MuonIO: Principled Norm-Aware Descent for Embedding Tables and Language Model Heads
Research
Research News
New papers, breakthroughs and academic AI research
Probe the Harness: Setup Checks for Stale-Data RL Comparisons in Language Models
Research
Conditional Capacity and Routing in Mixture-of-Experts Particle Transformers
Research
ChromaGS: Text-Driven Semantic Editing of 4D Gaussian Avatars
Research
Corrupted but Correct: Why Vision-Language Models Lie to Themselves Internally
Research
Benchmarking Candidate Coverage in Typed Decision Models
Research
A Near-Zero Monitor Readout Is Not Evidence of Behavioral Control
Research
From Benchmarks to Production: A Text-to-SQL System for Complex Financial Data
Research
Understanding Trajectory Heterogeneity in Federated World Model Learning
Research
Sentry: Learning to Recover from LLM Agent Failures at Test Time
Research
Tailoring the Quantization Space for 1-Bit KV Cache Compression
Research
Differential Privacy of Gradient Descent on Perturbed Objectives
Research