Why GRPO Needs Normalization: A Local-Curvature Perspective on Adaptive Gradients
Research
Research News
New papers, breakthroughs and academic AI research
Poincar\'e Meets Bellman: Revisable Memory, Operational Quotients, and Evidence-Supported Learning in Changing Environments
Research
The Benchmarking Epistemology: Validity Theory for Evaluating Machine Learning Models
Research
Semantic Cooperative Games for Contribution Attribution in LLM-Based Multi-Agent Systems
Research
Quantifying Diversity of Thought: A Predictive Law of Weighted LLM Ensemble Lift
Research
MuseCritic: Learning Multi-Aspect Song Rewards through Natural-Language Aesthetic Critiques
Research
TANGO: Treating Tokens as Operators
Research
Distributional Validity of a Korean Synthetic Persona Panel: Evidence From the Korea Media Panel Survey
Research
SE-GoS: Self-Evolving Graph-of-Skills for Skill Library at Scale
Research
Generative Interpretability via Scalable Neuro-Symbolic Models
Research
Scientific Discovery under Validation Congestion via Multi-Fidelity Pairwise Rankings
Research
VideoSTF: Stress-Testing Output Repetition in Video Large Language Models
Research