Tag: language model interpretability
Browse our exclusive articles!
Enhancing Harmonic Loss with Non-Euclidean Distance Metrics
Explore how non-Euclidean distance layers improve harmonic loss for better model accuracy, interpretability, and sustainability in deep learning.
DecompKAN: Accurate Long-Term Time Series Forecasting Model
Discover DecompKAN, a transparent and accurate model for long-term time series forecasting with trend decomposition and patch-wise analysis.
How Representational Curvature Affects Uncertainty in LLMs
Explore how representational curvature influences behavioral uncertainty and token prediction accuracy in large language models like GPT-2 and Pythia.
Systematic Debugging Techniques for Large Language Models
Discover a structured, model-agnostic approach to effectively debug large language models, enhancing transparency, scalability, and error analysis.
Feature Attribution Benefits in Supervised Contrastive Learning
Explore how supervised contrastive learning improves feature attribution for better model interpretability, robustness, and trustworthiness in AI systems.