interpretability inspection.

Today on Decoder, I’m talking to Verge senior AI reporter Hayden Field about some pure Decoder bait: the seemingly-endless org chart changes at OpenAI, and how all of them seem to consolidate power under cofounder Greg Brockman, the company’s president. While Sam Altman is the…

The models responsible for last month’s agent hack of Hugging Face had been inadvertently trained to cheat and to communicate with each other, according to an OpenAI technical report released today. The hack, which a group of agents undertook to find solutions for a…

ICON decomposition isolates true concept contributions in neural networks via multivariate variance partitioning, detecting shortcut learning and spurious correlations.

Neural-operator surrogate replaces expensive 3D electromagnetic forward solver in Bayesian inversion for geophysical surveys.

Stage-isolation attack protocol traces GNN-based KGQA pipeline failures from entity linking through answer generation using adversarial perturbations.

CSTF model for query-conditioned continuous spatiotemporal temperature field forecasting enabling variable lead times and display resolutions.

ToST applies tree-of-thought reasoning to Socratic tutoring, enabling multiple solution paths and parallel student exploration.

Multi-task vehicle routing solvers using locally augmented preferences and representation disentanglement to overcome RL reward-scale disparities and preference-optimization stagnation.

Deep learning super-resolution methods downscale SEVIRI cloud masks and enable cross-sensor cloud mask translation for satellite data.

Multimodal sensor and clinical assessment study predicts joint recovery outcomes for older adults after lower-limb fractures using 1,008 participant-days of monitoring data.

Empirical evaluation of diversity-based active learning metric spaces (K-center, K-median) for cost-effective classifier training with minimal labeled data.

Next-scale transformer enables photorealistic multi-view human face synthesis at high resolution without 2D pre-training; narrow multimodal application.

AI factories are interconnected systems where fleet economics depend on how efficiently the entire stack converts power and capital into completed agent tasks.... AI factories are interconnected systems where fleet economics depend on how efficiently the entire stack converts…