Latent Information Sharing for Accelerating Federated Learning
Research
Research News
New papers, breakthroughs and academic AI research
CARM: Cancellation-Aware Response Masking for LLM Reinforcement Learning
Research
Learning Rate Transfer for Hybrid Transformer-SSM Architectures
Research
CAGE-NAS: Certified Functional Descent for Efficient Model Growth
Research
Finetuning with Sampling: SFT Learns Better Than You Think
Research
Fully Online Decentralized Learning in Stochastic Games with Unknown Independent Chains
Research
When Guessing is Rewarded: Rethinking Language Model Evaluation with Distributional Uncertainty Scoring
Research
Mixture-Trained Merging for Unified Multi-Objective Models
Research
Token-Level Video Reinforcement Learning
Research
Every Ablation Is a Dose: Counterweights and the Semblance of Self-Repair
Research
Autoregressive Drillhole Modelling Under Distribution Shift
Research
MERGE: Minimal Expression-Replacement GEneralization Test for Natural Language Inference
Research