Blog
No more endless PDFs. Discover the core value of the latest top-tier research in one article.
Retrieval-Augmented LLM Agents: Learning to Learn from Experience
Context Bootstrapped Reinforcement Learning
The Simplicity of the Hodge Bundle
EntropyCache: Decoded Token Entropy Guided KV Caching for Diffusion Language Models
Spectrally-Guided Diffusion Noise Schedules
VISER: Visually-Informed System for Enhanced Robustness in Open-Set Iris Presentation Attack Detection
Tinted Frames: Question Framing Blinds Vision-Language Models
A simple understanding of quantum electrodynamics using Bohmian trajectories: detecting non-ontic photons
D-Mem: A Dual-Process Memory System for LLM Agents
Holographic Quantum Foam: Theoretical Underpinnings and Observational Evidence
Learning to Reason with Curriculum I: Provable Benefits of Autocurriculum
TARo: Token-level Adaptive Routing for LLM Test-time Alignment
Multi-stage Flow Scheduling for LLM Serving
ResNets of All Shapes and Sizes: Convergence of Training Dynamics in the Large-scale Limit
Total Recall QA: A Verifiable Evaluation Suite for Deep Research Agents
AndroTMem: From Interaction Trajectories to Anchored Memory in Long-Horizon GUI Agents
Black Hole--Entropy Container or Creator
Action Draft and Verify: A Self-Verifying Framework for Vision-Language-Action Model
Improvement on the Erdős-Kleitman conjecture via the KKL theorem
FILT3R: Latent State Adaptive Kalman Filter for Streaming 3D Reconstruction
ZEBRAARENA: A Diagnostic Simulation Environment for Studying Reasoning-Action Coupling in Tool-Augmented LLMs
When Only the Final Text Survives: Implicit Execution Tracing for Multi-Agent Attribution
Large Reward Models: Generalizable Online Robot Reward Generation with Vision-Language Models
Formalization of QFT
Governed Memory: A Production Architecture for Multi-Agent Workflows
MoRI: Learning Motivation-Grounded Reasoning for Scientific Ideation in Large Language Models
SCALE:Scalable Conditional Atlas-Level Endpoint transport for virtual cell perturbation prediction
Box Maze: A Process-Control Architecture for Reliable LLM Reasoning
ManiTwin: Scaling Data-Generation-Ready Digital Object Dataset to 100K
GHOST: Fast Category-agnostic Hand-Object Interaction Reconstruction from RGB Videos using Gaussian Splatting