This paper demonstrates how OpenAI's ChatGPT can be used in a few-shot setting to convert natural language instructions into a sequence of executable robot actions. The paper proposes easy-to-customize input prompts for ChatGPT that meet common requirements in practical applications, such as easy integration with robot execution systems and applicability to various environments while minimizing the impact of ChatGPT's token limit. The prompts encourage ChatGPT to output a sequence of predefined robot actions, represent the operating environment in a formalized style, and infer the updated state of the operating environment. Experiments confirmed that the proposed prompts enable ChatGPT to act according to requirements in various environments, and users can adjust ChatGPT's output with natural language feedback for safe and robust operation. The proposed prompts and source code are open-source and publicly available at https://github.com/microsoft/ChatGPT-Robot-Manipulation-Prompts

Cheap-to-Build Very Large-Language Models (CtB-LLMs) with affordable training are emerging as the next big revolution in natural language processing and understanding. These CtB-LLMs are democratizing access to trainable Very Large-Language Models (VLLMs) and, thus, may represent the building blocks of many NLP systems solving downstream tasks. Hence, a little or a large bias in CtB-LLMs may cause huge harm. In this paper, we performed a large investigation of the bias of three families of CtB-LLMs, and we showed that debiasing techniques are effective and usable. Indeed, according to current tests, the LLaMA and the OPT families have an important bias in gender, race, religion, and profession. In contrast to the analysis for other LLMs, we discovered that bias depends not on the number of parameters but on the perplexity. Finally, the debiasing of OPT using LoRA reduces bias up to 4.12 points in the normalized stereotype score.

Large Language Models (LLM) are a new class of computation engines, "programmed" via prompt engineering. We are still learning how to best "program" these LLMs to help developers. We start with the intuition that developers tend to consciously and unconsciously have a collection of semantics facts in mind when working on coding tasks. Mostly these are shallow, simple facts arising from a quick read. For a function, examples of facts might include parameter and local variable names, return expressions, simple pre- and post-conditions, and basic control and data flow, etc. One might assume that the powerful multi-layer architecture of transformer-style LLMs makes them inherently capable of doing this simple level of "code analysis" and extracting such information, implicitly, while processing code: but are they, really? If they aren't, could explicitly adding this information help? Our goal here is to investigate this question, using the code summarization task and evaluate whether automatically augmenting an LLM's prompt with semantic facts explicitly, actually helps. Prior work shows that LLM performance on code summarization benefits from few-shot samples drawn either from the same-project or from examples found via information retrieval methods (such as BM25). While summarization performance has steadily increased since the early days, there is still room for improvement: LLM performance on code summarization still lags its performance on natural-language tasks like translation and text summarization. We find that adding semantic facts actually does help! This approach improves performance in several different settings suggested by prior work, including for two different Large Language Models. In most cases, improvement nears or exceeds 2 BLEU; for the PHP language in the challenging CodeSearchNet dataset, this augmentation actually yields performance surpassing 30 BLEU.

ChatGPT is a recent chatbot service released by OpenAI and is receiving increasing attention over the past few months. While evaluations of various aspects of ChatGPT have been done, its robustness, i.e., the performance to unexpected inputs, is still unclear to the public. Robustness is of particular concern in responsible AI, especially for safety-critical applications. In this paper, we conduct a thorough evaluation of the robustness of ChatGPT from the adversarial and out-of-distribution (OOD) perspective. To do so, we employ the AdvGLUE and ANLI benchmarks to assess adversarial robustness and the Flipkart review and DDXPlus medical diagnosis datasets for OOD evaluation. We select several popular foundation models as baselines. Results show that ChatGPT shows consistent advantages on most adversarial and OOD classification and translation tasks. However, the absolute performance is far from perfection, which suggests that adversarial and OOD robustness remains a significant threat to foundation models. Moreover, ChatGPT shows astounding performance in understanding dialogue-related texts and we find that it tends to provide informal suggestions for medical tasks instead of definitive answers. Finally, we present in-depth discussions of possible research directions.

Americans have grown more worried about AI in the last nine months. A new survey from the Pew Research Center indicates 52 percent of respondents are more concerned than excited about rising artificial intelligence use, up 14 points since December. Meanwhile, only 10 percent say they're more excited than worried, while another 36 percent described their views as equally balanced. "Concern about AI outweighs excitement across all major demographic groups," the Pew Research Center wrote in a blog post today. It's been an eventful nine months since the Pew Center last surveyed people about AI.

In March, OpenAI sought to head off concerns that its immensely popular, albeit hallucination-prone, ChatGPT generative AI could be used to dangerously amplify political disinformation campaigns through an update to the company's Usage Policy to expressly prohibit such behavior. However, an investigation by The Washington Post shows that the chatbot is still easily incited to breaking those rules, with potentially grave repercussions for the 2024 election cycle. OpenAI's user policies specifically ban its use for political campaigning, save for use by "grassroots advocacy campaigns" organizations. This includes generating campaign materials in high volumes, targeting those materials at specific demographics, building campaign chatbots to disseminate information, engage in political advocacy or lobbying. Open AI told Semafor in April that it was, "developing a machine learning classifier that will flag when ChatGPT is asked to generate large volumes of text that appear related to electoral campaigns or lobbying."

OpenAI launched ChatGPT Enterprise today, the business-focused subscription it teased in April. The company says it won't train its AI models on any business data or conversations under the new plan. "Our models don't learn from your usage," the company wrote in an announcement blog post about the enterprise features. In addition, the new plan encrypts business chats (in transit and at rest) and is SOC 2 compliant. OpenAI says companies including Block, Canva, Carlyle, The Estée Lauder Companies, PwC and Zapier have already tested ChatGPT Enterprise.

This copy is for your personal, non-commercial use only. For non-personal use or to order multiple copies, please contact Dow Jones Reprints at 1-800-843-0008 or visit www.djreprints.com.

What is it with wave metaphors? Technological determinists – people who believe that technology drives history – love them. Think of Alvin Toffler, who saw the history of civilisation as a succession of three such waves (agricultural, industrial and post-industrial). The idea is of immense power, unstoppable, moving inexorably towards us as we cower before its immensity, much as the dinosaurs must have done when they saw the mile-high tsunami heading in their direction. Mustafa Suleyman says he is not a determinist, but at times he sounds awfully like one.

Recently, there has been a growing interest in the field of controllable Text-to-Speech (TTS). While previous studies have relied on users providing specific style factor values based on acoustic knowledge or selecting reference speeches that meet certain requirements, generating speech solely from natural text prompts has emerged as a new challenge for researchers. This challenge arises due to the scarcity of high-quality speech datasets with natural text style prompt and the absence of advanced text-controllable TTS models. In light of this, 1) we propose TextrolSpeech, which is the first large-scale speech emotion dataset annotated with rich text attributes. The dataset comprises 236,220 pairs of style prompt in natural text descriptions with five style factors and corresponding speech samples. Through iterative experimentation, we introduce a multi-stage prompt programming approach that effectively utilizes the GPT model for generating natural style descriptions in large volumes. 2) Furthermore, to address the need for generating audio with greater style diversity, we propose an efficient architecture called Salle. This architecture treats text controllable TTS as a language model task, utilizing audio codec codes as an intermediate representation to replace the conventional mel-spectrogram. Finally, we successfully demonstrate the ability of the proposed model by showing a comparable performance in the controllable TTS task. Audio samples are available at https://sall-e.github.io/