In recent years, Retrieval-Augmented Generation (RAG) systems have become a cornerstone of artificial intelligence, attracting considerable attention in a variety of fields. By integrating the strengths of information retrieval and generative models, these systems have shown immense potential to push the boundaries of machine learning applications. Nevertheless, RAG systems still face significant challenges and offer ample room for advancement and innovation.

In recent years, Retrieval Augmented Generation (RAG) systems have emerged as a pivotal component in the field of artificial intelligence, gaining significant attention and importance across various domains. These systems, which combine the strengths of information retrieval and generative models, have shown promise in enhancing the capabilities and performance of machine learning applications. However, despite their growing prominence, RAG systems are not without their limitations and continue to be in need of exploration and improvement. This workshop seeks to focus on the critical aspect of information retrieval and its integral role within RAG frameworks. We argue that current efforts have undervalued the role of Information Retrieval (IR) in the RAG and have concentrated their attention on the generative part. As the cornerstone of these systems, IR's effectiveness dramatically influences the overall performance and outcomes of RAG models. We call for papers that will seek to revisit and emphasize the fundamental principles underpinning RAG systems. At the end of the workshop, we aim to have a clearer understanding of how robust information retrieval mechanisms can significantly enhance the capabilities of RAG systems. The workshop will serve as a platform for experts, researchers, and practitioners. We intend to foster discussions, share insights, and encourage research that underscores the vital role of Information Retrieval in the future of generative systems.

Abstract While recent language models have the ability to take long contexts as input, relatively little is known about how well they use longer context. We analyze the performance of language models on two tasks that require identifying relevant information in their input contexts: multi-document question answering and key-value retrieval. We find that performance can degrade significantly when changing the position of relevant information, indicating that current language models do not robustly make use of information in long input contexts. In particular, we observe that performance is often highest when relevant information occurs at the beginning or end of the input context, and significantly degrades when models must access relevant information in the middle of long contexts, even for explicitly long-context models. Our analysis provides a better understanding of how language models use their input context and provides new evaluation protocols for future long-context language models.

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.

In this paper we consider the problem of classifying fine-grained, multi-step activities (e.g., cooking different recipes, making disparate home improvements, creating various forms of arts and crafts) from long videos spanning up to several minutes. Accurately categorizing these activities requires not only recognizing the individual steps that compose the task but also capturing their temporal dependencies. This problem is dramatically different from traditional action classification, where models are typically optimized on videos that span only a few seconds and that are manually trimmed to contain simple atomic actions. While step annotations could enable the training of models to recognize the individual steps of procedural activities, existing large-scale datasets in this area do not include such segment labels due to the prohibitive cost of manually annotating temporal boundaries in long videos. To address this issue, we propose to automatically identify steps in instructional videos by leveraging the distant supervision of a textual knowledge base (wikiHow) that includes detailed descriptions of the steps needed for the execution of a wide variety of complex activities. Our method uses a language model to match noisy, automatically-transcribed speech from the video to step descriptions in the knowledge base. We demonstrate that video models trained to recognize these automatically-labeled steps (without manual supervision) yield a representation that achieves superior generalization performance on four downstream tasks: recognition of procedural activities, step classification, step forecasting and egocentric video classification.

Existing work on Entity Linking mostly assumes that the reference knowledge base is complete, and therefore all mentions can be linked.In practice this is hardly ever the case, as knowledge bases are incomplete and because novel concepts arise constantly.We introduce the temporally segmented Unknown Entity Discovery and Indexing (EDIN) -benchmark where unknown entities, that is entities not part of the knowledge base and without descriptions and labeled mentions, have to be integrated into an existing entity linking system.By contrasting EDIN with zero-shot entity linking, we provide insight on the additional challenges it poses.Building on denseretrieval based entity linking, we introduce the end-to-end EDIN-pipeline that detects, clusters, and indexes mentions of unknown entities in context.Experiments show that indexing a single embedding per entity unifying the information of multiple mentions works better than indexing mentions independently.

Martin Josifoski, Nicola De Cao, Maxime Peyrard, Fabio Petroni, Robert West. Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2022.

Patrick Lewis, Barlas Oguz, Wenhan Xiong, Fabio Petroni, Scott Yih, Sebastian Riedel. Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2022.

Knowledge-intensive language tasks require NLP systems to both provide the correct answer and retrieve supporting evidence for it in a given corpus. Autoregressive language models are emerging as the de-facto standard for generating answers, with newer and more powerful systems emerging at an astonishing pace. In this paper we argue that all this (and future) progress can be directly applied to the retrieval problem with minimal intervention to the models' architecture. Previous work has explored ways to partition the search space into hierarchical structures and retrieve documents by autoregressively generating their unique identifier. In this work we propose an alternative that doesn't force any structure in the search space: using all ngrams in a passage as its possible identifiers. This setup allows us to use an autoregressive model to generate and score distinctive ngrams, that are then mapped to full passages through an efficient data structure. Empirically, we show this not only outperforms prior autoregressive approaches but also leads to an average improvement of at least 10 points over more established retrieval solutions for passage-level retrieval on the KILT benchmark, establishing new state-of-the-art downstream performance on some datasets, while using a considerably lighter memory footprint than competing systems. Code available in the supplementary materials. Pre-trained models will be made available.

