Papers

Unified opinion role labeling (ORL) aims to detect all possible opinion structures of 'opinion-holder-target' in one shot, given a text. The existing transition-based unified method, unfortunately, is subject to longer opinion terms and fails to solve the term overlap issue. Current top performance has been achieved by employing the span-based graph model, which however still suffers from both high model complexity and insufficient interaction among opinions and roles. In this work, we investigate a novel solution by revisiting the transition architecture, and augmenting it with a pointer network (PointNet). The framework parses out all opinion structures in linear-time complexity, meanwhile breaks through the limitation of any length of terms with PointNet. To achieve the explicit opinion-role interactions, we further propose a unified dependency-opinion graph (UDOG), co-modeling the syntactic dependency structure and the partial opinion-role structure. We then devise a relation-centered graph aggregator (RCGA) to encode the multi-relational UDOG, where the resulting high-order representations are used to promote the predictions in the vanilla transition system. Our model achieves new state-of-the-art results on the MPQA benchmark. Analyses further demonstrate the superiority of our methods on both efficacy and efficiency.

poster session 4 @ red 5, poster session 11 @ red 5, poster session 4, poster session 11

Transformer language models have made tremendous strides in natural language understanding tasks. However, the complexity of natural language makes it challenging to ascertain how accurately these models are tracking the world state underlying the text. Motivated by this issue, we consider the task of language modeling for the game of chess. Unlike natural language, chess notations describe a simple, constrained, and deterministic domain. Moreover, we observe that the appropriate choice of chess notation allows for directly probing the world state, without requiring any additional probing-related machinery. We find that: (a) With enough training data, transformer language models can learn to track pieces and predict legal moves with high accuracy when trained solely on move sequences. (b) For small training sets providing access to board state information during training can yield significant improvements. (c) The success of transformer language models is dependent on access to the entire game history i.e. “full attention”. Approximating this full attention results in a significant performance drop. We propose this testbed as a benchmark for future work on the development and analysis of transformer language models.

poster session 4 @ red 5, poster session 11 @ red 5, poster session 4, poster session 11

Chatbots are designed to carry out human-like conversations across different domains, such as general chit-chat, knowledge exchange, and persona-grounded conversations. To measure the quality of such conversational agents, a dialogue evaluator is expected to conduct assessment across domains as well. However, most of the state-of-the-art automatic dialogue evaluation metrics (ADMs) are not designed for multi-domain evaluation. We are motivated to design a general and robust framework, MDD-Eval, to address the problem. Specifically, we first train a teacher evaluator with human-annotated data to acquire a rating skill to tell good dialogue responses from bad ones in a particular domain and then, adopt a self-training strategy to train a new evaluator with teacher-annotated multi-domain data, that helps the new evaluator to generalize across multiple domains. MDD-Eval is extensively assessed on six dialogue evaluation benchmarks. Empirical results show that the MDD-Eval framework achieves a strong performance with an absolute improvement of 7% over the state-of-the-art ADMs in terms of mean Spearman correlation scores across all the evaluation benchmarks.

poster session 4 @ red 5, poster session 11 @ red 5, poster session 4, poster session 11

Commonsense question answering (CQA) aims to test if models can answer questions regarding commonsense knowledge that everyone knows. Prior works that incorporate external knowledge bases have shown promising results, but knowledge bases are expensive to construct and are often limited to a fixed set of relations. In this paper, we instead focus on better utilizing the \textit{implicit knowledge} stored in pre-trained language models. While researchers have found that the knowledge embedded in pre-trained language models can be extracted by having them fill in the blanks of carefully designed prompts for relation extraction and text classification, it remains unclear if we can adopt this paradigm in CQA where the inputs and outputs take much more flexible forms. To this end, we investigate four translation methods that can translate natural questions into cloze-style sentences to better solicit commonsense knowledge from language models, including a syntactic-based model, an unsupervised neural model, and two supervised neural models. In addition, to combine the different translation methods, we propose to encourage consistency among model predictions on different translated questions with unlabeled data. We demonstrate the effectiveness of our methods on three CQA datasets in zero-shot settings. We show that our methods are complementary to a knowledge base improved model, and combining them can lead to state-of-the-art zero-shot performance. Analyses also reveal distinct characteristics of the different cloze translation methods and provide insights on why combining them can lead to great improvements.

