AI
A Physical Response-and-Memory Model for Muon Optimization
Researchers have proposed a physical model for optimizing weight updates during deep learning traini...
AI
Coarse Indexing, Fine Evidence: Decoupling Temporal Granularity in Long-Video RAG
Researchers have proposed a new approach to improve the efficiency of graph-based retrieval-augmente...
AI
SplitLite: Low-Rank Residual Compression for Split Learning
Researchers have developed SplitLite, a method that reduces the high communication costs associated ...
AI
Unlearning Is Not Just Erasing: Temporal Decoupling via Generation Inequality
Researchers have developed a new framework called ADU to help large language models 'forget' sensiti...
AI
Beyond Surface Cues: Disentangling Sociocultural Signals in Multilingual LLMs
Researchers have developed an audit to disentangle sociocultural signals in mult...
AI
FedCC: Towards Addressing Label Distribution Skews in Distillation-Based Federated Learning
Researchers propose a new algorithm called FedCC to address the issue of label d...
AI
The Multilingual FrameNet Corpus
Researchers have created a new dataset called the Multilingual FrameNet Corpus (...
AI
Beyond Verdicts: A Graph-Based Analysis of Human and LLM Reasoning in Scientific Fact-Checking
Researchers have developed a graph-based framework to compare how humans and lar...