MoRE: Scaling mixture of experts with hardware-aware low-rank routing
Research
Research News
New papers, breakthroughs and academic AI research
Universality and Generalization of Causal Transformers Across Context Lengths
Research
Training LLMs to Verbalize Evaluation Awareness
Research
ATLAS: Aligned Transport of Latent Structure for Reliable World Model Planning
Research
AdaKerNet: Neural Kernel Decoding for Task-Adaptive Prediction with Multimodal Large Models
Research
Quantization Enables Private Dense Retrieval against Malicious Service Providers
Research
SpecDrop: Parameter-Free Category-Conditioned Routing for Modular Specialization
Research
ReCAP: Retrieval-Guided Capability Reuse for Multimodal Continual Instruction Tuning
Research
Predicting Task Difficulty Without Rollouts
Research
SYNCR: Diagnosing and Learning Cross-Video Reasoning from Simulation
Research
Which Decisions Low-Bit Quantization Breaks, and How to Predict Them
Research
From Behavior to Mechanism: Tracing Divergent Response Modes in Frontier Language Models
Research