poster session 4 @ red 5, poster session 11 @ red 5, oral session 4 @ red 5, poster session 4, poster session 11, oral session 4

As the labeling cost for different modules in task-oriented dialog (ToD) systems is high, a major challenge is to learn different tasks with the least amount of labeled data. Recently, pre-trained language models (PLMs) have shown promising results for few-shot learning in ToD. To better utilize the power of PLMs, this paper proposes Comprehensive Instruction (CINS) that exploits PLMs with extra task-specific instructions. We design a schema (definition, constraint, prompt) of instructions and their customized realizations for three important downstream tasks in ToD, ie. intent classification, dialog state tracking, and natural language generation. A sequence-to-sequence model (T5) is adopted to solve these three tasks in a unified framework. Extensive experiments are conducted on these ToD tasks in realistic few-shot learning scenarios with small validation data. Empirical results demonstrate that the proposed CINS approach consistently improves techniques that finetune PLMs with raw input or short prompt.

poster session 4 @ red 5, poster session 11 @ red 5, oral session 11 @ red 5, poster session 4, poster session 11, oral session 11

We target the task of cross-lingual Machine Reading Comprehension (MRC) in the direct zero-shot setting, by incorporating syntactic features from Universal Dependencies (UD), and the key features we use are the syntactic relations within each sentence. While previous work has demonstrated effective syntax-guided MRC models, we propose to adopt the inter-sentence syntactic relations, in addition to the rudimentary intra-sentence relations, to further utilize the syntactic dependencies in the multi-sentence input of the MRC task. In our approach, we build the Inter-Sentence Dependency Graph (ISDG) connecting dependency trees to form global syntactic relations across sentences. We then propose the ISDG encoder that encodes the global dependency graph, addressing the inter-sentence relations via both one-hop and multi-hop dependency paths explicitly. Experiments on three multilingual MRC datasets (XQuAD, MLQA, TyDiQA-GoldP) show that our encoder that is only trained on English is able to improve the zero-shot performance on all 14 test sets covering 8 languages, with up to 3.8 F1 / 5.2 EM improvement on-average, and 5.2 F1 / 11.2 EM on certain languages. Further analysis shows the improvement can be attributed to the attention on the cross-linguistically consistent syntactic path.

poster session 4 @ red 5, poster session 11 @ red 5, oral session 4 @ red 5, poster session 4, poster session 11, oral session 4

Knowledge Graph Question Answering (KGQA) involves retrieving facts from a Knowledge Graph (KG) using natural language queries. A KG is a curated set of facts consisting of entities linked by relations. Certain facts include also temporal information forming a Temporal KG (TKG). Although many natural questions involve explicit or implicit time constraints, question answering (QA) over TKGs has been a relatively unexplored area. Existing solutions are mainly designed for simple temporal questions that can be answered directly by a single TKG fact. This paper puts forth a comprehensive embedding-based framework for answering complex questions over TKGs. Our method termed temporal question reasoning (TempoQR) exploits TKG embeddings to ground the question to the specific entities and time scope it refers to. It does so by augmenting the question embeddings with context, entity and time-aware information by employing three specialized modules. The first computes a textual representation of a given question, the second combines it with the entity embeddings for entities involved in the question, and the third generates question-specific time embeddings. Finally, a transformer-based encoder learns to fuse the generated temporal information with the question representation, which is used for answer predictions. Extensive experiments show that TempoQR improves accuracy by 25--45 percentage points on complex temporal questions over state-of-the-art approaches and it generalizes better to unseen question types.

