Signal over volume.

New Framework for Proper Scoring Rules and Multivariate Modeling in Right-Cen...

New Framework for Proper Scoring Rules and Multivariate Modeling in Right-Censored Survival Data

Supervised Memory Training: Pretraining Recurrent Networks without Recurrence

SARDI: Leveraging Diffusion Trajectories for Faster and More Accurate Retriev...

OpAI-Bench: A New Benchmark for Tracking Progressive Human-to-AI Text Transfo...

Vortex: A Programmable and Efficient System for Sparse Attention Serving in A...

Agent Memory: First Systems Characterization of Stateful Long-Horizon LLM Wor...

MedReCo: An Entity-Aware Framework for Comparative Reasoning in Radiology

CURVBENCH: Challenging Flat Leaderboards with Curvature-Stratified Relational...

New Komi-Yazva–Russian Corpus and Protocol for Evaluating LLMs in Extreme Low...

OPRD: Lifting On-Policy Distillation from Output Space to Hidden-State Space

VIDEO2LORA: Parametric Video Internalization for Efficient Vision-Language Mo...

WLA: A Unified World-Language-Action Model for Embodied Foundation Intelligence

NF-CoT: Accelerating LLM Reasoning via Autoregressive Normalizing Flows

Flash-WAM: Accelerating World-Action Models for Real-Time Robotic Control

Rethinking Continual Experience Internalization for Self-Evolving LLM Agents

SePO: Closing the Loop with Self-Evolving Prompt Optimization

ADR: Scaling Code RLVR via Atomic Decomposition and Recombination

RE-Edit: A Multi-Dimensional Benchmark for Reasoning-Aware Image Editing

MMPO: Optimizing Long-Horizon LLM Agents via Metacognitive Belief Entropy

EvoDS: A Self-Evolving Autonomous Data Science Agent with Skill Learning

Reinforcement Learning Elicits Meta-Skills for Translating Unseen Low-Resourc...

Neural Networks Discover Group Irreducible Representations via Riemannian Gra...

SuperMemory-VQA: A New Benchmark for Long-Horizon Egocentric AI Memory Assist...

Token Budgets: Preventing LLM-Agent Financial Runaways via Rust Affine Types