About
Research & Development
KVComp: Boosting LLM Performance with Smart KV Cache Compression
TLDR: KVComp is a new high-performance, LLM-aware lossy compression framework for Key-Value (KV) cache in large language models. It tackles the significant memory bottleneck...
Research & Development
Sharing Health Data Securely: Private Kaplan-Meier Curves Across Institutions
TLDR: This research introduces a novel one-shot, node-level differentially private pipeline for calculating Kaplan-Meier survival curves across multiple healthcare institutions. It systematically compares four...
Research & Development
Fine-Tuning AI Music Style with Simple Descriptors
TLDR: This research introduces a method for fine-grained stylistic control in text-to-music generation using human-readable descriptors from a large language model, offering a policy-compliant...
Research & Development
Fusion to Enhance: A New Approach to Sharpen Multimodal AI’s Visual Perception
TLDR: A new research paper introduces Fusion to Enhance (FtZ), a novel vision tower framework for Multimodal Large Language Models (MLLMs). FtZ addresses the...
Research & Development
VECTOR+: A New AI Framework for Efficient Drug Design in Data-Scarce Environments
TLDR: VECTOR+ is a novel AI framework that combines contrastive learning and latent sampling to efficiently design new drug molecules, even with limited data....
Research & Development
Balancing Digital Discourse: How Queues and Graphs Can Reduce Online Anger
TLDR: This research introduces a framework called eImpact that combines a graph-based system for real-time emotion regulation with a comment queuing mechanism to reduce...