Large language models (LLMs) have been widely deployed in various applications, often functioning as autonomous agents that interact with each other in multi-agent systems. While these systems have shown promise in enhancing capabilities and enabling complex tasks, they also pose significant ethical challenges. This position paper outlines a research agenda aimed at ensuring the ethical behavior of multi-agent systems of LLMs (MALMs) from the perspective of mechanistic interpretability. We identify three key research challenges: (i) developing comprehensive evaluation frameworks to assess ethical behavior at individual, interactional, and systemic levels; (ii) elucidating the internal mechanisms that give rise to emergent behaviors through mechanistic interpretability; and (iii) implementing targeted parameter-efficient alignment techniques to steer MALMs towards ethical behaviors without compromising their performance.

Large language models (LLMs) are prone to three types of hallucination: Input-Conflicting, Context-Conflicting and Fact-Conflicting hallucinations. The purpose of this study is to mitigate the different types of hallucination by exploiting the interdependence between them. For this purpose, we propose a Multi-Information Adapter for Large Language Models (MALM). This framework employs a tailored multi-graph learning approach designed to elucidate the interconnections between original inputs, contextual information, and external factual knowledge, thereby alleviating the three categories of hallucination within a cohesive framework. Experiments were carried out on four benchmarking datasets: HaluEval, TruthfulQA, Natural Questions, and TriviaQA. We evaluated the proposed framework in two aspects: (1) adaptability to different base LLMs on HaluEval and TruthfulQA, to confirm if MALM is effective when applied on 7 typical LLMs. MALM showed significant improvements over LLaMA-2; (2) generalizability to retrieval-augmented generation (RAG) by combining MALM with three representative retrievers (BM25, Spider and DPR) separately. Furthermore, automated and human evaluations were conducted to substantiate the correctness of experimental results, where GPT-4 and 3 human volunteers judged which response was better between LLaMA-2 and MALM. The results showed that both GPT-4 and human preferred MALM in 79.4% and 65.6% of cases respectively. The results validate that incorporating the complex interactions between the three types of hallucination through a multilayered graph attention network into the LLM generation process is effective to mitigate the them. The adapter design of the proposed approach is also proven flexible and robust across different base LLMs.

More of the creative work these days is not being done by humans at all. When Dixons Carphone wanted to push shoppers towards its Black Friday sale, the company turned to Artificial Intelligence (AI) software and got the winning line "The time is now". Saul Lopes, head of customer marketing at Dixons Carphone, thinks it worked because it didn't have the words Black Friday in it. His human copywriters had produced dozens of potentially successful sentences but they all mentioned Black Friday. It was technology that broke this chain of thought.

The Transform Technology Summits start October 13th with Low-Code/No Code: Enabling Enterprise Agility. Retailers are already preparing for a 2021 holiday ecommerce season that mirrors last year. A recent survey by Radial found that 65% of consumers plan to spend the same as or even more than last year. Radial itself plans to hire 27,000 seasonal workers ahead of the shopping season to fulfill ecommerce orders. But to snag a share of that market, retailers will need to use the right email marketing language and tactics, according to a new report from AI-powered copywriting platform Phrasee.

Artificial intelligence (AI) has been used for decades now to augment human intelligence and prowess. In a recent article, the BBC questions whether it can replace humans in creative endeavours such as copywriting, and comes up with a collaborative solution. A carphone company was seeking a catchphrase for a Black Friday sale, and all the options human copywriters were coming up with contained the words'Black Friday.' Stepping in to save the day was a software company whose technology ran through thousands of options and came up with "The Time is Now." Phrasee is the brainchild of Parry Malm, a Canadian living in the UK. Malm, who works in marketing, was frustrated that technology to boost human creativity didn't already exist, and set out to create the software in 2015.

What springs to mind when you think of advertising? Or perhaps trendy people swapping catch phrases in a converted warehouse?' Well, more of the creative work these days is not being done by humans at all. When Dixons Carphone wanted to push shoppers towards its Black Friday sale, the company turned to Artificial Intelligence (AI) software and got the winning line "The time is now". Saul Lopes, head of customer marketing at Dixons Carphone, thinks it worked because it didn't have the words Black Friday in it.

There's a science behind getting people to open your emails. One tech startup is helping companies master that. London-based marketing technology software company Phrasee uses artificial intelligence to deliver human-sounding copy for companies like eBay and Domino's. The software injects AI-optimized language into a company's email subject lines, push notifications, and ads to boost clicks and conversations from copy. And it can do it all in matter of minutes.

We're all aware that artificial intelligence (AI) is predicted to transform every sector of the economy, and marketing is no exception. Already, the industry is using this new technology to analyse data or give product recommendations to customers. Of course, as adoption of the technology increases, so too is anxiety over the future of jobs. Many workers fear that AI-powered robots might replace them. But what about the actual creation of ads? Surely that will remain the preserve of humans?

Can artificial intelligence replace the human brain?Will it? "Humans were are not built to spend more than two hours looking at a screen or scrolling through excel sheets. Humans are best at being human. Artificial Intelligence will do the rest." Telling words from Jim Stolze, Co-founder of aigency -- an Amsterdam-based company that recruits AI and humans for work.

"Humans were are not built to spend more than two hours looking at a screen or scrolling through excel sheets. Humans are best at being human. Artificial Intelligence will do the rest." Kind of an employment company run by three humans overseeing 59 robots (actually computers working on algorithms created at the University of Amsterdam to solve problems). Stolze was addressing reporters in StartUp Village at the Amsterdam Science Park on the sidelines of the first World Summit AI in Amsterdam October 11-12.