Through additional training, we explore embedding specialized scientific knowledge into the Llama 2 Large Language Model (LLM). Key findings reveal that effective knowledge integration requires reading texts from multiple perspectives, especially in instructional formats. We utilize text augmentation to tackle the scarcity of specialized texts, including style conversions and translations. Hyperparameter optimization proves crucial, with different size models (7b, 13b, and 70b) reasonably undergoing additional training. Validating our methods, we construct a dataset of 65,000 scientific papers. Although we have succeeded in partially embedding knowledge, the study highlights the complexities and limitations of incorporating specialized information into LLMs, suggesting areas for further improvement.

A company called PolyAI offers voice assistant technology that enables businesses to automate customer support. As its technology has already been taught on trillions of real-world interactions, no more training data is needed, and its voice assistants can swiftly pick up new languages. It comes with no maintenance costs and a low cost of ownership. It provides services to clients in the banking, hospitality, retail, telecommunications, insurance, and medical fields. Save my name, email, and website in this browser for the next time I comment.

A software developer with PolyAI who was testing the system, asked about booking a table for himself and a Serbian friend. "Yes, we allow children at the restaurant," the voice bot replied, according to PolyAI founder Nikola Mrksic. Seemingly out of nowhere, the bot was trying make an obnoxious joke about people from Serbia. When it was asked about bringing a Polish friend, it replied, "Yes, but you can't bring your own booze." Mrksic, who is Serbian, admits that the system appeared to think people from Serbia were immature.

Google Duplex impressed and scared the world in equal parts when it was unveiled, and now we've seen how a conversation goes with another chatbot. Duplex, for a quick primer, is Google's AI-powered voice bot which can call businesses on a person's behalf for things such as booking hair appointments. It's so realistic that everyone has decided that bots must declare themselves as such before chatting with a human. A company known as PolyAI – which specialises in "enterprise-ready voice assistants" – has posted an account of what happened when Duplex called one of its restaurant assistants. Duplex was calling businesses over the summer to update opening hours on Google Maps.

Running for the 11th time, the Startups 100 is an annual campaign designed to find the UK's most exciting new businesses. Having highlighted a huge number of fast-growing companies in its tenure thus far, this year rings another exciting cohort of successful startups to the fore. In 2019, AI and future tech startups stand out as a significant trend in British business. With other Startups 100 themes this year including disruptive fintech apps and the rise of health-related technologies, AI and future tech are in good company in the UK business market. AI startups – considered by many to sit under the umbrella of'future tech' – are at the cutting edge of both the business and technology worlds.

PolyAI, a London startup founded by experts in the field of "conversational AI" -- including CEO Nikola Mrkšić, who was previously the first engineer at Apple-acquired VocalIQ -- has raised $12 million in Series A funding to deploy its tech in customer support contact centres. The round was led by Point72 Ventures, with participation from Sands Capital Ventures, Amadeus Capital Partners, Passion Capital and Entrepreneur First (EF). PolyAI's founders are graduates of EF, although they didn't meet during the company building program but already knew each other from their time at Cambridge's Dialog Systems Group, part of the Machine Intelligence Lab at the University of Cambridge. "We started PolyAI in 2017, straight after submitting our PhD theses," Mrkšić tells me. "At Cambridge, we developed state-of-the-art conversational technology, and starting a company was the best way to get this tech used in the real world. We brought many of our Cambridge colleagues with us and started building the commercial version of our conversational platform."

This is a deep-dive into one of the problems we face when we model dialogue: understanding mentions of people's names in a restaurant booking system. This article presents how we approached the problem and solved it using some creative neural network structures. At PolyAI, we use datasets of billions of conversations and unstructured natural language texts to learn powerful deep neural models of conversational response. These models allow us to embed any conversational context or response into a shared high-dimensional vector space, so we can retrieve relevant responses, answers, entities and even photos from large databases comprising in-domain knowledge. Comparison of embedding vectors can also facilitate intent detection, i.e. classification of spoken language into specific categories such as'make a booking' or'confirm booking'. In this way, we can exploit a large ranker model and its internal implicit semantic vector space to solve many of the problems in dialogue, without hand-designing any explicit semantic structures like dialogue acts.