Click-through rate (CTR) prediction plays as a core function module in various personalized online services. The traditional ID-based models for CTR prediction take as inputs the one-hot encoded ID features of tabular modality, which capture the collaborative signals via feature interaction modeling. But the one-hot encoding discards the semantic information conceived in the original feature texts. Recently, the emergence of Pretrained Language Models (PLMs) has given rise to another paradigm, which takes as inputs the sentences of textual modality obtained by hard prompt templates and adopts PLMs to extract the semantic knowledge. However, PLMs generally tokenize the input text data into subword tokens and ignore field-wise collaborative signals. Therefore, these two lines of research focus on different characteristics of the same input data (i.e., textual and tabular modalities), forming a distinct complementary relationship with each other. In this paper, we propose to conduct Fine-grained feature-level ALignment between ID-based Models and Pretrained Language Models (FLIP) for CTR prediction. We design a novel joint reconstruction pretraining task for both masked language and tabular modeling. Specifically, the masked data of one modality (i.e., tokens or features) has to be recovered with the help of the other modality, which establishes the feature-level interaction and alignment via sufficient mutual information extraction between dual modalities. Moreover, we propose to jointly finetune the ID-based model and PLM for downstream CTR prediction tasks, thus achieving superior performance by combining the advantages of both models. Extensive experiments on three real-world datasets demonstrate that FLIP outperforms SOTA baselines, and is highly compatible for various ID-based models and PLMs.

This paper introduces WordArt Designer, a user-driven framework for artistic typography synthesis, relying on the Large Language Model (LLM). The system incorporates four key modules: the LLM Engine, SemTypo, StyTypo, and TexTypo modules. 1) The LLM Engine, empowered by the LLM (e.g., GPT-3.5), interprets user inputs and generates actionable prompts for the other modules, thereby transforming abstract concepts into tangible designs. 2) The SemTypo module optimizes font designs using semantic concepts, striking a balance between artistic transformation and readability. 3) Building on the semantic layout provided by the SemTypo module, the StyTypo module creates smooth, refined images. 4) The TexTypo module further enhances the design's aesthetics through texture rendering, enabling the generation of inventive textured fonts. Notably, WordArt Designer highlights the fusion of generative AI with artistic typography. Experience its capabilities on ModelScope: https://www.modelscope.cn/studios/WordArt/WordArt.

Despite breakthroughs in audio generation models, their capabilities are often confined to domain-specific conditions such as speech transcriptions and audio captions. However, real-world audio creation aims to generate harmonious audio containing various elements such as speech, music, and sound effects with controllable conditions, which is challenging to address using existing audio generation systems. We present WavJourney, a novel framework that leverages Large Language Models (LLMs) to connect various audio models for audio creation. WavJourney allows users to create storytelling audio content with diverse audio elements simply from textual descriptions. Specifically, given a text instruction, WavJourney first prompts LLMs to generate an audio script that serves as a structured semantic representation of audio elements. The audio script is then converted into a computer program, where each line of the program calls a task-specific audio generation model or computational operation function. The computer program is then executed to obtain a compositional and interpretable solution for audio creation. Experimental results suggest that WavJourney is capable of synthesizing realistic audio aligned with textually-described semantic, spatial and temporal conditions, achieving state-of-the-art results on text-to-audio generation benchmarks. Additionally, we introduce a new multi-genre story benchmark. Subjective evaluations demonstrate the potential of WavJourney in crafting engaging storytelling audio content from text. We further demonstrate that WavJourney can facilitate human-machine co-creation in multi-round dialogues. To foster future research, the code and synthesized audio are available at: https://audio-agi.github.io/WavJourney_demopage/.

Best Buy's Black Friday deals have already been running for weeks and show no sign of slowing down. The classic big box electronics store is going head-to-head with Amazon and has been touting a price match guarantee--at this point, though, we know where the deals are going to land. If you have a paid My Best Buy Membership, you can access some exclusive deals, but they're not worth the yearly fee if you haven't already joined. We test products year-round and handpicked these deals. The discount amounts we show are based on actual street prices at retailers in the past few months. Products that are sold out or no longer discounted as of publishing will be crossed out. We'll update this guide periodically. If you buy something using links in our stories, we may earn a commission. This helps support our journalism. This is our top TV recommendation if you're looking for the best bang for your buck. The Hisense U8K (8/10, WIRED Recommends) has a mini-LED display that delivers excellent black levels, and it comes with a Google TV interface that's easy to use. The feet are on the ends though, so you'll want to mount it or pair it with a long TV stand. We've been fans of TVs with Roku's interface for a while, and Roku's in-house line makes our favorite entry-level TV.

