Boost LLM Consistency with Group Relative Policy Optimization
Improve large language model reliability using Group Relative Policy Optimization for consistent, stable recommendations across varied prompts.
TPA: Detecting Hallucinations in RAG with Token Attribution
Discover TPA, a novel method for detecting hallucinations in Retrieval-Augmented Generation by attributing next token probabilities to key model components...
Post-Training Alignment for Medical Audio Understanding
Enhance medical audio models with AcuLa, a framework aligning audio encoders to language models for improved clinical diagnostics and AI healthcare innovat...
Reducing LLM Errors in Hardware Logic Design Automation
Discover how to mitigate hallucinations and omissions in LLMs for accurate hardware logic design automation, improving productivity and error detection.
Predictive Maintenance & Fault Detection in Heating Substations
Discover an open-source framework and labelled dataset enabling early fault detection in district heating substations for improved efficiency and reliabili...