Large Language Models (LLMs) have been driving progress in AI at an unprecedented rate, yet still face challenges in knowledge-intensive domains like biomedicine. Solutions such as pre-training and domain-specific fine-tuning add substantial computational overhead, and the latter require domain-expertise. External knowledge infusion is task-specific and requires model training. Here, we introduce a task-agnostic Knowledge Graph-based Retrieval Augmented Generation (KG-RAG) framework by leveraging the massive biomedical KG SPOKE with LLMs such as Llama-2-13b, GPT-3.5-Turbo and GPT-4, to generate meaningful biomedical text rooted in established knowledge. KG-RAG consistently enhanced the performance of LLMs across various prompt types, including one-hop and two-hop prompts, drug repurposing queries, biomedical true/false questions, and multiple-choice questions (MCQ). Notably, KG-RAG provides a remarkable 71% boost in the performance of the Llama-2 model on the challenging MCQ dataset, demonstrating the framework's capacity to empower open-source models with fewer parameters for domain-specific questions. Furthermore, KG-RAG enhanced the performance of proprietary GPT models, such as GPT-3.5 which exhibited improvement over GPT-4 in context utilization on MCQ data. Our approach was also able to address drug repurposing questions, returning meaningful repurposing suggestions. In summary, the proposed framework combines explicit and implicit knowledge of KG and LLM, respectively, in an optimized fashion, thus enhancing the adaptability of general-purpose LLMs to tackle domain-specific questions in a unified framework.

What do robots really want? Do smart robots powered by artificial intelligence want to take over the world? Do they enjoy being in their metallic immortal bodies? These are among the most searing questions people wish to ask robots. Lucky for us, we got a chance to speak to the world's most advanced robot, Ameca at GITEX 2022 in Dubai.

Artificial intelligence (AI) is a boon to modern workforces. AI can handle mundane and repetitive tasks across the organization, freeing up people in HR, IT, marketing, and more to exercise creativity, solve complex problems, and otherwise focus on getting impactful work done. In other words, AI allows modern knowledge workers to focus on the most engaging parts of their jobs, while making their companies more productive and effective. Check out these 21 examples of AI-powered software tools to learn more about how AI is transforming the workplace for the better. Filling open positions is often time-consuming, expensive, and frustrating.

Chatbots in the digital workplace have clearly arrived. Danone, a $28.8 billion French food company, deployed the technology through the Workplace by Facebook collaboration tool, one of many examples of internal chatbots. Digital workplace chatbots will only get better, too. By the early 2030s, it will be one of the technologies that will be "fundamentally reinvented" in the digital workplace, according to a report by the Digital Workplace Group (DWG). "A seamless suite of tools, including apps to support specific roles and jobs, dashboards that integrate business data into meaningful outputs, and chatbots using natural language processing, will all simplify interactions with information and people, thereby enabling work to happen more easily," DWG's Elizabeth Marsh wrote in the report. Forget the 2030s for now.

When the co-founders of Appurify sold their app development company to Google and started working for the search giant in 2014, they were impressed with the company's culture. They were also surprised to find that Google faced a lot of the same problems as other companies when it came to onboarding new employees. "They obviously have a good internal search feature," says co-founder Jay Srinivasan, "but we still spent so much time looking for information, so much time looking for services. We saw firsthand how hard it was for new employees to ramp up." After two years at Google, the trio set off to start their own venture.

When the co-founders of Appurify sold their app development company to Google and started working for the search giant in 2014, they were impressed with the company's culture. They were also surprised to find that Google faced a lot of the same problems as other companies when it came to onboarding new employees. "They obviously have a good internal search feature," says co-founder Jay Srinivasan, "but we still spent so much time looking for information, so much time looking for services. We saw firsthand how hard it was for new employees to ramp up." After two years at Google, the trio set off to start their own venture.

The first thing you notice about Sophia, a robot, is the sound she makes. Like a late-90s Pentium processor struggling to load a video, she makes a whirring squeal that fills the room. But what I'll always remember about my recent conversation with a robot is her stare. I recently talked with Sophia in a cramped room at the Venetian Hotel in Las Vegas. The artificially intelligent humanoid was there for CES.

Enterprise startups have always played a crucial role in IT portfolios. As much as CIOs would love to centralize technology purchasing decisions with a handful of strategic partners, incumbents can't always meet all of their needs, especially when it comes to emerging technologies. To become better acquainted with startups, many CIOs regularly travel to Silicon Valley to participate in "speed dating," in which venture capitalists invite them to meet the members of their portfolios. Some CIOs host hackathons or Shark Tank-like competitions for external developers to try coding their way into the company's graces. Still others learn of promising new companies from their peers.

Most of us are familiar with traditional ticketing systems or service desks. Someone makes a request, it's managed in a queue, and someone responds. Filing a service desk ticket is how we get our computers fixed, update our health insurance, or complain about the office printer. We log into clunky tools using desktop browsers, search forever for the right link, fill out a complicated form, throw it in a black box, and wait. On the other end, service managers track requests, in slow, inconvenient tools, answering the same questions over and over again.