poster session 4 @ red 5, poster session 11 @ red 5, poster session 4, poster session 11

Dialogue is an essential part of human communication and cooperation. Existing research mainly focuses on short dialogue scenarios in a one-on-one fashion. However, multi-person interactions in the real world, such as meetings or interviews, are frequently over a few thousand words. There is still a lack of corresponding research and powerful tools to understand and process such long dialogues. Therefore, in this work, we present a pre-training framework for long dialogue understanding and summarization. Considering the nature of long conversations, we propose a window-based denoising approach for generative pre-training. For a dialogue, it corrupts a window of text with dialogue-inspired noise, and guides the model to reconstruct this window based on the content of the remaining conversation. Furthermore, to process longer input, we augment the model with sparse attention which is combined with conventional attention in a hybrid manner. We conduct extensive experiments on five datasets of long dialogues, covering tasks of dialogue summarization, abstractive question answering and topic segmentation. Experimentally, we show that our pre-trained model \DialogLM significantly surpasses the state-of-the-art models across datasets and tasks.

poster session 4 @ red 5, poster session 11 @ red 5, oral session 11 @ red 5, poster session 4, poster session 11, oral session 11

The demands of the modern world are increasingly responsible for causing psychological burdens and bringing adverse impacts on our mental health. As a result, neural conversational agents with empathetic responding and distress management capabilities have recently gained popularity. However, existing end-to-end empathetic conversational agents often generate generic and repetitive empathetic statements such as "I am sorry to hear that", which fail to convey specificity to a given situation. Due to the lack of controllability in such models, they also impose the risk of generating toxic responses. Chatbots leveraging reasoning over knowledge graphs is seen as an efficient and fail-safe solution over end-to-end models. However, such resources are limited in the context of emotional distress. To address this, we introduce HEAL, a knowledge graph developed based on 1M distress narratives and their corresponding consoling responses curated from Reddit. It consists of 22K nodes identifying different types of stressors, speaker expectations, responses, and feedback types associated with distress dialogues and forms 104K connections between different types of nodes. Each node is associated with one of 41 affective states. Statistical and visual analysis conducted on HEAL reveals emotional dynamics between speakers and listeners in distress-oriented conversations and identifies useful response patterns leading to emotional relief. Automatic and human evaluation experiments show that HEAL's responses are more diverse, empathetic, and reliable compared to the baselines.

poster session 4 @ red 5, poster session 11 @ red 5, oral session 11 @ red 5, poster session 4, poster session 11, oral session 11

Recently, there has been an increasing interest in building question answering (QA) models that reason across multiple modalities, such as text and images. However, QA using images is often limited to just picking the answer from a pre-defined set of options. In addition, images in the real world, especially in news, have objects that are co-referential to the text, with complementary information from both modalities. In this paper, we present a new QA evaluation benchmark with 1,384 questions over news articles that require cross-media grounding of objects in images onto text. Specifically, the task involves multi-hop questions that require reasoning over image-caption pairs to identify the grounded visual object being referred to and then predicting a span from the news body text to answer the question. In addition, we introduce a novel multimedia data augmentation framework, based on cross-media knowledge extraction and synthetic question-answer generation, to automatically augment data that can provide weak supervision for this task. We evaluate both pipeline-based and end-to-end pretraining-based multimedia QA models on our benchmark, and show that they achieve promising performance, while considerably lagging behind human performance hence leaving large room for future work on this challenging new task.

poster session 4 @ red 5, poster session 11 @ red 5, poster session 4, poster session 11

