About
Research & Development
Enhancing Aviation Safety AI with Logic-Constrained Transformers
TLDR: NASP-T is a neuro-symbolic transformer framework that improves the classification of aviation safety reports by integrating Answer Set Programming (ASP) rules. It uses...
Research & Development
Forecasting Taiwan’s CO2 Emissions: A Decade-Long Outlook with Advanced Machine Learning
TLDR: This research paper presents an advanced ensemble model for forecasting CO2 emissions in Taiwan over the next decade (2023-2032). By comparing 21 univariate...
Research & Development
MixReasoning: A Smart Approach to Efficient Language Model Thinking
TLDR: MixReasoning is a new framework that allows large language models (LLMs) to dynamically adjust their reasoning depth within a single response. Instead of...
Research & Development
Unpacking LLM Memory: New Research Reveals Early Forgetting in Complex Reasoning Tasks
TLDR: A new study introduces a graph reconstruction benchmark to evaluate LLMs' ability to induce structured knowledge from long, noisy texts. It finds that...
Research & Development
The Surprising Disconnect: Interpretability vs. Utility in AI Steering
TLDR: Research on Sparse Autoencoders (SAEs) in Large Language Models (LLMs) reveals that higher interpretability does not necessarily lead to better steering utility. A...
Research & Development
Automated Safety Monitoring: A Vision-Language Framework for Mining Operations
TLDR: MonitorVLM is a new vision-language framework designed to automatically detect safety violations in mining operations from surveillance video streams. It introduces three key...