Abstract We present mGENRE, a sequence-to- sequence system for the Multilingual Entity Linking (MEL) problem—the task of resolving language-specific mentions to a multilingual Knowledge Base (KB). For a mention in a given language, mGENRE predicts the name of the target entity left-to-right, token-by-token in an autoregressive fashion. The autoregressive formulation allows us to effectively cross-encode mention string and entity names to capture more interactions than the standard dot product between mention and entity vectors. It also enables fast search within a large KB even for mentions that do not appear in mention tables and with no need for large-scale vector indices. While prior MEL works use a single representation for each entity, we match against entity names of as many languages as possible, which allows exploiting language connections between source input and target name. Moreover, in a zero-shot setting on languages with no training data at all, mGENRE treats the target language as a latent variable that is marginalized at prediction time. This leads to over 50% improvements in average accuracy. We show the efficacy of our approach through extensive evaluation including experiments on three popular MEL benchmarks where we establish new state-of-the-art results. Source code available at https://github.com/facebookresearch/GENRE.

The binary similarity problem consists in determining if two functions are similar considering only their compiled form. Advanced techniques for binary similarity recently gained momentum as they can be applied in several fields, such as copyright disputes, malware analysis, vulnerability detection, etc. In this article we describe SAFE, a novel architecture for function representation based on a self-attentive neural network. SAFE works directly on disassembled binary functions, does not require manual feature extraction, is computationally more efficient than existing solutions, and is more general as it works on stripped binaries and on multiple architectures. Results from our experimental evaluation show how SAFE provides a performance improvement with respect to previous solutions. Furthermore, we show how SAFE can be used in widely different use cases, thus providing a general solution for several application scenarios.

Jean Maillard, Vladimir Karpukhin, Fabio Petroni, Wen-tau Yih, Barlas Oguz, Veselin Stoyanov, Gargi Ghosh. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

In many applications that rely on machine learning, the availability of labelled data is a matter of primary importance. However, when tackling new tasks, labels are usually missing and must be collected from scratch by the users. In this work, we address the problem of learning classifiers when the amount of labels is very scarce. We do so by learning multiple vectors, called prototypes, that represent relevant semantic concepts for the task at hand. We propose a theoretically inspired mechanism that computes probabilities of matching between the prototypes and the input elements, and we combine these probabilities to increase the expressiveness of the classifier. Moreover, by leveraging low-cost extra annotations in the training data, a simple error-boosting technique guides the learning process and provides substantial performance improvements. Empirical results confirm the benefits of the proposed approach in both balanced and unbalanced datasets. Our methodology is thus of practical use when gathering and labelling new examples is more expensive than annotating what we already have.

Fabio Petroni, Aleksandra Piktus, Angela Fan, Patrick Lewis, Majid Yazdani, Nicola De Cao, James Thorne, Yacine Jernite, Vladimir Karpukhin, Jean Maillard, Vassilis Plachouras, Tim Rocktäschel, Sebastian Riedel. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2021.

Angela Fan, Aleksandra Piktus, Fabio Petroni, Guillaume Wenzek, Marzieh Saeidi, Andreas Vlachos, Antoine Bordes, Sebastian Riedel. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP). 2020.

This paper introduces a conceptually simple, scalable, and highly effective BERT-based entity linking model, along with an extensive evaluation of its accuracy-speed trade-off.We present a two-stage zero-shot linking algorithm, where each entity is defined only by a short textual description.The first stage does retrieval in a dense space defined by a bi-encoder that independently embeds the mention context and the entity descriptions.Each candidate is then re-ranked with a crossencoder, that concatenates the mention and entity text.Experiments demonstrate that this approach is state of the art on recent zeroshot benchmarks (6 point absolute gains) and also on more established non-zero-shot evaluations (e.g.TACKBP-2010), despite its relative simplicity (e.g.no explicit entity embeddings or manually engineered mention tables).We also show that bi-encoder linking is very fast with nearest neighbour search (e.g.linking with 5.9 million candidates in 2 milliseconds), and that much of the accuracy gain from the more expensive crossencoder can be transferred to the bi-encoder via knowledge distillation.Our code and models are available at https://github. com/facebookresearch/BLINK. wikipedia dense spaceMy kids really enjoyed a ride in the Jaguar!Jaguar is the luxury vehicle brand. Jaguar_carsJaguar! is a junior roller coaster.

