Britain is to develop new air defence weapons alongside the EU's four biggest military powers, deepening ties with the European defence sector. The project will invite manufacturers in the UK, Germany, France, Italy and Poland to submit plans to build low-cost missiles and autonomous drones. The allies are pledging a speedy process to build the weapons together, inspired by Ukraine's development of cheap drones to counter attacks from Russia. The UK's Ministry of Defence (MoD) says the programme will prioritise a lightweight, affordable surface-to-air weapon, with the first project to be delivered by next year. The plan, announced at a meeting of the five countries' defence ministers in the Polish city of Krakow, marks a boost to UK-Europe ties after the failure of talks last year over UK participation in the EU's new €150bn (£130bn) defence fund.

Instruction tuning for large language models (LLMs) can drive them to produce results consistent with human goals in specific downstream tasks. However, the process of continual instruction tuning (CIT) for LLMs may bring about the catastrophic forgetting (CF) problem, where previously learned abilities are degraded. Recent methods try to alleviate the CF problem by modifying models or replaying data, which may only remember the surface-level pattern of instructions and get confused on held-out tasks. In this paper, we propose a novel continual instruction tuning method based on Key-part Information Gain (KPIG). Our method computes the information gain on masked parts to dynamically replay data and refine the training objective, which enables LLMs to capture task-aware information relevant to the correct response and alleviate overfitting to general descriptions in instructions. In addition, we propose two metrics, P-score and V-score, to measure the generalization and instruction-following abilities of LLMs. Experiments demonstrate our method achieves superior performance on both seen and held-out tasks.

AI is already having an impact on the recruitment industry by automating and streamlining key parts of the hiring process. While this has led to better efficiency for employers, it also raises concerns about bias and job displacement-two issues that must be addressed right now before they become widespread problems. AI has become more common in the recruitment industry, and it's here to stay. As a result of AI's growing presence in recruiting, many companies are looking for ways to use it to help them with their hiring needs. Let's say an employer is looking to hire a Branding Expert – whether they are using AI for automated resume screening and basic vetting, or using it to provide company culture fit insights into applicants' personality traits and behaviours, it can help them learn more about potential employees before making an offer. In recruitment, AI has the potential to reduce bias in hiring decisions.

As machine learning becomes a more integral part of running businesses, the model-building process still requires iteration and experimentation. Comet has created an entire platform to get models from idea to product, and today the company announced a $50 million Series B. The investment comes on the heels of the company's $13 million A round in April. OpenView led the round with participation from existing investors Scale Venture Partners, Trilogy Equity Partners and Two Sigma Ventures. The company has now raised almost $70 million, according to Crunchbase data. Company co-founder and CEO Gideon Mendels said that the product works on any platform from a laptop to the cloud to an on-prem cluster.

A range of the biggest and most advanced companies are doubling down on artificial intelligence and making it a key part of their research and development, if not the primary focus of their products and services. A prime example is Telsa, where Elon Musk has been making artificial intelligence a key part of the product development business. As reported in Investor Place, "Musk and company could have an AI prototype ready by 2022…Ten years ago, you could have counted the number of these advanced stocks on one or two hands. Now, however, it seems that many companies of various sizes are working with this kind of technology." And where technology companies take a concept and build it into their core product offerings, the professionals working in those arenas come too, with a huge boost in opportunities to work in these new arenas, and the potential to fast-track your own path.

TORONTO and LONDON, Aug. 20, 2021 (GLOBE NEWSWIRE) -- ACCA (the Association of Chartered Certified Accountants) and Chartered Accountants Australia and New Zealand (CA ANZ) reveal in a new report the pressing need for the accountancy profession to make the necessary connections between Artificial Intelligence (AI) and its relationship to environmental, social and governance (ESG) dimensions. Polling over 5,700 respondents globally, including an expert panel of ACCA members across North America, the research reveals a cautious tone, with fewer than half (43%) believing that the impact of AI on their rights as an individual is positive – such as safety and personal security, levels of fairness, levels of choice, levels of transparency. In North America, 30% of respondents believe this to be the case. ACCA and CA ANZ say in Ethics for sustainable AI adoption: Connecting AI and ESG that accountants, with their explicit and long-standing commitment to ethical practices, are well placed to guide organizations along a responsible path for AI adoption. Jillian Couse, Head of ACCA North America says: 'Our findings present a wake-up call for the accountancy profession to lead the way and become the super connectors needed to ensure an ethical approach.

TMI's editorial team is comprised of researchers, writers, editors, proofreaders, and contributors with extensive experience in the HR and Talent Management space. The team scrupulously researches, analyzes and deliberates on each submission before concluding its suitability for our publications. In the early part of 2020, the war for talent was as heightened as ever, and poised to become even more so. Several industries and sectors expected ambitious hiring to cater to strong growth. And automation, artificial intelligence (AI), and other forms of HR technology were frequently tapped to aid HR in efficient, quick, and scalable discharge of HR duties.

For three years, I've spoken at schools about video games and social media. The small groups of parents who have attended often tell me similar stories: their sons and daughters are "addicted" to Fortnite, to Pokemon, to Minecraft, and they feel powerless about what they can do. When I tell them about parental controls – built-in tools on all games machines that let you limit access to games – many are bewildered or agitated, even though such controls have been a feature of console design for a decade. Video game trade body Ukie has launched a campaign to encourage more parents and carers to use parental controls, teaming up with ex-footballer Rio Ferdinand. The organisation wants to raise awareness that everything from screen time to in-game purchases can be monitored and controlled.

With the new year approaching and a new decade on the horizon (yes, technically it starts in 2021), we wanted to look at transformations in media that are likely to define the 2020s. Especially in the media and computing fields, the decade proves to be a useful marker. Consider that at the end of 2009, Digital Cinema and 3D film was just being introduced with the release of Avatar, GPU rendering was in its infancy, YouTube was less than 1/20th the size it is today, and Netflix video streaming had yet to be launched internationally. Meanwhile, cloud computing had just started its adoption curve, the first augmented reality print campaign was launched, and blockchain mining was in its first year. It's fair to expect that when we look back a decade later, the end of 2019 will similarly look distant - if not more remote with the accelerating rate of change in media and technology.

A council in England is using virtual assistants (VAs) to check it is paying staff and schools correctly. Wiltshire Council is using Microsoft technology to complete repetitive tasks such as checking payrolls in its Human Resources department. This frees up staff to work on more critical tasks that directly help the 435,000 residents in Wiltshire and the council. VAs are currently checking more than 43 payrolls each month, which cover schools and academies in the region as well as council staff. Payroll checks need to be completed before payments can be made to ensure the correct amount of money is paid to the right person or organisation.