About
Research & Development
ReInAgent: A Collaborative AI for Smarter Mobile Task Automation
TLDR: ReInAgent is a new multi-agent AI framework designed to make mobile phone task automation more reliable and user-friendly. Unlike fully autonomous agents, ReInAgent...
Research & Development
AILoRA: Tailoring Low-Rank Adaptation for Enhanced LLM Efficiency
TLDR: AILoRA is a novel Parameter-Efficient Fine-Tuning (PEFT) method that improves upon traditional Low-Rank Adaptation (LoRA) for Large Language Models (LLMs). It addresses LoRA's...
Research & Development
Advancing Data Selection with Multi-Condition Conformal Methods
TLDR: A new algorithm, Multi-Condition Conformal Selection (MCCS), extends existing conformal selection methods to handle complex "multi-condition" scenarios (like "Y is between C1 and...
Research & Development
Balancing Speed and Depth in AI Conversations: The Prepared Mind, Fast Response Framework
TLDR: The PMFR (Prepared Mind, Fast Response) framework addresses the latency-quality tradeoff in open-domain dialogue AI by asynchronously orchestrating knowledge. It uses a lightweight...
Research & Development
Co-TAP: A New Protocol Framework for Intelligent Multi-Agent Systems
TLDR: Co-TAP is a three-layer agent interaction protocol (HAI, UAP, MEK) designed to overcome challenges in multi-agent systems. HAI standardizes human-agent interaction for real-time,...
Research & Development
Designing Intuitive Tools for Crafting AI Agent Experiences
TLDR: This research paper introduces AgentBuilder, a no-code prototyping system designed to help a wider range of individuals, beyond AI experts, create and refine...