Recently, neural networks based purely on self-attention, such as the Vision Transformer (ViT), have been shown to outperform deep learning models constructed with convolutional neural networks (CNNs) on various vision tasks, thus extending the success of Transformers, which were originally developed for language processing, to the vision domain. A recent study showed that a similar methodology can also be applied to the audio domain. Specifically, the Audio Spectrogram Transformer (AST) achieves state-of-the-art results on various audio classification benchmarks. However, pure Transformer models tend to require more training data compared to CNNs, and the success of the AST relies on supervised pretraining that requires a large amount of labeled data and a complex training pipeline, thus limiting the practical usage of AST. This paper focuses on audio and speech classification, and aims to reduce the need for large amounts of labeled data for the AST by leveraging self-supervised learning using unlabeled data. Specifically, we propose to pretrain the AST model with joint discriminative and generative masked spectrogram patch modeling (MSPM) using unlabeled audio from AudioSet and Librispeech. We evaluate our pretrained models on both audio and speech classification tasks including audio event classification, keyword spotting, emotion recognition, and speaker identification. The proposed self-supervised framework significantly boosts AST performance on all tasks, with an average improvement of 60.9%, leading to similar or even better results than a supervised pretrained AST. To the best of our knowledge, it is the first patch-based self-supervised learning framework in the audio and speech domain, and also the first self-supervised learning framework for AST.

poster session 4 @ red 5, poster session 11 @ red 5, poster session 4, poster session 11

We introduce a Recursive INsertion-based Encoder (RINE), a novel approach for semantic parsing in task-oriented dialog. Our model consists of an encoder network that incrementally builds the semantic parse tree by predicting the non-terminal label and its positions in the linearized tree. At the generation time, the model constructs the semantic parse tree by recursively inserting the predicted non-terminal labels at the predicted positions until termination. RINE achieves state-of-the-art exact match accuracy on low- and high-resource versions of the conversational semantic parsing benchmark TOP (Gupta et al., 2018; Chen et al., 2020), outperforming strong sequence-to-sequence models and transition-based parsers. We also show that our model design is applicable to nested named entity recognition task, where it performs on par with state-of-the-art approach designed for that task. Finally, we demonstrate that our approach is 2-3.5 times faster than the sequence-to-sequence model at inference time.

poster session 4 @ red 5, poster session 11 @ red 5, poster session 4, poster session 11

Program translation aims to translate source code from one programming language to another. It is particularly useful in applications such as multiple-platform adaptation and legacy code migration. Traditional rule-based program translation methods usually rely on meticulous manual rule-crafting, which is costly both in terms of time and effort. Recently, neural network based methods have been developed to address this problem. However, the absence of high-quality parallel code data is one of the main bottlenecks which impedes the development of program translation models. In this paper, we introduce CoST, a new multilingual Code Snippet Translation dataset that contains parallel data from 7 commonly used programming languages. The dataset is parallel at the level of code snippets, which provides much more fine-grained alignments between different languages than the existing translation datasets. We also propose a new program translation model that leverages cross-lingual snippet denoising auto-encoding and Multilingual Snippet Translation (MuST) pre-training. Extensive experiments shows that the multilingual snippet training is effective in improving program translation performance, especially for low-resource languages. Moreover, our training method shows good generalizability and consistently improves the translation performance of a number of baseline models. The proposed model outperforms the baselines on both snippet-level and program-level translation, and achieves state-of-the-art performance on CodeXGLUE translation task.

poster session 4 @ red 5, poster session 11 @ red 5, poster session 4, poster session 11

