Quincy Surasmith is a radio journalist and actor, but you may also hear his voice and never realize it. That's because he's been the voice of Thai-speaking cartoons, chattering background crowds, and characters without major speaking roles. "I'm making grunting noises, getting beat up by some guy," Suarasmith says. "It takes specific improv and acting skills." Soon those grunting and background chatter performances could be at risk of being replaced by artificial intelligence.

ChatGPT is one of the most popular language models which achieve amazing performance on various natural language tasks. Consequently, there is also an urgent need to detect the texts generated ChatGPT from human written. One of the extensively studied methods trains classification models to distinguish both. However, existing studies also demonstrate that the trained models may suffer from distribution shifts (during test), i.e., they are ineffective to predict the generated texts from unseen language tasks or topics. In this work, we aim to have a comprehensive investigation on these methods' generalization behaviors under distribution shift caused by a wide range of factors, including prompts, text lengths, topics, and language tasks. To achieve this goal, we first collect a new dataset with human and ChatGPT texts, and then we conduct extensive studies on the collected dataset. Our studies unveil insightful findings which provide guidance for developing future methodologies or data collection strategies for ChatGPT detection.

Despite the advancements of open-source large language models (LLMs), e.g., LLaMA, they remain significantly limited in tool-use capabilities, i.e., using external tools (APIs) to fulfill human instructions. The reason is that current instruction tuning largely focuses on basic language tasks but ignores the tool-use domain. This is in contrast to the excellent tool-use capabilities of state-of-the-art (SOTA) closed-source LLMs, e.g., ChatGPT. To bridge this gap, we introduce ToolLLM, a general tool-use framework encompassing data construction, model training, and evaluation. We first present ToolBench, an instruction-tuning dataset for tool use, which is constructed automatically using ChatGPT. Specifically, the construction can be divided into three stages: (i) API collection: we collect 16,464 real-world RESTful APIs spanning 49 categories from RapidAPI Hub; (ii) instruction generation: we prompt ChatGPT to generate diverse instructions involving these APIs, covering both single-tool and multi-tool scenarios; (iii) solution path annotation: we use ChatGPT to search for a valid solution path (chain of API calls) for each instruction. To enhance the reasoning capabilities of LLMs, we develop a novel depth-first search-based decision tree algorithm. It enables LLMs to evaluate multiple reasoning traces and expand the search space. Moreover, to evaluate the tool-use capabilities of LLMs, we develop an automatic evaluator: ToolEval. Based on ToolBench, we fine-tune LLaMA to obtain an LLM ToolLLaMA, and equip it with a neural API retriever to recommend appropriate APIs for each instruction. Experiments show that ToolLLaMA demonstrates a remarkable ability to execute complex instructions and generalize to unseen APIs, and exhibits comparable performance to ChatGPT. Our ToolLLaMA also demonstrates strong zero-shot generalization ability in an out-of-distribution tool-use dataset: APIBench.

Camera Image Signal Processing (ISP) pipelines can get appealing results in different image signal processing tasks. Nonetheless, the majority of these methods, including those employing an encoder-decoder deep architecture for the task, typically utilize a uniform filter applied consistently across the entire image. However, it is natural to view a camera image as heterogeneous, as the color intensity and the artificial noise are distributed vastly differently, even across the two-dimensional domain of a single image. Varied Moire ringing, motion blur, color-bleaching, or lens-based projection distortions can all potentially lead to a heterogeneous image artifact filtering problem. In this paper, we present a specific patch-based, local subspace deep neural network that improves Camera ISP to be robust to heterogeneous artifacts (especially image denoising). We call our three-fold deep-trained model the Patch Subspace Learning Autoencoder (PSL-AE). The PSL-AE model does not make assumptions regarding uniform levels of image distortion. Instead, it first encodes patches extracted from noisy a nd clean image pairs, with different artifact types or distortion levels, by contrastive learning. Then, the patches of each image are encoded into corresponding soft clusters within their suitable latent sub-space, utilizing a prior mixture model. Furthermore, the decoders undergo training in an unsupervised manner, specifically trained for the image patches present in each cluster. The experiments highlight the adaptability and efficacy through enhanced heterogeneous filtering, both from synthesized artifacts but also realistic SIDD image pairs.

