Large language models have shown impressive few-shot results on a wide range of tasks. However, when knowledge is key for such results, as is the case for tasks such as question answering and fact checking, massive parameter counts to store knowledge seem to be needed. Retrieval augmented models are known to excel at knowledge intensive tasks without the need for as many parameters, but it is unclear whether they work in few-shot settings. In this work we present Atlas, a carefully designed and pre-trained retrieval augmented language model able to learn knowledge intensive tasks with very few training examples. We perform evaluations on a wide range of tasks, including MMLU, KILT and NaturalQuestions, and study the impact of the content of the document index, showing that it can easily be updated. Notably, Atlas reaches over 42% accuracy on Natural Questions using only 64 examples, outperforming a 540B parameters model by 3% despite having 50x fewer parameters.

Large Language Models (LLMs) have been transformative. They are pre-trained foundational models that are self-supervised and can be adapted with fine tuning to a wide range of natural language tasks, each of which previously would have required a separate network model. This is one step closer to the extraordinary versatility of human language. GPT-3 and more recently LaMDA can carry on dialogs with humans on many topics after minimal priming with a few examples. However, there has been a wide range of reactions and debate on whether these LLMs understand what they are saying or exhibit signs of intelligence. This high variance is exhibited in three interviews with LLMs reaching wildly different conclusions. A new possibility was uncovered that could explain this divergence. What appears to be intelligence in LLMs may in fact be a mirror that reflects the intelligence of the interviewer, a remarkable twist that could be considered a Reverse Turing Test. If so, then by studying interviews we may be learning more about the intelligence and beliefs of the interviewer than the intelligence of the LLMs. As LLMs become more capable they may transform the way we interact with machines and how they interact with each other. Increasingly, LLMs are being coupled with sensorimotor devices. LLMs can talk the talk, but can they walk the walk? A road map for achieving artificial general autonomy is outlined with seven major improvements inspired by brain systems. LLMs could be used to uncover new insights into brain function by downloading brain data during natural behaviors.

In open-ended natural-language generation, existing text decoding methods typically struggle to produce text which is both diverse and high-quality. Greedy and beam search are known to suffer from text degeneration and linguistic diversity issues, while temperature, top-k, and nucleus sampling often yield diverse but low-quality outputs. In this work, we present crowd sampling, a family of decoding methods based on Bayesian risk minimization, to address this diversity-quality trade-off. Inspired by the principle of "the wisdom of the crowd," crowd sampling seeks to select a candidate from a pool of candidates that has the least expected risk (i.e., highest expected reward) under a generative model according to a given utility function. Crowd sampling can be seen as a generalization of numerous existing methods, including majority voting, and in practice, it can be used as a drop-in replacement for existing sampling methods. Extensive experiments show that crowd sampling delivers improvements of 3-7 ROUGE and BLEU points across a wide range of tasks, including summarization, data-to-text, translation, and textual style transfer, while achieving new state-of-the-art results on WebNLG and WMT'16.

A recurring challenge of crowdsourcing NLP datasets at scale is that human writers often rely on repetitive patterns when crafting examples, leading to a lack of linguistic diversity. We introduce a novel approach for dataset creation based on worker and AI collaboration, which brings together the generative strength of language models and the evaluative strength of humans. Starting with an existing dataset, MultiNLI for natural language inference (NLI), our approach uses dataset cartography to automatically identify examples that demonstrate challenging reasoning patterns, and instructs GPT-3 to compose new examples with similar patterns. Machine generated examples are then automatically filtered, and finally revised and labeled by human crowdworkers. The resulting dataset, WANLI, consists of 107,885 NLI examples and presents unique empirical strengths over existing NLI datasets. Remarkably, training a model on WANLI improves performance on eight out-of-domain test sets we consider, including by 11% on HANS and 9% on Adversarial NLI, compared to training on the 4x larger MultiNLI. Moreover, it continues to be more effective than MultiNLI augmented with other NLI datasets. Our results demonstrate the promise of leveraging natural language generation techniques and re-imagining the role of humans in the dataset creation process.

Ukrainian President Volodymyr Zelenskyy has backed a fundraising campaign to help Ukraine build a naval drone fleet to protect cities against Russian missiles launched from warships on the Black Sea. United24, an initiative Zelenskyy launched to raise charitable donations following Russia's invasion in February, said Ukraine needed 100 drones operating from the sea, each of which costs 10 million hryvnias (around $274,000). The fundraising site said that since the invasion began, Russian has launched over 4,500 missiles into Ukraine and "every fifth strike came from the sea". "We must defend the waters of our seas and peaceful cities from Russian missiles launched from ships," Zelenskyy wrote on the Telegram messaging app on Friday. "Naval drones will also help unblock the corridor for civilian ships transporting grain for the world," he said.

