Open-HumanScale: An Open Recipe for Scaling Human and Robot Experience for Embodied Foundation Models
Research
Research News
New papers, breakthroughs and academic AI research
Context-Aware Autoregressive Image Generation for Emerging Reasoning Properties
Research
Same Layout, Different Count: What Number Words Do Past Ten in Text-to-Image Models
Research
Understanding Reasoning from Pretraining to Post-Training
Research
Transforming Integer Grids to Spatial Intuition: Perception, Not Planning, Governs ARC-AGI-3
Research
KL for a KL: On-Policy Distillation with Control Variate Baseline
Research
Final Checkpoints Are Not Enough: Analyzing Latent Reasoning Faithfulness Along Training Trajectories
Research
FedFiOT: Fisher-Weighted Optimal Transport for Model Merging in One-Shot and Few-Round Federated Learning
Research
Context--Answer Metamodels: Forecasting LLM Activations Before the First Answer Token
Research
UpSafe$^\circ$C: Upcycling for Controllable Safety in Large Language Models
Research
AdaMamba: Adaptive Frequency-Gated Mamba for Long-Term Time Series Forecasting
Research
LITMUS: Benchmarking Behavioral Jailbreaks of LLM Agents in Real OS Environments
Research