Retrieval-Augmented Generation (RAG), by incorporating external knowledge with parametric memory of language models, has become the state-of-the-art architecture for open-domain QA tasks. However, common knowledge bases are inherently constrained by limited coverage and noisy information, making retrieval-based approaches inadequate to answer implicit reasoning questions. In this paper, we propose an Induction-Augmented Generation (IAG) framework that utilizes inductive knowledge along with the retrieved documents for implicit reasoning. We leverage large language models (LLMs) for deriving such knowledge via a novel prompting method based on inductive reasoning patterns. On top of this, we implement two versions of IAG named IAG-GPT and IAG-Student, respectively. IAG-GPT directly utilizes the knowledge generated by GPT-3 for answer prediction, while IAG-Student gets rid of dependencies on GPT service at inference time by incorporating a student inductor model. The inductor is firstly trained via knowledge distillation and further optimized by back-propagating the generator feedback via differentiable beam scores. Experimental results show that IAG outperforms RAG baselines as well as ChatGPT on two Open-Domain QA tasks. Notably, our best models have won the first place in the official leaderboards of CSQA2.0 (since Nov 1, 2022) and StrategyQA (since Jan 8, 2023).

Are Large language models (LLMs) temporally grounded? Since LLMs cannot perceive and interact with the environment, it is impossible to answer this question directly. Instead, we provide LLMs with textual narratives and probe them with respect to their common-sense knowledge of the structure and duration of events, their ability to order events along a timeline, and self-consistency within their temporal model (e.g., temporal relations such as after and before are mutually exclusive for any pair of events). We evaluate state-of-the-art LLMs (such as LLaMA 2 and GPT-4) on three tasks reflecting these abilities. Generally, we find that LLMs lag significantly behind both human performance as well as small-scale, specialised LMs. In-context learning, instruction tuning, and chain-of-thought prompting reduce this gap only to a limited degree. Crucially, LLMs struggle the most with self-consistency, displaying incoherent behaviour in at least 27.23% of their predictions. Contrary to expectations, we also find that scaling the model size does not guarantee positive gains in performance. To explain these results, we study the sources from which LLMs may gather temporal information: we find that sentence ordering in unlabelled texts, available during pre-training, is only weakly correlated with event ordering. Moreover, public instruction tuning mixtures contain few temporal tasks. Hence, we conclude that current LLMs lack a consistent temporal model of textual narratives. Code, datasets, and LLM outputs are available at https://github.com/yfqiu-nlp/temporal-llms.

Information-seeking questions in long-form question answering (LFQA) often prove misleading due to ambiguity or false presupposition in the question. While many existing approaches handle misleading questions, they are tailored to limited questions, which are insufficient in a real-world setting with unpredictable input characteristics. In this work, we propose PreWoMe, a unified approach capable of handling any type of information-seeking question. The key idea of PreWoMe involves extracting presuppositions in the question and exploiting them as working memory to generate feedback and action about the question. Our experiment shows that PreWoMe is effective not only in tackling misleading questions but also in handling normal ones, thereby demonstrating the effectiveness of leveraging presuppositions, feedback, and action for real-world QA settings.

The United States will likely face a "great depression" if China seizes Taiwan's semiconductor industry, a hedge fund chief has said. Speaking at the Bloomberg New Economy Forum in Singapore last week, Citadel CEO Ken Griffin said the U.S. GDP would take a hit of between 5% to 10%, causing "an immediate Great Depression." "The United States has no ability to produce anywhere near the number of semiconductors it needs to run its economy," he was quoted as saying by Fortune. "If we lose access to Taiwanese semiconductors, the hit to U.S. GDP is probably in the order of magnitude of 5% to 10%. While it is unclear when, or if, that scenario will happen, Griffin said America's recent export controls, which include measures to cut China off from semiconductor chips and chip-making equipment, amounts to the U.S. "playing with fire." "You can argue that by depriving the Chinese of access to semiconductors, we up the risk that they seize Taiwan," Griffin added. In October, the Biden administration imposed sweeping export controls that banned U.S. companies from selling advanced semiconductors or equipment used to fabricate newer chips to China. Only companies that acquire a license from the Commerce Department will be allowed to export semiconductors and chip-making equipment to Chinese companies, according to The Wall Street Journal. In addition, the Biden administration also banned international companies from exporting chips that were built using U.S. technology. American citizens and green-card holders were also banned from working on certain technology for Chinese companies and entities. China has prioritized the development of semiconductor chips that are used in a variety of technology equipment, including artificial intelligence products. Commerce Secretary Gina Raimondo this month also said China will likely use advanced semiconductor technology "for surveillance." "China is crystal clear," she said, adding, "They will use this technology for surveillance.

Ken Burns has spent the last 40 years chronicling the most poignant and influential events in American history. The 66-year-old Oscar-nominated filmmaker has crafted definitive and multifaceted histories of the Civil War, baseball, the Roosevelts, cancer, country music and jazz. In an age of short Tweets and shorter attention spans, Burns's films are sprawling, deep-dive studies on topics that simultaneously reveal the best and worst of America. We are living through one of those moments right now as the coronavirus shakes every aspect of American life. With most of the country stuck at home and weathering a torrent of fear and breaking news, Burns is offering an alternative.

This will be the definitive forum on the shape of the next economy. Be part of the discussion and understand how the technological revolution will shape the future of work and business. "Could a machine do your job?" ask Michael Chui, James Manyika, and Mehdi Miremadi in a recent McKinsey Quarterly article, "Where Machines Could Replace Humans and Where They Can't Yet." "As automation technologies such as machine learning and robotics play an increasingly great role in everyday life, their potential effect on the workplace has, unsurprisingly, become a major focus of research and public concern. The discussion tends toward a Manichean guessing game: which jobs will or won't be replaced by machines? In fact, as our research has begun to show, the story is more nuanced. While automation will eliminate very few occupations entirely in the next decade, it will affect portions of almost all jobs to a greater or lesser degree, depending on the type of work they entail."

If we let machines put us out of work, it will be because of a failure of imagination and the will to make a better future. "Could a machine do your job?" ask Michael Chui, James Manyika, and Mehdi Miremadi in a recent McKinsey Quarterly article, "Where Machines Could Replace Humans and Where They Can't Yet." "As automation technologies such as machine learning and robotics play an increasingly great role in everyday life, their potential effect on the workplace has, unsurprisingly, become a major focus of research and public concern. The discussion tends toward a Manichean guessing game: which jobs will or won't be replaced by machines? In fact, as our research has begun to show, the story is more nuanced. While automation will eliminate very few occupations entirely in the next decade, it will affect portions of almost all jobs to a greater or lesser degree, depending on the type of work they entail."