Effectively integrating knowledge into end-to-end task-oriented dialog systems remains a challenge. It typically requires incorporation of an external knowledge base (KB) and capture of the intrinsic semantics of the dialog history. Recent research shows promising results by using Sequence-to-Sequence models, Memory Networks, and even Graph Convolutional Networks. However, current state-of-the-art models are less effective at integrating dialog history and KB into task-oriented dialog systems in the following ways: \textbf{1}. The KB representation is not fully context-aware. The dynamic interaction between the dialog history and KB is seldom explored. \textbf{2}. Both the sequential and structural information in the dialog history can contribute to capturing the dialog semantics, but they are not studied concurrently. In this paper, we propose a novel Graph Memory Network (GMN) based Seq2Seq model, GraphMemDialog, to effectively learn the inherent structural information hidden in dialog history, and to model the dynamic interaction between dialog history and KBs. We adopt a modified graph attention network to learn the rich structural representation of the dialog history, whereas the context-aware representation of KB entities are learnt by our novel GMN. To fully exploit this dynamic interaction, we design a learnable memory controller coupled with external KB entity memories to recurrently incorporate dialog history context into KB entities through a multi-hop reasoning mechanism. Experiments on three public datasets show that our GraphMemDialog model achieves state-of-the-art performance and outperforms strong baselines by a large margin, especially on datatests with more complicated KB information.

poster session 4 @ red 5, poster session 11 @ red 5, poster session 4, poster session 11

Transformer models have achieved promising results on natural language processing (NLP) tasks including extractive question answering (QA). Common Transformer encoders used in NLP tasks process the hidden states of all input tokens in the context paragraph throughout all layers. However, different from other tasks such as sequence classification, answering the raised question does not necessarily need all the tokens in the context paragraph. Following this motivation, we propose Block-skim, which learns to skim unnecessary context in higher hidden layers to improve and accelerate the Transformer performance. The key idea of Block-Skim is to identify the context that must be further processed and those that could be safely discarded early on during inference. Critically, we find that such information could be sufficiently derived from the self-attention weights inside the Transformer model. We further prune the hidden states corresponding to the unnecessary positions early in lower layers, achieving significant inference-time speedup. To our surprise, we observe that models pruned in this way outperform their full-size counterparts. Block-Skim improves QA models' accuracy on different datasets and achieves 3 times speedup on BERT-base model.

poster session 4 @ red 5, poster session 11 @ red 5, poster session 4, poster session 11

Utterance rewriting aims to recover coreferences and omitted information from the latest turn of a multi-turn dialogue. Recently, methods that tag rather than linearly generate sequences have proven stronger in both in- and out-of-domain rewriting settings. This is due to a tagger's smaller search space as it can only copy tokens from the dialogue context. However, these methods may suffer from low coverage when phrases that must be added to a source utterance cannot be covered by a single context span. This can occur in languages like English that introduce tokens such as prepositions into the rewrite for grammaticality. We propose a hierarchical context tagger (HCT) that mitigates this issue by predicting slotted rules (e.g., "besides _") whose slots are later filled with context spans. HCT (i) tags the source string with token-level edit actions and slotted rules and (ii) fills in the resulting rule slots with spans from the dialogue context. This rule tagging allows HCT to add out-of-context tokens and multiple spans at once; we further cluster the rules to truncate the long tail of the rule distribution. Experiments on several benchmarks show that HCT can outperform state-of-the-art rewriting systems by up to 17.8 BLEU points.

poster session 4 @ red 5, poster session 11 @ red 5, poster session 4, poster session 11

We present a generative neural model for open and multi-turn dialog response generation that relies on a multi-dimension attention process to account for the semantic interdependence between the generated words and the conversational history, so as to identify all the words and utterances that influence each generated response. The performance of the model is evaluated on the wide scope DailyDialog corpus and a comparison is made with two other generative neural architectures, using several machine metrics. The results show that the proposed model improves the state of the art for generation accuracy, and its multi-dimension attention allows for a more detailed tracking of the influential words and utterances in the dialog history for response explainability by the dialog history.

poster session 4 @ red 5, poster session 11 @ red 5, poster session 4, poster session 11

