About
Research & Development
CommonKV: A Training-Free Approach to Efficient LLM Memory Management
TLDR: CommonKV is a novel, training-free method designed to significantly reduce the memory footprint of Large Language Models (LLMs) by compressing their Key-Value (KV)...
Research & Development
MIEL: Helping Robots Understand Ambiguous Instructions and Out-of-View Users
TLDR: MIEL is a new framework for robots to interpret ambiguous verbal instructions, like “Bring me that cup,” even when the user or object...
Research & Development
Adaptive Prompt Tuning: A New AI Framework for Smarter Anomaly Detection
TLDR: A new AI framework called Adaptive Prompt Tuning (APT) revolutionizes anomaly detection by moving beyond human-designed prompts. It uses self-generated anomaly samples and...
Research & Development
Advancing Nanoscale Localization in the Bloodstream with Set Transformers and AI-Generated Data
TLDR: This research explores using Set Transformer neural networks and synthetic data generation to improve Flow-guided Localization (FGL) of nanodevices in the human bloodstream....
Research & Development
MM-ORIENT: A New AI Framework for Deeper Multimodal Content Understanding
TLDR: MM-ORIENT is a novel AI framework designed to improve the semantic comprehension of multimodal content, such as memes, for multiple tasks like sentiment,...
Research & Development
Vevo2: A Unified Approach to Controllable Speech and Singing Voice Generation
TLDR: Vevo2 is a new framework that unifies controllable speech and singing voice generation. It uses two novel audio tokenizers—a notation-free prosody tokenizer and...