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.

We propose an algorithm for learning a conditional generative model of a molecule given a target. Specifically, given a receptor molecule that one wishes to bind to, the conditional model generates candidate ligand molecules that may bind to it. The distribution should be invariant to rigid body transformations that act $\textit{jointly}$ on the ligand and the receptor; it should also be invariant to permutations of either the ligand or receptor atoms. Our learning algorithm is based on a continuous normalizing flow. We establish semi-equivariance conditions on the flow which guarantee the aforementioned invariance conditions on the conditional distribution. We propose a graph neural network architecture which implements this flow, and which is designed to learn effectively despite the vast differences in size between the ligand and receptor. We evaluate our method on the CrossDocked2020 dataset, attaining a significant improvement in binding affinity over competing methods.

A key challenge in science and engineering is to design experiments to learn about some unknown quantity of interest. Classical experimental design optimally allocates the experimental budget to maximize a notion of utility (e.g., reduction in uncertainty about the unknown quantity). We consider a rich setting, where the experiments are associated with states in a {\em Markov chain}, and we can only choose them by selecting a {\em policy} controlling the state transitions. This problem captures important applications, from exploration in reinforcement learning to spatial monitoring tasks. We propose an algorithm -- \textsc{markov-design} -- that efficiently selects policies whose measurement allocation \emph{provably converges to the optimal one}. The algorithm is sequential in nature, adapting its choice of policies (experiments) informed by past measurements. In addition to our theoretical analysis, we showcase our framework on applications in ecological surveillance and pharmacology.

Providing better language tools for low-resource and endangered languages is imperative for equitable growth. Recent progress with massively multilingual pretrained models has proven surprisingly effective at performing zero-shot transfer to a wide variety of languages. However, this transfer is not universal, with many languages not currently understood by multilingual approaches. It is estimated that only 72 languages possess a "small set of labeled datasets" on which we could test a model's performance, the vast majority of languages not having the resources available to simply evaluate performances on. In this work, we attempt to clarify which languages do and do not currently benefit from such transfer. To that end, we develop a general approach that requires only unlabelled text to detect which languages are not well understood by a cross-lingual model. Our approach is derived from the hypothesis that if a model's understanding is insensitive to perturbations to text in a language, it is likely to have a limited understanding of that language. We construct a cross-lingual sentence similarity task to evaluate our approach empirically on 350, primarily low-resource, languages.

Misinformation spread over social media has become an undeniable infodemic. However, not all spreading claims are made equal. If propagated, some claims can be destructive, not only on the individual level, but to organizations and even countries. Detecting claims that should be prioritized for fact-checking is considered the first step to fight against spread of fake news. With training data limited to a handful of languages, developing supervised models to tackle the problem over lower-resource languages is currently infeasible. Therefore, our work aims to investigate whether we can use existing datasets to train models for predicting worthiness of verification of claims in tweets in other languages. We present a systematic comparative study of six approaches for cross-lingual check-worthiness estimation across pairs of five diverse languages with the help of Multilingual BERT (mBERT) model. We run our experiments using a state-of-the-art multilingual Twitter dataset. Our results show that for some language pairs, zero-shot cross-lingual transfer is possible and can perform as good as monolingual models that are trained on the target language. We also show that in some languages, this approach outperforms (or at least is comparable to) state-of-the-art models.

Our team, Hibikino-Musashi@Home (the shortened name is HMA), was founded in 2010. It is based in the Kitakyushu Science and Research Park, Japan. We have participated in the RoboCup@Home Japan open competition open platform league every year since 2010. Moreover, we participated in the RoboCup 2017 Nagoya as open platform league and domestic standard platform league teams. Currently, the Hibikino-Musashi@Home team has 20 members from seven different laboratories based in the Kyushu Institute of Technology. In this paper, we introduce the activities of our team and the technologies.

Streets networks provide an invaluable source of information about the different temporal and spatial patterns emerging in our cities. These streets are often represented as graphs where intersections are modelled as nodes and streets as links between them. Previous work has shown that raster representations of the original data can be created through a learning algorithm on low-dimensional representations of the street networks. In contrast, models that capture high-level urban network metrics can be trained through convolutional neural networks. However, the detailed topological data is lost through the rasterisation of the street network. The models cannot recover this information from the image alone, failing to capture complex street network features. This paper proposes a model capable of inferring good representations directly from the street network. Specifically, we use a variational autoencoder with graph convolutional layers and a decoder that outputs a probabilistic fully-connected graph to learn latent representations that encode both local network structure and the spatial distribution of nodes. We train the model on thousands of street network segments and use the learnt representations to generate synthetic street configurations. Finally, we proposed a possible application to classify the urban morphology of different network segments by investigating their common characteristics in the learnt space.

Google last week unpacked a host of critical new artificial intelligence (AI) advancements, particularly in the areas around generative AI such as language analysis, text-to-video, and assisted writing capabilities. Along with other enticing developments around ethical AI, Google revealed it would be working on AI in at least one thousand languages going forward. At the Google AI event held at the company's Pier 57 offices in New York City on November 2, the global technology titan highlighted areas of focus to include democratizing AI development pathways, "building for everyone" responsible, controlled models which can identify generative AI -- unsupervised artificial intelligence learning algorithms that can create new digital audio, text, imagery, video, or code -- along with advancing language translation, disaster management, and health AI. Google shared the first rendering of a video that shares both of the company's complementary text-to-video research approaches -- Imagen Video and Phenaki. The interest of global communities was piqued by the announcement that language abilities are being developed using AI for the world's one thousand most spoken languages, showcasing another major area of artificial intelligence focus as tech giants compete to dominate the internet's next battleground.

Burkina Faso is a multi-cultural and diverse country with a rich history. In this article, we explore how personal names can be interpreted to reflect regional, ethnic appartenance within the country. Then we illustrate how the use of a personal name can affect a black-box Artificial Intelligence – such as OpenAI's DALL-E. This is a first article in our series of blog posts with tag #thisnamedpersondoesnotexist.

Blockchain and AI, particularly machine learning, are two quite recent revolutionary technologies that are being adopted by Governments and Businesses in all sorts of ways. In a series of articles (in 4 parts) I reflect in what ways these technologies have the potential to impact the lives of ordinary citizens. A lot of discussion has been going on about how blockchain and AI affects governments and businesses. If AI has been around for quite a long time, the more recent Blockchain technology, has taken the world by storm, as being a database system that provides us with a simple protocol that allows transactions to be simultaneously anonymous and secure, peer-to-peer, immediate and in constant flow. The beautiful promise of blockchain is that is distributes the trust which is currently allocated to centralized, large and powerful intermediaries, to a large global network of people engaged in massive collaboration, facilitated by clever coding and cryptography.