About
Research & Development
Advancing Radiology Report Generation with a New Multi-modal Knowledge Graph
TLDR: R2GenKG is a new framework that uses a hierarchical multi-modal knowledge graph (M3KG) to improve X-ray medical report generation by large language models....
Research & Development
Beyond Discrete Tokens: Enhancing Reasoning in LLMs with Randomized Soft Thinking
TLDR: Large Language Models (LLMs) using "Soft Thinking" (reasoning with continuous probability distributions instead of discrete tokens) were thought to explore multiple reasoning paths...
Research & Development
SonicMaster: A Unified AI for Music Restoration and Mastering
TLDR: SonicMaster is the first unified, text-guided AI model for music restoration and mastering. It uses a generative flow-matching approach trained on a new...
Research & Development
Advancing Multilingual Fact-Checking: A Bi-Encoder Approach for Claim Retrieval
TLDR: This research paper details a bi-encoder model for SemEval-2025 Task 7, focusing on multilingual and cross-lingual fact-checked claim retrieval. The model, fine-tuned from...
Research & Development
VideoGuard: A New Defense Against Unauthorized Video Editing
TLDR: VideoGuard is a novel method that protects video content from unauthorized editing by generative AI models. It introduces subtle, nearly imperceptible perturbations into...
Research & Development
Unlocking Carbon Footprint Data: A New AI Model for Sustainability Reports
TLDR: A new research paper introduces CF-RAG, a method and dataset for answering carbon footprint questions from complex PDF sustainability reports. The CarbonPDF-QA dataset...