Vector databases typically manage large collections of embedding vectors. As AI applications are growing rapidly, the number of embeddings that need to be stored and indexed is increasing. The Faiss library is dedicated to vector similarity search, a core functionality of vector databases. Faiss is a toolkit of indexing methods and related primitives used to search, cluster, compress and transform vectors. This paper describes the trade-offs in vector search and the design principles of Faiss in terms of structure, approach to optimization and interfacing. We benchmark key features of the library and discuss a few selected use cases to highlight its broad applicability.

Retrieval-augmented language models (RALMs) improve performance by accessing long-tail and up-to-date knowledge from external data stores, but are challenging to build. Existing approaches require either expensive retrieval-specific modifications to LM pre-training or use post-hoc integration of the data store that leads to suboptimal performance. We introduce Retrieval-Augmented Dual Instruction Tuning (RA-DIT), a lightweight fine-tuning methodology that provides a third option by retrofitting any LLM with retrieval capabilities. Our approach operates in two distinct fine-tuning steps: (1) one updates a pre-trained LM to better use retrieved information, while (2) the other updates the retriever to return more relevant results, as preferred by the LM. By fine-tuning over tasks that require both knowledge utilization and contextual awareness, we demonstrate that each stage yields significant performance improvements, and using both leads to additional gains. Our best model, RA-DIT 65B, achieves state-of-the-art performance across a range of knowledge-intensive zero- and few-shot learning benchmarks, significantly outperforming existing in-context RALM approaches by up to +8.9% in 0-shot setting and +1.4% in 5-shot setting on average.

Language models are currently trained to predict tokens given document prefixes, enabling them to zero shot long form generation and prompting-style tasks which can be reduced to document completion. We instead present IN-CONTEXT PRETRAINING, a new approach where language models are trained on a sequence of related documents, thereby explicitly encouraging them to read and reason across document boundaries. Our approach builds on the fact that current pipelines train by concatenating random sets of shorter documents to create longer context windows; this improves efficiency even though the prior documents provide no signal for predicting the next document. Given this fact, we can do IN-CONTEXT PRETRAINING by simply changing the document ordering so that each context contains related documents, and directly applying existing pretraining pipelines. However, this document sorting problem is challenging. There are billions of documents and we would like the sort to maximize contextual similarity for every document without repeating any data. To do this, we introduce approximate algorithms for finding related documents with efficient nearest neighbor search and constructing coherent batches with a graph cover algorithm. Our experiments show IN-CONTEXT PRETRAINING offers a scalable and simple approach to significantly enhance LM performance: we see notable improvements in tasks that require more complex contextual reasoning, including in-context learning (+8%), reading comprehension (+15%), faithfulness to previous contexts (+16%), long-context reasoning (+5%), and retrieval augmentation (+9%).

Language models (LMs) exhibit remarkable abilities to solve new tasks from just a few examples or textual instructions, especially at scale. They also, paradoxically, struggle with basic functionality, such as arithmetic or factual lookup, where much simpler and smaller specialized models excel. In this paper, we show that LMs can teach themselves to *use external tools* via simple APIs and achieve the best of both worlds. We introduce *Toolformer*, a model trained to decide which APIs to call, when to call them, what arguments to pass, and how to best incorporate the results into future token prediction. This is done in a self-supervised way, requiring nothing more than a handful of demonstrations for each API. We incorporate a range of tools, including a calculator, a Q&A system, a search engine, a translation system, and a calendar. Toolformer achieves substantially improved zero-shot performance across a variety of downstream tasks, often competitive with much larger models, without sacrificing its core language modeling abilities.

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 Natural Questions, 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 parameter model by 3% despite having 50x fewer parameters.