Correspondent Griff Jenkins caught up with the singer-songwriter to discuss the inspiration behind his music. The University of North Carolina in Chapel Hill released photos of a suspect who allegedly committed a sexual assault at one of the campus's residence halls on Monday night. At about 10:40 p.m. on Monday, police put out an alert to students, faculty and staff, saying they were investigating a report of a groping or sexual assault at McClinton Residence Hall. The incident occurred at about 6:10 p.m., and according to the preliminary investigation, the suspect followed the victim into the building's lobby and stairwell. UNC Police are searching for man who allegedly followed a student into a residence hall, groped her and left on Oct. 1, 2023.

Fox News Flash top headlines are here. Check out what's clicking on Foxnews.com. If you have a pulse and an internet connection, companies want all the details they can get on what you're willing to buy -- and it's getting harder to tell where they're getting all that info. Retailers can now track what customers purchase to influence their buying patterns. Loyalty programs collect data on your purchases, frequency and preferences -- in exchange for deals.

Amazon's second Prime-related event for 2023 is officially called Prime Big Deal Days and will happen October 10 and 11. This is the second year in a row for a fall-based, site-wide Amazon sale and we're already seeing discounts pop up. You'll need a Prime membership to access many of the deals, though a few are available to everyone. This week, there are early Prime Day deals on the AirPods Pro, Amazon's Echo Dot, Amazon Music Unlimited, Eero 6 mesh Wi-Fi systems, Ring Video Doorbells and security systems and Amazon Fire Omni smart TVs. Here are the best early October Prime Day deals you can get right now.

Rap legend Tupac Shakur, soulful English singer-songwriter Amy Winehouse and many other beloved, 'once in a lifetime' talents have been tragically robbed of a full lifetime to share their gifts with the world. So we put the image-making artificial intelligence (AI) Midjourney to work to help imagine what these stars might have looked like at age 80. The results were unusual and uncanny, as might be expected of a machine manifesting snaps from an alternate dimension of what could have been. Scroll down to see if you recognize these famous figures in their AI-generated old age. The results might surprise you.

An advertiser reportedly used a deepfake of Tom Hanks to promote dental plans without the actor's permission. Hanks shared a warning on Instagram on Sunday alerting his followers about the AI-generated video, which he wrote he had "nothing to do with." Hanks has been outspoken about the challenges AI poses for the industry, and the use of actors' digital likenesses is one of the major points of concern voiced by striking SAG-AFTRA workers. Just last spring, Hanks said in an appearance on The Adam Buxton Podcast that AI and deepfakes present both artistic and legal challenges. "I could be hit by a bus tomorrow and that's it," Hanks said, "but my performances can go on and on and on and on and on, and outside of the understanding that it's been done with AI or deepfake, there'll be nothing to tell you that it's not me."

Tom Hanks and wife Rita Wilson walked the Pre-Grammy GALA red carpet discussing what they believe is the success to a great relationship, after being married for 34 years. Tom Hanks is warning fans about a potential AI-generated scam. The "Forrest Gump" actor says his name and likeness are being used without his consent in a dental promotion, and that users should "beware." I have nothing to do with it," he wrote, signing his name in a post on Instagram. Tom Hanks condemned a dental promotion using his name and likeness to promote their plan. A representative for Hanks did not immediately return Fox News Digital's request for comment. It is unclear where the image originated. Hanks recently gave his own two cents on artificial intelligence – noting that its use in the industry is nothing new but has "always been" lingering. "The first time we did a movie that had a huge amount of our own data locked in a computer, literally what we looked like, was a movie called'The Polar Express,'" Hanks said on "The Adam Buxton Podcast" about his 2004 animated film that used the technology. Film preparation for "The Polar Express" included motion capture. "And we saw this coming.