Ant Group Co., the fintech affiliate of Alibaba Group Holding Ltd., received approval from the Chinese government to roll out products powered by its large language model Bailing to the public. Bailing will be applied to Ant's various services and help with innovation, Xu Peng, vice president of Ant Group said in a statement on Monday. Chinese tech firms from Alibaba to Tencent Holdings Ltd. and Baidu Inc. have joined startups Baichuan and Zhipu to release ChatGPT-like products, joining a global race to capitalize on the potential of generative AI. Ant, the owner of Alipay, can leverage the popularity of the mobile payment service to gain more data and insight on user habits. Hangzhou-based Alibaba has defined AI as one of its two core strategies as it goes through a six-way spinoff.

In a sign of great interest, OpenAI's developer conference took up multiple floors of a large event space in downtown San Francisco, spilling onto the roof where a breakfast spread of burritos and avocado toast, complete with a separate table featuring multiple hot sauces, awaited the hundreds of attendees lining up outside to get in. Those who weren't able to score a coveted invitation to the event gathered at so-called "watch parties" at separate corporate offices to keep tabs on the live stream.

Users will soon be able to make customized versions of ChatGPT, the maker of the tool OpenAI said Monday as it made a series of announcements at its first Developer Day conference in San Francisco. OpenAI is calling the customizable versions of ChatGPT "GPTs," which it says will be able to comply with specified instructions and have access to user-provided information. "The upsides of this are going to be tremendous," OpenAI CEO Sam Altman said on stage on Monday. "It gives agency to everyone." ChatGPT currently has 100 million weekly active users, Altman added.

OpenAI's ChatGPT became a phenomenon thanks to its wide-ranging abilities, such as drafting college essays, writing working computer programs, and digging up information from across the web. Now the company aims to further widen the range of tricks up ChatGPT's sleeve by making it possible for anyone to build a custom chatbot powered by the technology--without any coding skills. OpenAI suggests people might want to build custom bots to help with specific problems or interests in their life, such as helping someone learn the rules of a board game, teach their kids math, or help design stickers using AI-generated art. To create one of these custom bots or AI agents, which OpenAI calls "GPTs," a user need only specify, by talking with ChatGPT, what they would like the bot to do. Behind the scenes, ChatGPT will write the code needed to create and run the new bot. The bots can plug into other sites and services to do things like access databases, search emails, and automate ecommerce orders, OpenAI says.

Last week, Elon Musk flew to the UK to hype up the existential risk posed by artificial intelligence. A couple of days later, he announced that his latest company, xAI, had developed a powerful AI--one with fewer guardrails than the competition. The AI model, called Grok (a name that means "to understand" in tech circles), "is designed to answer questions with a bit of wit and has a rebellious streak, so please don't use it if you hate humor!" reads an announcement on the company's website. "It will also answer spicy questions that are rejected by most other AI systems." The announcement does not explain what a "spicy" or "rebellious" means, but most commercial AI models will refuse to generate sexually explicit, violent, or illegal content, and they are designed to avoid expressing biases picked up from their training data.

People primarily consult tables to conduct data analysis or answer specific questions. Text generation systems that can provide accurate table summaries tailored to users' information needs can facilitate more efficient access to relevant data insights. Motivated by this, we define a new query-focused table summarization task, where text generation models have to perform human-like reasoning and analysis over the given table to generate a tailored summary. We introduce a new benchmark named QTSumm for this task, which contains 7,111 human-annotated query-summary pairs over 2,934 tables covering diverse topics. We investigate a set of strong baselines on QTSumm, including text generation, table-to-text generation, and large language models. Experimental results and manual analysis reveal that the new task presents significant challenges in table-to-text generation for future research. Moreover, we propose a new approach named ReFactor, to retrieve and reason over query-relevant information from tabular data to generate several natural language facts. Experimental results demonstrate that ReFactor can bring improvements to baselines by concatenating the generated facts to the model input. Our data and code are publicly available at https://github.com/yale-nlp/QTSumm.