Large language models (LLMs) have led to a series of breakthroughs in natural language processing (NLP), owing to their excellent understanding and generation abilities. Remarkably, what further sets these models apart is the massive amounts of world knowledge they internalize during pretraining. While many downstream applications provide the model with an informational context to aid its performance on the underlying task, how the model's world knowledge interacts with the factual information presented in the context remains under explored. As a desirable behavior, an LLM should give precedence to the context whenever it contains task-relevant information that conflicts with the model's memorized knowledge. This enables model predictions to be grounded in the context, which can then be used to update or correct specific model predictions without frequent retraining. By contrast, when the context is irrelevant to the task, the model should ignore it and fall back on its internal knowledge. In this paper, we undertake a first joint study of the aforementioned two properties, namely controllability and robustness, in the context of LLMs. We demonstrate that state-of-the-art T5 and PaLM (both pretrained and finetuned) could exhibit poor controllability and robustness, which do not scale with increasing model size. As a solution, we propose a novel method - Knowledge Aware FineTuning (KAFT) - to strengthen both controllability and robustness by incorporating counterfactual and irrelevant contexts to standard supervised datasets. Our comprehensive evaluation showcases the utility of KAFT across model architectures and sizes.

Having AIs Train Robot Dogs to Balance Makes Them a Lot Cheaper Jeremy Tsu New Scientist "An AI has been used to train a small robot dog to perform cleaning tasks. The hardware cost a total of $6300, which is less than a tenth of the price tag of the well-known robot dogs built by US tech firm Boston Dynamics. This type of self-taught robotic body coordination relies on an AI training regimen that could pave the way for affordable robot dogs and possibly even humanoid robots that could be used as helpers in homes and workplaces." Google Plans Giant AI Language Model Supporting World's 1,000 Most Spoken Languages James Vincent The Verge "i'The way we get to 1,000 languages is not by building 1,000 different models. Languages are like organisms, they've evolved from one another and they have certain similarities. And we can find some pretty spectacular advances in what we call zero-shot learning when we incorporate data from a new language into our 1,000 language model and get the ability to translate [what it's learned] from a high-resource language to a low-resource language,' says [Zoubin Ghahramani, vice president of research at Google AI]. Genetically Modified Mosquitoes Cut the Insect's Number by 96 Percent Miriam Fauzia New Scientist "Although not a permanent fix, periodically releasing such mosquitoes could reduce the burden of infections including dengue, malaria, and Zika.

In the 36th week of war in Ukraine, Russia backed out of a United Nations-sponsored agreement guaranteeing the safe passage of grain ships through the Black Sea, only to rejoin it three days later. Moscow's withdrawal over the weekend renewed fears of a global food crisis – concerns that have not been completely quelled since it rejoined because its return came with conditions. President Vladimir Putin said he reserved the right to back out again if Kyiv used the humanitarian corridor for attacks, the reason Russia gave for the initial pullout. The Kremlin has also warned that it has not yet decided whether to extend the grain deal, which expires in two weeks. Officials in Moscow had said that grain ships may have acted as a cloak for an attack on its naval base on Saturday at Sevastopol on the Crimean Peninsula.

In this paper, we present Pangu-Weather, a deep learning based system for fast and accurate global weather forecast. For this purpose, we establish a data-driven environment by downloading $43$ years of hourly global weather data from the 5th generation of ECMWF reanalysis (ERA5) data and train a few deep neural networks with about $256$ million parameters in total. The spatial resolution of forecast is $0.25^\circ\times0.25^\circ$, comparable to the ECMWF Integrated Forecast Systems (IFS). More importantly, for the first time, an AI-based method outperforms state-of-the-art numerical weather prediction (NWP) methods in terms of accuracy (latitude-weighted RMSE and ACC) of all factors (e.g., geopotential, specific humidity, wind speed, temperature, etc.) and in all time ranges (from one hour to one week). There are two key strategies to improve the prediction accuracy: (i) designing a 3D Earth Specific Transformer (3DEST) architecture that formulates the height (pressure level) information into cubic data, and (ii) applying a hierarchical temporal aggregation algorithm to alleviate cumulative forecast errors. In deterministic forecast, Pangu-Weather shows great advantages for short to medium-range forecast (i.e., forecast time ranges from one hour to one week). Pangu-Weather supports a wide range of downstream forecast scenarios, including extreme weather forecast (e.g., tropical cyclone tracking) and large-member ensemble forecast in real-time. Pangu-Weather not only ends the debate on whether AI-based methods can surpass conventional NWP methods, but also reveals novel directions for improving deep learning weather forecast systems.

Network intrusion detection systems are evolving into intelligent systems that perform data analysis while searching for anomalies in their environment. Indeed, the development of deep learning techniques paved the way to build more complex and effective threat detection models. However, training those models may be computationally infeasible in most Edge or IoT devices. Current approaches rely on powerful centralized servers that receive data from all their parties - violating basic privacy constraints and substantially affecting response times and operational costs due to the huge communication overheads. To mitigate these issues, Federated Learning emerged as a promising approach, where different agents collaboratively train a shared model, without exposing training data to others or requiring a compute-intensive centralized infrastructure. This work presents GowFed, a novel network threat detection system that combines the usage of Gower Dissimilarity matrices and Federated averaging. Different approaches of GowFed have been developed based on state-of the-art knowledge: (1) a vanilla version; and (2) a version instrumented with an attention mechanism. Furthermore, each variant has been tested using simulation oriented tools provided by TensorFlow Federated framework. In the same way, a centralized analogous development of the Federated systems is carried out to explore their differences in terms of scalability and performance - across a set of designed experiments/scenarios. Overall, GowFed intends to be the first stepping stone towards the combined usage of Federated Learning and Gower Dissimilarity matrices to detect network threats in industrial-level networks.