About
Research & Development
SSGaussian: Advanced 3D Style Transfer for Coherent and Detailed Virtual Worlds
TLDR: SSGaussian is a novel 3D style transfer pipeline that addresses limitations in existing methods by effectively integrating prior knowledge from 2D diffusion models....
Research & Development
A Unified Framework for LLM Post-Training: Introducing Hybrid Adaptive Learning
TLDR: This research paper introduces the Unified Policy Gradient Estimator (UPGE), a theoretical framework that unifies Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL) as...
Research & Development
PG-Agent: Enhancing AI Navigation in User Interfaces with Page Graphs
TLDR: PG-Agent is a new AI framework that improves how agents interact with graphical user interfaces (GUIs). Instead of just using sequential action records,...
Research & Development
LLM Agents Learn to Plan Strategically, Not Constantly
TLDR: This research introduces a new method for training Large Language Model (LLM) agents to dynamically decide when to plan, rather than always planning...
Research & Development
Unveiling the Logic: Making Knowledge Graph-Augmented AI Transparent with KG-SMILE
TLDR: Researchers developed KG-SMILE, a new framework that makes Retrieval-Augmented Generation (RAG) systems using Knowledge Graphs (KGs) more transparent. By systematically altering parts of...
Research & Development
Embodied-LM: Bridging Language Models and Human-like Reasoning with Image Schemas
TLDR: Embodied-LM is a new neurosymbolic system that improves Large Language Models' (LLMs) logical reasoning by grounding their understanding in 'image schemas' – fundamental...