Artificial intelligence is either going to save humanity or finish it off, depending on who you speak to. Either way, every week there are new developments and breakthroughs. "Just accept the tech, architects!" Oliver Wainwright, our architecture and design critic, looks at whether AI will wipe out architects. Teaser: it can quickly show you what mosques in Abu Dhabi could look like, summarises local planning policies and allows the public to experiment with projects. If architects want to explore the endless world of AI, they can start by viewing AI as their perfectly on-time, organised and eager studio assistant.

Quantifying uncertainty about a policy's long-term performance is important to solve sequential decision-making tasks. We study the problem from a model-based Bayesian reinforcement learning perspective, where the goal is to learn the posterior distribution over value functions induced by parameter (epistemic) uncertainty of the Markov decision process. Previous work restricts the analysis to a few moments of the distribution over values or imposes a particular distribution shape, e.g., Gaussians. Inspired by distributional reinforcement learning, we introduce a Bellman operator whose fixed-point is the value distribution function. Based on our theory, we propose Epistemic Quantile-Regression (EQR), a model-based algorithm that learns a value distribution function that can be used for policy optimization. Evaluation across several continuous-control tasks shows performance benefits with respect to established model-based and model-free algorithms.

Information-seeking conversation, which aims to help users gather information through conversation, has achieved great progress in recent years. However, the research is still stymied by the scarcity of training data. To alleviate this problem, we propose AutoConv for synthetic conversation generation, which takes advantage of the few-shot learning ability and generation capacity of large language models (LLM). Specifically, we formulate the conversation generation problem as a language modeling task, then finetune an LLM with a few human conversations to capture the characteristics of the information-seeking process and use it for generating synthetic conversations with high quality. Experimental results on two frequently-used datasets verify that AutoConv has substantial improvements over strong baselines and alleviates the dependence on human annotation. In addition, we also provide several analysis studies to promote future research.

Recently developed large language models have achieved remarkable success in generating fluent and coherent text. However, these models often tend to 'hallucinate' which critically hampers their reliability. In this work, we address this crucial problem and propose an approach that actively detects and mitigates hallucinations during the generation process. Specifically, we first identify the candidates of potential hallucination leveraging the model's logit output values, check their correctness through a validation procedure, mitigate the detected hallucinations, and then continue with the generation process. Through extensive experiments with GPT-3.5 (text-davinci-003) on the 'article generation task', we first demonstrate the individual efficacy of our detection and mitigation techniques. Specifically, the detection technique achieves a recall of ~88% and the mitigation technique successfully mitigates 57.6% of the correctly detected hallucinations. Importantly, our mitigation technique does not introduce new hallucinations even in the case of incorrectly detected hallucinations, i.e., false positives. Then, we show that the proposed active detection and mitigation approach successfully reduces the hallucinations of the GPT-3.5 model from 47.5% to 14.5% on average. We further demonstrate the effectiveness and wide applicability of our approach through additional studies including performance on different types of questions (multi-hop and false premise questions) and with another LLM from a different model family (Vicuna). In summary, our work contributes to improving the reliability and trustworthiness of large language models, a crucial step en route to enabling their widespread adoption in real-world applications.

Recent years have witnessed the rapid development of heterogeneous graph neural networks (HGNNs) in information retrieval (IR) applications. Many existing HGNNs design a variety of tailor-made graph convolutions to capture structural and semantic information in heterogeneous graphs. However, existing HGNNs usually represent each node as a single vector in the multi-layer graph convolution calculation, which makes the high-level graph convolution layer fail to distinguish information from different relations and different orders, resulting in the information loss in the message passing. %insufficient mining of information. To this end, we propose a novel heterogeneous graph neural network with sequential node representation, namely Seq-HGNN. To avoid the information loss caused by the single vector node representation, we first design a sequential node representation learning mechanism to represent each node as a sequence of meta-path representations during the node message passing. Then we propose a heterogeneous representation fusion module, empowering Seq-HGNN to identify important meta-paths and aggregate their representations into a compact one. We conduct extensive experiments on four widely used datasets from Heterogeneous Graph Benchmark (HGB) and Open Graph Benchmark (OGB). Experimental results show that our proposed method outperforms state-of-the-art baselines in both accuracy and efficiency. The source code is available at https://github.com/nobrowning/SEQ_HGNN.

Mere hours after the arrival of "Heart on My Sleeve," the AI-generated "Drake" song that went viral last spring, the doomsday projections began pouring in. The main reason the song generated so much buzz is that Drake is one of the world's most popular musicians. But part of what gave us pause is that hip hop--which celebrates its 50th birthday this week--is driven by a spontaneity that feels as authentically human as anything humans have ever come up with. That is, rap is a unique form of human language, and if AI can mimic that, maybe nothing is safe. If the future is already here, then the impacts of generative AI will be even greater in the next decades, especially when it comes to hip hop.

A California-based company is developing a new drone for delivery services. Drone delivery is the way of the future, revolutionizing the speed and convenience of getting products and food right to your doorstep. Just ask Paul and Susie Sensmeier who've already used it over 1,200 times. CLICK TO GET KURT'S FREE CYBERGUY NEWSLETTER WITH SECURITY ALERTS, QUICK TIPS, TECH REVIEWS AND EASY HOW-TO'S TO MAKE YOU SMARTER They are early adopters of Wing's drone delivery service. The married couple from Virginia in their 80s have been using the drone delivery service since 2019, and now it's the only way they want to shop.

SHOPPING – Contains affiliated content. Products featured in this Mail Best article are selected by our shopping writers. If you make a purchase using links on this page, Dailymail.co.uk will earn an affiliate commission. Would you like a newly-improved Echo Dot that can wake you up, play music, audiobooks, answer your questions and more? Look no further than the Echo Dot (5th generation, 2022 release).

Influencer who used AI to dupe internet into thinking she caught her boyfriend cheating reveals she was inspired to do the skit because of real artificial voice scams. An influencer who made up a prank video claiming she used AI to catch her boyfriend cheating told Fox News her skit was a farce, but the voice-cloning technology she used to power the trick was not, and the skit was inspired by real scams. Mia Dio, a social media influencer with over 5 million TikTok followers, filmed a video of her using artificial intelligence to clone her boyfriend Billy's voice to see if he had cheated on her. The inspiration for the viral video came from reports of AI voice-cloning scams, she told Fox News. Dio used voicemails left by her boyfriend Billy to recreate his voice using AI software.

Actress Sandra Miska, a member of SAG-AFTRA, explains what it was like having her likeness scanned for AI and if she would agree to it again. The Writers Guild of America, which has been on strike for over three months, is meeting with the Alliance of Motion Picture and Television Producers (AMPTP) on Friday to potentially discuss next steps. Meanwhile, SAG-AFTRA, the actors' union that joined the WGA on strike last month, still hasn't seen any movement towards resuming talks with the AMPTP. One of the major sticking points: the potential use of artificial intelligence (AI) to replace background actors with digital copies. Some background actors have already been scanned by productions but remain unaware if or how their likeness is being used, causing concern about their future career prospects.