Soft prompts have been recently proposed as a tool for adapting large frozen language models (LMs) to new tasks. In this work, we repurpose soft prompts to the task of injecting world knowledge into LMs. We introduce a method to train soft prompts via self-supervised learning on data from knowledge bases. The resulting soft knowledge prompts (KPs) are task independent and work as an external memory of the LMs. We perform qualitative and quantitative experiments and demonstrate that: (1) KPs can effectively model the structure of the training data; (2) KPs can be used to improve the performance of LMs in different knowledge intensive tasks.
Much recent research on information retrieval has focused on how to transfer from one task (typically with abundant supervised data) to various other tasks where supervision is limited, with the implicit assumption that it is possible to generalize from one task to all the rest. However, this overlooks the fact that there are many diverse and unique retrieval tasks, each targeting different search intents, queries, and search domains. In this paper, we suggest to work on Few-shot Dense Retrieval, a setting where each task comes with a short description and a few examples. To amplify the power of a few examples, we propose Prompt-base Query Generation for Retriever (Promptagator), which leverages large language models (LLM) as a few-shot query generator, and creates task-specific retrievers based on the generated data. Powered by LLM's generalization ability, Promptagator makes it possible to create task-specific end-to-end retrievers solely based on a few examples {without} using Natural Questions or MS MARCO to train %question generators or dual encoders. Surprisingly, LLM prompting with no more than 8 examples allows dual encoders to outperform heavily engineered models trained on MS MARCO like ColBERT v2 by more than 1.2 nDCG on average on 11 retrieval sets. Further training standard-size re-rankers using the same generated data yields another 5.0 point nDCG improvement. Our studies determine that query generation can be far more effective than previously observed, especially when a small amount of task-specific knowledge is given.
Cross-domain recommendation (CDR) can help customers find more satisfying items in different domains. Existing CDR models mainly use common users or mapping functions as bridges between domains but have very limited exploration in fully utilizing extra knowledge across domains. In this paper, we propose to incorporate the knowledge graph (KG) for CDR, which enables items in different domains to share knowledge. To this end, we first construct a new dataset AmazonKG4CDR from the Freebase KG and a subset (two domain pairs: movies-music, movie-book) of Amazon Review Data. This new dataset facilitates linking knowledge to bridge within- and cross-domain items for CDR. Then we propose a new framework, KG-aware Neural Collective Matrix Factorization (KG-NeuCMF), leveraging KG to enrich item representations. It first learns item embeddings by graph convolutional autoencoder to capture both domain-specific and domain-general knowledge from adjacent and higher-order neighbours in the KG. Then, we maximize the mutual information between item embeddings learned from the KG and user-item matrix to establish cross-domain relationships for better CDR. Finally, we conduct extensive experiments on the newly constructed dataset and demonstrate that our model significantly outperforms the best-performing baselines.
Dual encoders have been used for question-answering (QA) and information retrieval (IR) tasks with good results. There are two major types of dual encoders, Siamese Dual Encoders (SDE), with parameters shared across two encoders, and Asymmetric Dual Encoder (ADE), with two distinctly parameterized encoders. In this work, we explore the dual encoder architectures for QA retrieval tasks. By evaluating on MS MARCO and the MultiReQA benchmark, we show that SDE performs significantly better than ADE. We further propose three different improved versions of ADEs. Based on the evaluation of QA retrieval tasks and direct analysis of the embeddings, we demonstrate that sharing parameters in projection layers would enable ADEs to perform competitively with SDEs.
Recent neural network-based language models have benefited greatly from scaling up the size of training datasets and the number of parameters in the models themselves. Scaling can be complicated due to various factors including the need to distribute computation on supercomputer clusters (e.g., TPUs), prevent bottlenecks when infeeding data, and ensure reproducible results. In this work, we present two software libraries that ease these issues: $\texttt{t5x}$ simplifies the process of building and training large language models at scale while maintaining ease of use, and $\texttt{seqio}$ provides a task-based API for simple creation of fast and reproducible training data and evaluation pipelines. These open-source libraries have been used to train models with hundreds of billions of parameters on datasets with multiple terabytes of training data. Along with the libraries, we release configurations and instructions for T5-like encoder-decoder models as well as GPT-like decoder-only architectures. $\texttt{t5x}$ and $\texttt{seqio}$ are open source and available at https://github.com/google-research/t5x and https://github.com/google/seqio, respectively.
