As demonstrated by GPT-3 and T5, transformers grow in capability as parameter spaces become larger and larger. However, for tasks that require a large amount of knowledge, non-parametric memory allows models to grow dramatically with a sub-linear increase in computational cost and GPU memory requirements. Recent models such as RAG and REALM have introduced retrieval into conditional generation. These models incorporate neural initial retrieval from a corpus of passages. We build on this line of research, proposing Re2G, which combines both neural initial retrieval and reranking into a BART-based sequence-to-sequence generation. Our reranking approach also permits merging retrieval results from sources with incomparable scores, enabling an ensemble of BM25 and neural initial retrieval. To train our system end-to-end, we introduce a novel variation of knowledge distillation to train the initial retrieval, reranker and generation using only ground truth on the target sequence output. We find large gains in four diverse tasks: zero-shot slot filling, question answering, fact checking and dialog, with relative gains of 9% to 34% over the previous state-of-the-art on the KILT leaderboard. We make our code available as open source.

Automatically inducing high quality knowledge graphs from a given collection of documents still remains a challenging problem in AI. One way to make headway for this problem is through advancements in a related task known as slot filling. In this task, given an entity query in form of [Entity, Slot, ?], a system is asked to ‘fill’ the slot by generating or extracting the missing value exploiting evidence extracted from relevant passage(s) in the given document collection. The recent works in the field try to solve this task in an end-to-end fashion using retrieval-based language models. In this paper, we present a novel approach to zero-shot slot filling that extends dense passage retrieval with hard negatives and robust training procedures for retrieval augmented generation models. Our model reports large improvements on both T-REx and zsRE slot filling datasets, improving both passage retrieval and slot value generation, and ranking at the top-1 position in the KILT leaderboard. Moreover, we demonstrate the robustness of our system showing its domain adaptation capability on a new variant of the TACRED dataset for slot filling, through a combination of zero/few-shot learning. We release the source code and pre-trained models.

Noun phrases and Relation phrases in open knowledge graphs are not canonicalized, leading to an explosion of redundant and ambiguous subject-relation-object triples. Existing approaches to solve this problem take a two-step approach. First, they generate embedding representations for both noun and relation phrases, then a clustering algorithm is used to group them using the embeddings as features. In this work, we propose Canonicalizing Using Variational AutoEncoders and Side Information (CUVA), a joint model to learn both embeddings and cluster assignments in an end-to-end approach, which leads to a better vector representation for the noun and relation phrases. Our evaluation over multiple benchmarks shows that CUVA outperforms the existing state-of-the-art approaches. Moreover, we introduce CanonicNell, a novel dataset to evaluate entity canonicalization systems.

We address relation extraction as an analogy problem by proposing a novel approach to learn representations of relations expressed by their textual mentions. In our assumption, if two pairs of entities belong to the same relation, then those two pairs are analogous. Following this idea, we collect a large set of analogous pairs by matching triples in knowledge bases with web-scale corpora through distant supervision. We leverage this dataset to train a hierarchical siamese network in order to learn entity-entity embeddings which encode relational information through the different linguistic paraphrasing expressing the same relation. We evaluate our model in a one-shot learning task by showing a promising generalization capability in order to classify unseen relation types, which makes this approach suitable to perform automatic knowledge base population with minimal supervision. Moreover, the model can be used to generate pre-trained embeddings which provide a valuable signal when integrated into an existing neural-based model by outperforming the state-of-the-art methods on a downstream relation extraction task.

The textual similarity is a crucial aspect for many extractive text summarization methods. A bag-of-words representation does not allow to grasp the semantic relationships between concepts when comparing strongly related sentences with no words in common. To overcome this issue, in this paper we propose a centroid-based method for text summarization that exploits the compositional capabilities of word embeddings. The evaluations on multi-document and multilingual datasets prove the effectiveness of the continuous vector representation of words compared to the bag-of-words model. Despite its simplicity, our method achieves good performance even in comparison to more complex deep learning models. Our method is unsupervised and it can be adopted in other summarization tasks.