Every single year I tell myself that I'm not going to go overboard for the holidays. Then I start shopping for 20 people. If you're in the same boat, you'll be happy to know we've tracked down the Black Friday deals under $50 on reviewer-approved gadgets. WIRED tests products year-round and handpicked these deals based on the actual discounts. Products that are sold out or no longer discounted as of publishing will be crossed out . We'll update this guide through Black Friday. If you buy something using links in our stories, we may earn a commission. This helps support our journalism. If you watch mainly Amazon Prime content, the Fire TV Stick is the streaming device you need, and this one was just released at the end of September. You can stream from any service, of course, but it's heavily geared toward the company's own content. This 4K Max is faster, with 16 gigabytes of storage versus 8, but if you want to spend less and still stay in the Amazon ecosystem, the also-new 4K Stick is on sale for $25 ($25 off). The Echo Show 5 is typically overpriced at $90, but the deal price of $40 matches the best we have tracked. Just keep in mind that it goes on sale quite frequently. We prefer the Echo Show 10 for its larger size, but the smaller footprint of the Echo Show 5 is good for a nightstand or tiny kitchen.

Nicolas Cage knows he's a meme. After making the mistake of googling himself a few years back, the charismatic actor discovered that his big on-screen performances had been translated into single-frame quips and supercuts, taken--like all memes, really--out of context, played for lolz, and in a manner that, frankly, makes Cage seem like a graduate from the Jim Carrey school of rubber-faced acting. "Something like'Nick Cage loses his shit,' where they cherry-pick meltdowns from different movies I'd made over the years," he says. "I get that it's all done for laughs, and in that context it is funny, but at the same time, there's no regard to how the character got there. This, Cage says, is not why he got into making movies.

When a beloved actor or musician passes away, they always leave behind that lingering thought of what they might have been able to create if they'd only had a little more time. However, as John Lennon's posthumous chart success has shown, for the stars of the future, death doesn't need to be the final curtain call. From actors making ghostly returns to the screen or famous figures narrating the story of their own lives, AI is reviving more than just celebs' careers. But when stardom doesn't end with death, who gets the final say on a celebrity's legacy? Here, MailOnline reveals how these creepy technologies are bringing back your favourite figures from the past to perform again.

Large language models (LLMs) have demonstrated remarkable capabilities across a broad spectrum of tasks. They have attracted significant attention and been deployed in numerous downstream applications. Nevertheless, akin to a double-edged sword, LLMs also present potential risks. They could suffer from private data leaks or yield inappropriate, harmful, or misleading content. Additionally, the rapid progress of LLMs raises concerns about the potential emergence of superintelligent systems without adequate safeguards. To effectively capitalize on LLM capacities as well as ensure their safe and beneficial development, it is critical to conduct a rigorous and comprehensive evaluation of LLMs. This survey endeavors to offer a panoramic perspective on the evaluation of LLMs. We categorize the evaluation of LLMs into three major groups: knowledge and capability evaluation, alignment evaluation and safety evaluation. In addition to the comprehensive review on the evaluation methodologies and benchmarks on these three aspects, we collate a compendium of evaluations pertaining to LLMs' performance in specialized domains, and discuss the construction of comprehensive evaluation platforms that cover LLM evaluations on capabilities, alignment, safety, and applicability. We hope that this comprehensive overview will stimulate further research interests in the evaluation of LLMs, with the ultimate goal of making evaluation serve as a cornerstone in guiding the responsible development of LLMs. We envision that this will channel their evolution into a direction that maximizes societal benefit while minimizing potential risks. A curated list of related papers has been publicly available at https://github.com/tjunlp-lab/Awesome-LLMs-Evaluation-Papers.

Black Friday means savings on cameras for content creation, travel photography and more. This year is no exception as all the major manufacturers, including Sony, Canon, DJI, Nikon, GoPro, Fujifilm and Nikon have some stellar deals. Sony is offering its latest compact vlogging camera, the ZV-1F, at just $398 ($100 off), while Canon's new EOS R100 mirrorless APS-C camera is available with a kit lens for just $449. Panasonic has its full frame Lumix S5 on sale for $1,298 (38 percent off), DJI's Avata Pro-View combo is $999 (30 percent off) and GoPro's Hero 11 is down to $300, for a savings of 14 percent. Sony's ZV-1F for budding content creators is 20 percent off for Black Friday.

Fox News Flash top sports headlines are here. Check out what's clicking on Foxnews.com. Over the past nearly three decades, FOX Sports, a unit of FOX Corp., parent to Fox News and FOX Business, has accumulated a countless amount of video footage. Millions of hours' worth of sports-related content live within vast archives. At any given time, various individuals have been tasked with sorting through the seemingly endless amount of footage in order to produce new pieces of content.