EMODE: Dynamic Para-Semantic Experts for Emotion-Aware Speech Language Modeling
Research
Research News
New papers, breakthroughs and academic AI research
HARL-A: An Extensible Benchmark Framework for Heterogeneous Multi-Agent Adversarial Reinforcement Learning in IsaacLab
Research
Metonymic Circuits for Abstract Concept Grounding in Vision Transformers
Research
Component and Dimension Sparsity in Transformer Refusal Mechanisms
Research
Verdicts Without Annotated Evidence: Rejection Sampling or Label-Only Post-Training for Evidence Recovery?
Research
Stabilizing language models under continual learning via condition-anchored distillation
Research
Investigating Model Compression for Neural Machine Translation in the Biomedical Domain
Research
DistDF: Time-Series Forecasting Needs Joint-Distribution Wasserstein Alignment
Research
Text2Dashboard: A Governed Agent Architecture for Natural-Language Dashboard Generation over Enterprise DataBrain
Research
WavePrune: One period is often enough for RoPE
Research
Time-multiplexed layer reuse for physical neural networks
Research
Offline AI Modules: Voice-First Offline Architecture, Hardware Reference Stack, Quantization and Benchmarking
Research