In this paper, we demonstrate that information retrieval can be accomplished with a single Transformer, in which all information about the corpus is encoded in the parameters of the model. To this end, we introduce the Differentiable Search Index (DSI), a new paradigm that learns a text-to-text model that maps string queries directly to relevant docids; in other words, a DSI model answers queries directly using only its parameters, dramatically simplifying the whole retrieval process. We study variations in how documents and their identifiers are represented, variations in training procedures, and the interplay between models and corpus sizes. Experiments demonstrate that given appropriate design choices, DSI significantly outperforms strong baselines such as dual encoder models. Moreover, DSI demonstrates strong generalization capabilities, outperforming a BM25 baseline in a zero-shot setup.
Recent work has shown that either (1) increasing the input length or (2) increasing model size can improve the performance of Transformer-based neural models. In this paper, we present a new model, called LongT5, with which we explore the effects of scaling both the input length and model size at the same time. Specifically, we integrated attention ideas from long-input transformers (ETC), and adopted pre-training strategies from summarization pre-training (PEGASUS) into the scalable T5 architecture. The result is a new attention mechanism we call {\em Transient Global} (TGlobal), which mimics ETC's local/global attention mechanism, but without requiring additional side-inputs. We are able to achieve state-of-the-art results on several summarization tasks and outperform the original T5 models on question answering tasks.
It has been shown that dual encoders trained on one domain often fail to generalize to other domains for retrieval tasks. One widespread belief is that the bottleneck layer of a dual encoder, where the final score is simply a dot-product between a query vector and a passage vector, is too limited to make dual encoders an effective retrieval model for out-of-domain generalization. In this paper, we challenge this belief by scaling up the size of the dual encoder model {\em while keeping the bottleneck embedding size fixed.} With multi-stage training, surprisingly, scaling up the model size brings significant improvement on a variety of retrieval tasks, especially for out-of-domain generalization. Experimental results show that our dual encoders, \textbf{G}eneralizable \textbf{T}5-based dense \textbf{R}etrievers (GTR), outperform %ColBERT~\cite{khattab2020colbert} and existing sparse and dense retrievers on the BEIR dataset~\cite{thakur2021beir} significantly. Most surprisingly, our ablation study finds that GTR is very data efficient, as it only needs 10\% of MS Marco supervised data to achieve the best out-of-domain performance. All the GTR models are released at https://tfhub.dev/google/collections/gtr/1.
Despite the recent success of multi-task learning and transfer learning for natural language processing (NLP), few works have systematically studied the effect of scaling up the number of tasks during pre-training. Towards this goal, this paper introduces ExMix (Extreme Mixture): a massive collection of 107 supervised NLP tasks across diverse domains and task-families. Using ExMix, we study the effect of multi-task pre-training at the largest scale to date, and analyze co-training transfer amongst common families of tasks. Through this analysis, we show that manually curating an ideal set of tasks for multi-task pre-training is not straightforward, and that multi-task scaling can vastly improve models on its own. Finally, we propose ExT5: a model pre-trained using a multi-task objective of self-supervised span denoising and supervised ExMix. Via extensive experiments, we show that ExT5 outperforms strong T5 baselines on SuperGLUE, GEM, Rainbow, Closed-Book QA tasks, and several tasks outside of ExMix. ExT5 also significantly improves sample efficiency while pre-training.
We provide the first exploration of text-to-text transformers (T5) sentence embeddings. Sentence embeddings are broadly useful for language processing tasks. While T5 achieves impressive performance on language tasks cast as sequence-to-sequence mapping problems, it is unclear how to produce sentence embeddings from encoder-decoder models. We investigate three methods for extracting T5 sentence embeddings: two utilize only the T5 encoder and one uses the full T5 encoder-decoder model. Our encoder-only models outperforms BERT-based sentence embeddings on both transfer tasks and semantic textual similarity (STS). Our encoder-decoder method achieves further improvement on STS. Scaling up T5 from millions to billions of parameters is found to produce consistent improvements on downstream tasks. Finally, we introduce a two-stage contrastive learning approach that achieves a new state-of-art on STS using sentence embeddings, outperforming both Sentence BERT and SimCSE.