A cyber–cyber digital twin is a simulation of a software system. By contrast, a cyber–physical digital twin is a simulation of a non-software (physical) system. Although cyber–physical digital twins have received a lot of recent attention, their cyber–cyber counterparts have been comparatively overlooked. In this paper we show how the unique properties of cyber–cyber digital twins open up exciting opportunities for research and development. Like all digital twins, the cyber–cyber digital twin is both informed by and informs the behaviour of the twin it simulates. It is therefore a software system that simulates another software system, making it conceptually truly a twin, blurring the distinction between the simulated and the simulator. Cyber–cyber digital twins can be twins of other cyber–cyber digital twins, leading to a hierarchy of twins. As we shall see, these apparently philosophical observations have practical ramifications for the design, implementation and deployment of digital twins at Facebook.

A cyber cyber digital twin is a deployed software model that executes in tandem with the system it simulates, contributing to, and drawing from, the system's behaviour. This paper outlines Facebook's cyber cyber digital twin, dubbed WW, a twin of Facebook's WWW platform, built using web-enabled simulation. The paper focuses on the current research challenges and opportunities in the area of measurement. Measurement challenges lie at the heart of modern simulation. They directly impact how we use simulation outcomes for automated online and semi-automated offline decision making. Measurements also encompas how we verify and validate those outcomes. Modern simulation systems are increasingly becoming more like cyber cyber digital twins, effectively moving from manual to automated decision making, hence, these measurement challenges acquire ever greater significance.

Latent variable models allow capturing the hidden structure underlying the data. In particular, feature allocation models represent each observation by a linear combination of latent variables. These models are often used to make predictions either for new observations or for missing information in the original data, as well as to perform exploratory data analysis. Although there is an extensive literature on latent feature allocation models for homogeneous datasets, where all the attributes that describe each object are of the same (continuous or discrete) type, there is no general framework for practical latent feature modeling for heterogeneous datasets. In this paper, we introduce a general Bayesian nonparametric latent feature allocation model suitable for heterogeneous datasets, where the attributes describing each object can be arbitrary combinations of real-valued, positive real-valued, categorical, ordinal and count variables. The proposed model presents several important properties. First, it is suitable for heterogeneous data while keeping the properties of conjugate models, which enables us to develop an inference algorithm that presents linear complexity with respect to the number of objects and attributes per MCMC iteration. Second, the Bayesian nonparametric component allows us to place a prior distribution on the number of features required to capture the latent structure in the data. Third, the latent features in the model are binary-valued, which facilitates the interpretability of the obtained latent features in exploratory data analysis. Finally, a software package, called GLFM toolbox, is made publicly available for other researchers to use and extend. It is available at https://ivaleram.github.io/GLFM/. We show the flexibility of the proposed model by solving both prediction and data analysis tasks on several real-world datasets.

This paper concerns the introduction of a new Markov Chain Monte Carlo scheme for posterior sampling in Bayesian nonparametric mixture models with priors that belong to the general Poisson-Kingman class. We present a novel and compact way of representing the infinite dimensional component of the model such that while explicitly representing this infinite component it has less memory and storage requirements than previous MCMC schemes. We describe comparative simulation results demonstrating the efficacy of the proposed MCMC algorithm against existing marginal and conditional MCMC samplers.

A Kernel Adaptive Metropolis-Hastings algorithm is introduced, for the purpose of sampling from a target distribution with strongly nonlinear support. The algorithm embeds the trajectory of the Markov chain into a reproducing kernel Hilbert space (RKHS), such that the feature space covariance of the samples informs the choice of proposal. The procedure is computationally efficient and straightforward to implement, since the RKHS moves can be integrated out analytically: our proposal distribution in the original space is a normal distribution whose mean and covariance depend on where the current sample lies in the support of the target distribution, and adapts to its local covariance structure. Furthermore, the procedure requires neither gradients nor any other higher order information about the target, making it particularly attractive for contexts such as Pseudo-Marginal MCMC. Kernel Adaptive Metropolis-Hastings outperforms competing fixed and adaptive samplers on multivariate, highly nonlinear target distributions, arising in both real-world and synthetic examples.