About
Research & Development
AI’s Hidden Weaknesses: How Adversarial Attacks Both Threaten and Protect Computer Vision
TLDR: This paper surveys adversarial attacks in computer vision, showing how subtle, often invisible, manipulations can fool AI systems. It categorizes attacks by their...
Research & Development
ChairPose: A Pressure-Based System for Detailed Sitting Posture Analysis
TLDR: ChairPose is a new system that accurately estimates full-body sitting posture using only a pressure-sensing mat, eliminating the need for cameras or wearable...
Research & Development
Advancing Web Agents with Human-Inspired Cognitive Learning
TLDR: The research introduces Web-CogReasoner, a novel web agent trained using the Web-CogKnowledge Framework, inspired by Bloom's Taxonomy of human learning. This framework categorizes...
Research & Development
Empowering Service Robots with Lifelong Learning and Planning through Knowledge Graphs
TLDR: L3M+P is a new framework that enhances large language models (LLMs) for lifelong robot planning. It uses an external knowledge graph to represent...
Research & Development
Unpacking Word Overuse in AI Models: The Role of Human Feedback
TLDR: A study found that Large Language Models (LLMs) overuse certain words like "delve" and "nuanced" because human evaluators, during Learning from Human Feedback...
Research & Development
SpeechR: Unpacking AI’s Ability to Reason from Spoken Language
TLDR: SpeechR is a new benchmark evaluating how well large audio-language models (LALMs) reason from speech. It tests factual, procedural, and normative reasoning using...