App reviews from app stores are crucial for improving software requirements. A large number of valuable reviews are continually being posted, describing software problems and expected features. Effectively utilizing user reviews necessitates the extraction of relevant information, as well as their subsequent summarization. Due to the substantial volume of user reviews, manual analysis is arduous. Various approaches based on natural language processing (NLP) have been proposed for automatic user review mining. However, the majority of them requires a manually crafted dataset to train their models, which limits their usage in real-world scenarios. In this work, we propose Mini-BAR, a tool that integrates large language models (LLMs) to perform zero-shot mining of user reviews in both English and French. Specifically, Mini-BAR is designed to (i) classify the user reviews, (ii) cluster similar reviews together, (iii) generate an abstractive summary for each cluster and (iv) rank the user review clusters. To evaluate the performance of Mini-BAR, we created a dataset containing 6,000 English and 6,000 French annotated user reviews and conducted extensive experiments. Preliminary results demonstrate the effectiveness and efficiency of Mini-BAR in requirement engineering by analyzing bilingual app reviews. (Replication package containing the code, dataset, and experiment setups on https://github.com/Jl-wei/mini-bar )

Generalization remains one of the most important desiderata for robust robot learning systems. While recently proposed approaches show promise in generalization to novel objects, semantic concepts, or visual distribution shifts, generalization to new tasks remains challenging. For example, a language-conditioned policy trained on pick-andplace tasks will not be able to generalize to a folding task, even if the arm trajectory of folding is similar to pick-and-place. Our key insight is that this kind of generalization becomes feasible if we represent the task through rough trajectory sketches. We propose a policy conditioning method using such rough trajectory sketches, which we call RT-Trajectory, that is practical, easy to specify, and allows the policy to effectively perform new tasks that would otherwise be challenging to perform. We find that trajectory sketches strike a balance between being detailed enough to express low-level motioncentric guidance while being coarse enough to allow the learned policy to interpret the trajectory sketch in the context of situational visual observations. In addition, we show how trajectory sketches can provide a useful interface to communicate with robotic policies - they can be specified through simple human inputs like drawings or videos, or through automated methods such as modern image-generating or waypoint-generating methods. We evaluate RT-Trajectory at scale on a variety of real-world robotic tasks, and find that RT-Trajectory is able to perform a wider range of tasks compared to languageconditioned and goal-conditioned policies, when provided the same training data. Evaluation videos can be found at https://rt-trajectory.github.io/. The pursuit of generalist robot policies has been a perennial challenge in robotics. The goal is to devise policies that not only perform well on known tasks but can also generalize to novel objects, scenes, and motions that are not represented in the training dataset.

Despite the recent success of large, pretrained neural language models (LLMs) on a variety of prompting tasks, these models can be alarmingly brittle to small changes in inputs or application contexts. To better understand such behavior and motivate the design of more robust LLMs, we provide a causal formulation of linguistic competence in the context of LLMs and propose a general framework to study and measure LLM competence. Our framework, CALM (Competence-based Analysis of Language Models), establishes the first quantitative measure of LLM competence, which we study by damaging models' internal representations of various linguistic properties in the course of performing various tasks using causal probing and evaluating models' alignment under these interventions with a given causal model. We also develop a novel approach for performing causal probing interventions using gradient-based adversarial attacks, which can target a broader range of properties and representations than existing techniques. We carry out a case study of CALM using these interventions to analyze BERT and RoBERTa's competence across a variety of lexical inference tasks, showing that the CALM framework and competence metric can be valuable tools for explaining and predicting their behavior across these tasks.

Knowledge Base Question Answering (KBQA) aims to answer factoid questions based on knowledge bases. However, generating the most appropriate knowledge base query code based on Natural Language Questions (NLQ) poses a significant challenge in KBQA. In this work, we focus on the CCKS2023 Competition of Question Answering with Knowledge Graph Inference for Unmanned Systems. Inspired by the recent success of large language models (LLMs) like ChatGPT and GPT-3 in many QA tasks, we propose a ChatGPT-based Cypher Query Language (CQL) generation framework to generate the most appropriate CQL based on the given NLQ. Our generative framework contains six parts: an auxiliary model predicting the syntax-related information of CQL based on the given NLQ, a proper noun matcher extracting proper nouns from the given NLQ, a demonstration example selector retrieving similar examples of the input sample, a prompt constructor designing the input template of ChatGPT, a ChatGPT-based generation model generating the CQL, and an ensemble model to obtain the final answers from diversified outputs. With our ChatGPT-based CQL generation framework, we achieved the second place in the CCKS 2023 Question Answering with Knowledge Graph Inference for Unmanned Systems competition, achieving an F1-score of 0.92676.