Large pre-trained language models have been shown to store factual knowledge in their parameters, and achieve state-of-the-art results when fine-tuned on downstream NLP tasks. However, their ability to access and precisely manipulate knowledge is still limited, and hence on knowledge-intensive tasks, their performance lags behind task-specific architectures. Additionally, providing provenance for their decisions and updating their world knowledge remain open research problems. Pre-trained models with a differentiable access mechanism to explicit non-parametric memory can overcome this issue, but have so far been only investigated for extractive downstream tasks. We explore a general-purpose fine-tuning recipe for retrieval-augmented generation (RAG) -- models which combine pre-trained parametric and non-parametric memory for language generation. We introduce RAG models where the parametric memory is a pre-trained seq2seq model and the non-parametric memory is a dense vector index of Wikipedia, accessed with a pre-trained neural retriever. We compare two RAG formulations, one which conditions on the same retrieved passages across the whole generated sequence, the other can use different passages per token. We fine-tune and evaluate our models on a wide range of knowledge-intensive NLP tasks and set the state-of-the-art on three open domain QA tasks, outperforming parametric seq2seq models and task-specific retrieve-and-extract architectures. For language generation tasks, we find that RAG models generate more specific, diverse and factual language than a state-of-the-art parametric-only seq2seq baseline.

Fabio Petroni, Tim Rocktäschel, Sebastian Riedel, Patrick Lewis, Anton Bakhtin, Yuxiang Wu, Alexander Miller. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.

The automatic extraction of breaking news events from natural language text is a valuable capability for decision support systems. Traditional systems tend to focus on extracting events from a single media source and often ignore cross-media references. Here, we describe a large-scale automated system for extracting natural disasters and critical events from both newswire text and social media. We outline a comprehensive architecture that can identify, categorize and summarize seven different event types - namely floods, storms, fires, armed conflict, terrorism, infrastructure breakdown, and labour unavailability. The system comprises fourteen modules and is equipped with a novel coreference mechanism, capable of linking events extracted from the two complementary data sources. Additionally, the system is easily extensible to accommodate new event types. Our experimental evaluation demonstrates the effectiveness of the system.

Fabio Petroni, Vassilis Plachouras, Timothy Nugent, Jochen L. Leidner. Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers). 2018.

We present a contrastive study of document-level event classification of a range of seven different event types, namely floods, storms, fires, armed conflict, terrorism, infrastructure breakdown and labour unavailability from English-language news. Our study compares different supervised classification approaches, namely Support Vector Machine (SVM), Random Forest (RF), Convolutional Neural Network (CNN) and Hierarchical Attention Network (HAN). While past systems for Topic Detection and Tracking (TDT) and event extraction have proposed different machine learning models, to date SVMs, RFs, CNNs and HANs have not been compared on this task. Our classifiers are also informed by word embeddings trained on large amounts of high-quality agency news, which leads to improvements compared to the use of pre-trained embedding vectors. We report a detailed quantitative error analysis.

Balanced graph partitioning is a fundamental problem that is receiving growing attention with the emergence of distributed graph-computing (DGC) frameworks. In these frameworks, the partitioning strategy plays an important role since it drives the communication cost and the workload balance among computing nodes, thereby affecting system performance. However, existing solutions only partially exploit a key characteristic of natural graphs commonly found in the real-world: their highly skewed power-law degree distributions. In this paper, we propose High-Degree (are) Replicated First (HDRF), a novel streaming vertex-cut graph partitioning algorithm that effectively exploits skewed degree distributions by explicitly taking into account vertex degree in the placement decision. We analytically and experimentally evaluate HDRF on both synthetic and real-world graphs and show that it outperforms all existing algorithms in partitioning quality.

We propose CORE, a novel matrix factorization model that leverages contextual information for open relation extraction.Our model is based on factorization machines and integrates facts from various sources, such as knowledge bases or open information extractors, as well as the context in which these facts have been observed.We argue that integrating contextual information-such as metadata about extraction sources, lexical context, or type information-significantly improves prediction performance.Open information extractors, for example, may produce extractions that are unspecific or ambiguous when taken out of context.Our experimental study on a large real-world dataset indicates that CORE has significantly better prediction performance than state-ofthe-art approaches when contextual information is available.

Matrix completion latent factors models are known to be an effective method to build recommender systems. Currently, stochastic gradient descent (SGD) is considered one of the best latent factor-based algorithm for matrix completion. In this paper we discuss GASGD, a distributed asynchronous variant of SGD for large-scale matrix completion, that (i) leverages data partitioning schemes based on graph partitioning techniques, (ii) exploits specific characteristics of the input data and (iii) introduces an explicit parameter to tune synchronization frequency among the computing nodes. We empirically show how, thanks to these features, GASGD achieves a fast convergence rate incurring in smaller communication cost with respect to current asynchronous distributed SGD implementations.