Zero/few-shot transfer to unseen services is a critical challenge in task-oriented dialogue research. The Schema-Guided Dialogue (SGD) dataset introduced a paradigm for enabling models to support an unlimited number of services without additional data collection or re-training through the use of schemas. Schemas describe APIs in natural language, which models consume to understand the services they need to support. However, the impact of the choice of language in these schemas on model performance remains unexplored. We address this by releasing SGD-X, a benchmark for measuring the robustness of dialogue systems to linguistic variations in schemas. SGD-X extends the SGD dataset with crowdsourced variants for every schema, where variants are semantically similar yet stylistically diverse. We evaluate two top-performing dialogue state tracking models on SGD-X and observe that neither generalizes well across schema variants, measured by joint goal accuracy and a novel metric for measuring schema sensitivity. Finally, we present a simple model-agnostic data augmentation method to improve schema robustness and zero-shot generalization to unseen services.

poster session 4 @ red 5, poster session 11 @ red 5, poster session 4, poster session 11

Knowledge-Enhanced Pre-trained Language Models (KEPLMs) are pre-trained models with relation triples injecting from knowledge graphs to improve language understanding abilities.Experiments show that our model outperforms other KEPLMs significantly over zero-shot knowledge probing tasks and multiple knowledge-aware language understanding tasks. To guarantee effective knowledge injection, previous studies integrate models with knowledge encoders for representing knowledge retrieved from knowledge graphs. The operations for knowledge retrieval and encoding bring significant computational burdens, restricting the usage of such models in real-world applications that require high inference speed. In this paper, we propose a novel KEPLM named DKPLM that decomposes knowledge injection process of the pre-trained language models in pre-training, fine-tuning and inference stages, which facilitates the applications of KEPLMs in real-world scenarios. Specifically, we first detect knowledge-aware long-tail entities as the target for knowledge injection, enhancing the KEPLMs' semantic understanding abilities and avoiding injecting redundant information. The embeddings of long-tail entities are replaced by ``pseudo token representations'' formed by relevant knowledge triples. We further design the relational knowledge decoding task for pre-training to force the models to truly understand the injected knowledge by relation triple reconstruction. Experiments show that our model outperforms other KEPLMs significantly over zero-shot knowledge probing tasks and multiple knowledge-aware language understanding tasks. We further show that DKPLM has a higher inference speed than other competing models due to the decomposing mechanism.

poster session 4 @ red 5, poster session 11 @ red 5, poster session 4, poster session 11

Cross-lingual Machine Reading Comprehension (xMRC) is a challenging task due to the lack of training data in low-resource languages. Recent approaches use training data only in a resource-rich language (such as English) to fine-tune large-scale cross-lingual pre-trained language models, which transfer knowledge from resource-rich languages (source) to low-resource languages (target). Due to the big difference between languages, the model fine-tuned only by the source language may not perform well for target languages. In our study, we make an interesting observation that while the top 1 result predicted by the previous approaches may often fail to hit the ground-truth answer, there are still good chances for the correct answer to be contained in the set of top k predicted results. Intuitively, the previous approaches have empowered the model certain level of capability to roughly distinguish good answers from bad ones. However, without sufficient training data, it is not powerful enough to capture the nuances between the accurate answer and those approximate ones. Based on this observation, we develop a two-stage approach to enhance the model performance. The first stage targets at recall; we design a hard-learning (HL) algorithm to maximize the likelihood that the top k predictions contain the accurate answer. The second stage focuses on precision, where an answer-aware contrastive learning (AA-CL) mechanism is developed to learn the minute difference between the accurate answer and other candidates. Extensive experiments show that our model significantly outperforms strong baselines on two cross-lingual MRC benchmark datasets.

poster session 4 @ red 5, poster session 11 @ red 5, poster session 4, poster session 11

Spatio-temporal scene-graph approaches to video-based reasoning tasks such as video question-answering (QA) typically construct such graphs for every video frame. Such approaches often ignore the fact that videos are essentially sequences of 2D ``views'' of events happening in a 3D space, and that the semantics of the 3D scene can thus be carried over from frame to frame. Leveraging this insight, we propose a (2.5+1)D scene graph representation to better capture the spatio-temporal information flows inside the videos. Specifically, we first create a 2.5D (pseudo-3D) scene graph by transforming every 2D frame to have an inferred 3D structure using an off-the-shelf 2D-to-3D transformation module, following which we register the video frames into a shared (2.5+1)D spatio-temporal space and ground each 2D scene graph within it. Such a (2.5+1)D graph is then segregated into a static sub-graph and a dynamic sub-graph, corresponding to whether the objects within them usually move in the world. The nodes in the dynamic graph are enriched with motion features capturing their interactions with other graph nodes. Next, for the video QA task, we present a novel transformer-based reasoning pipeline that embeds the (2.5+1)D graph into a spatio-temporal hierarchical latent space, where the sub-graphs and their interactions are captured at varied granularity. To demonstrate the effectiveness of our approach, we present experiments on the NExT-QA and AVSD-QA datasets. Our results show that our proposed (2.5+1)D representation leads to faster training and inference, while our hierarchical model showcases superior performance on the video QA task versus the state of the art.

poster session 4 @ red 5, poster session 11 @ red 5, oral session 4 @ red 5, poster session 4, poster session 11, oral session 4

Keyphrase generation aims to summarize long documents with a collection of salient phrases. Deep neural models have demonstrated remarkable success in this task, capable of predicting keyphrases that are even absent from a document. However, such abstractiveness is acquired at the expense of a substantial amount of annotated data. In this paper, we present a novel method for keyphrase generation, AutoKeyGen, without the supervision of any human annotation. Motivated by the observation that an absent keyphrase in one document can appear in other places, in whole or in part, we first construct a phrase bank by pooling all phrases in a corpus. With this phrase bank, we then draw candidate absent keyphrases for each document through a partial matching process. To rank both types of candidates, we combine their lexical- and semantic-level similarities to the input document. Moreover, we utilize these top-ranked candidates as to train a deep generative model for more absent keyphrases. Extensive experiments demonstrate that AutoKeyGen outperforms all unsupervised baselines and can even beat strong supervised methods in certain cases.

poster session 4 @ red 5, poster session 11 @ red 5, poster session 4, poster session 11

We propose JoTA (Job Taxonomy Alignment), a domain-independent, knowledge-poor method for automatic taxonomy alignment of lexical taxonomies via word embeddings. JoTA associates all the leaf terms of the origin taxonomy to one or many concepts in the destination one, employing a scoring function, which merges the score of a hierarchical method and the score of a classification task. JoTA is developed in the context of an EU Grant aiming at bridging the national taxonomies of EU countries towards the European Skills, Competences, Qualifications and Occupations taxonomy (ESCO) through AI. The method reaches a 0.8 accuracy on recommending top-5 occupations and a wMRR of 0.72.

poster session 4 @ red 5, poster session 11 @ red 5, poster session 4, poster session 11

Large transformer-based language models have achieved incredible success at various tasks which require narrative comprehension, including story completion, answering questions about stories, and generating stories ex nihilo. However, due to the limitations of finite context windows, these language models struggle to produce or understand stories longer than several thousand tokens. In order to mitigate the document length limitations that come with finite context windows, we introduce a novel architecture that augments story processing with an external dynamic knowledge graph. In contrast to static commonsense knowledge graphs which hold information about the real world, these dynamic knowledge graphs reflect facts extracted from the story being processed. Our architecture uses these knowledge graphs to create information-rich prompts which better facilitate story comprehension than prompts composed only of story text. We apply our architecture to the tasks of question answering and story completion. To complement this line of research, we introduce two long-form question answering tasks, LF-SQuAD and LF-QUOREF, in which the document length exceeds the size of the language model's context window, and introduce a story completion evaluation method that bypasses the stochastic nature of language model generation. We demonstrate broad improvement over typical prompt formulation methods for both question answering and story completion using GPT-2, GPT-3 and XLNet.

poster session 4 @ red 5, poster session 11 @ red 5, poster session 4, poster session 11