Marilyn Walker

Omkar Patil, Lena Reed, Kevin K. Bowden et al. · cmu

Conversational agents are consistently growing in popularity and many people interact with them every day. While many conversational agents act as personal assistants, they can have many different goals. Some are task-oriented, such as providing customer support for a bank or making a reservation. Others are designed to be empathetic and to form emotional connections with the user. The Alexa Prize Challenge aims to create a socialbot, which allows the user to engage in coherent conversations, on a range of popular topics that will interest the user. Here we describe Athena 2.0, UCSC's conversational agent for Amazon's Socialbot Grand Challenge 4. Athena 2.0 utilizes a novel knowledge-grounded discourse model that tracks the entity links that Athena introduces into the dialogue, and uses them to constrain named-entity recognition and linking, and coreference resolution. Athena 2.0 also relies on a user model to personalize topic selection and other aspects of the conversation to individual users.

Angela Ramirez, Mamon Alsalihy, Kartik Aggarwal et al.

Prompt-based or in-context learning has achieved high zero-shot performance on many natural language generation (NLG) tasks. Here we explore the performance of prompt-based learning for simultaneously controlling the personality and the semantic accuracy of an NLG for task-oriented dialogue. We experiment with prompt-based learning on the PERSONAGE restaurant recommendation corpus to generate semantically and stylistically-controlled text for 5 different Big-5 personality types: agreeable, disagreeable, conscientious, unconscientious, and extravert. We test two different classes of discrete prompts to generate utterances for a particular personality style: (1) prompts that demonstrate generating directly from a meaning representation that includes a personality specification; and (2) prompts that rely on first converting the meaning representation to a textual pseudo-reference, and then using the pseudo-reference in a textual style transfer (TST) prompt. In each case, we show that we can vastly improve performance by over-generating outputs and ranking them, testing several ranking functions based on automatic metrics for semantic accuracy, personality-match, and fluency. We also test whether NLG personality demonstrations from the restaurant domain can be used with meaning representations for the video game domain to generate personality stylized utterances about video games. Our findings show that the TST prompts produces the highest semantic accuracy (78.46% for restaurants and 87.6% for video games) and personality accuracy (100% for restaurants and 97% for video games). Our results on transferring personality style to video game utterances are surprisingly good. To our knowledge, there is no previous work testing the application of prompt-based learning to simultaneously controlling both style and semantic accuracy in NLG.

Cat P. Le, Luke Dai, Michael Johnston et al.

Effective evaluation methods remain a significant challenge for research on open-domain conversational dialogue systems. Explicit satisfaction ratings can be elicited from users, but users often do not provide ratings when asked, and those they give can be highly subjective. Post-hoc ratings by experts are an alternative, but these can be both expensive and complex to collect. Here, we explore the creation of automated methods for predicting both expert and user ratings of open-domain dialogues. We compare four different approaches. First, we train a baseline model using an end-to-end transformer to predict ratings directly from the raw dialogue text. The other three methods are variants of a two-stage approach in which we first extract interpretable features at the turn level that capture, among other aspects, user dialogue behaviors indicating contradiction, repetition, disinterest, compliments, or criticism. We project these features to the dialogue level and train a dialogue-level MLP regression model, a dialogue-level LSTM, and a novel causal inference model called counterfactual-LSTM (CF-LSTM) to predict ratings. The proposed CF-LSTM is a sequential model over turn-level features which predicts ratings using multiple regressors depending on hypotheses derived from the turn-level features. As a causal inference model, CF-LSTM aims to learn the underlying causes of a specific event, such as a low rating. We also bin the user ratings and perform classification experiments with all four models. In evaluation experiments on conversational data from the Alexa Prize SocialBot, we show that the CF-LSTM achieves the best performance for predicting dialogue ratings and classification.

Jon Z. Cai, Brendan King, Margaret Perkoff et al.

In this paper, we introduce Dependency Dialogue Acts (DDA), a novel framework for capturing the structure of speaker-intentions in multi-party dialogues. DDA combines and adapts features from existing dialogue annotation frameworks, and emphasizes the multi-relational response structure of dialogues in addition to the dialogue acts and rhetorical relations. It represents the functional, discourse, and response structure in multi-party multi-threaded conversations. A few key features distinguish DDA from existing dialogue annotation frameworks such as SWBD-DAMSL and the ISO 24617-2 standard. First, DDA prioritizes the relational structure of the dialogue units and the dialog context, annotating both dialog acts and rhetorical relations as response relations to particular utterances. Second, DDA embraces overloading in dialogues, encouraging annotators to specify multiple response relations and dialog acts for each dialog unit. Lastly, DDA places an emphasis on adequately capturing how a speaker is using the full dialog context to plan and organize their speech. With these features, DDA is highly expressive and recall-oriented with regard to conversation dynamics between multiple speakers. In what follows, we present the DDA annotation framework and case studies annotating DDA structures in multi-party, multi-threaded conversations.

Vrindavan Harrison, Rishi Rajasekaran, Marilyn Walker

Many open-domain dialogue systems rely on multiple response generators, any of which can contribute a response to the dialogue in a particular context. Thus the ability to compare potential responses and then select the best plays an important role in ensuring a dialogue system is coherent and engaging. Dialogue coherence goes beyond simply remaining on topic -- some trivia may be on topic and engaging when mentioned out of the blue, but may not be coherent and grounded in the context of the conversation. We carry out experiments on response selection in the Athena system, an Alexa Prize SocialBot that has dedicated content and multiple topic-specific response generators for a large number of topics. First, we collect a corpus of Athena conversations with live human traffic, where potential responses from all enabled response generators are logged and subsequently annotated for response quality. We compare several off-the-shelf response ranking methods for open-domain dialogue to Athena-Heuristic, a heuristic response ranker that was field-tested in Athena during the third Alexa Prize competition. We also compare these to a transformer-based response ranker we call Athena-RR, that we train on our Athena conversations. Athena-RR uses both the conversational context and the dialogue state to rank the potential responses. We find that Athena-RR with a Recall@1 of 70.79\% outperforms Athena-Heuristic and all of the off-the-shelf rankers by a large margin. We then conduct a live A/B study comparing Athena-Heuristic to Athena-RR in a 6,358 conversations with Alexa users. We show that Athena-RR leads to significantly longer conversations that receive significantly higher user ratings than the heuristic rule-based ranker.

Kevin K. Bowden, Marilyn Walker

There has been an increased focus on creating conversational open-domain dialogue systems in the spoken dialogue community. Unlike traditional dialogue systems, these conversational systems cannot assume any specific information need or domain restrictions, i.e., the only inherent goal is to converse with the user on an unknown set of topics. While massive improvements in Natural Language Understanding (NLU) and the growth of available knowledge resources can partially support a robust conversation, these conversations generally lack the rapport between two humans that know each other. We developed a robust open-domain conversational system, Athena, that real Amazon Echo users access and evaluate at scale in the context of the Alexa Prize competition. We experiment with methods intended to increase intimacy between Athena and the user by heuristically developing a rule-based user model that personalizes both the current and subsequent conversations and evaluating specific personal opinion question strategies in A/B studies. Our results show a statistically significant positive impact on perceived conversation quality and length when employing these strategies.

Juraj Juraska, Kevin K. Bowden, Lena Reed et al.

Athena 2.0 is an Alexa Prize SocialBot that has been a finalist in the last two Alexa Prize Grand Challenges. One reason for Athena's success is its novel dialogue management strategy, which allows it to dynamically construct dialogues and responses from component modules, leading to novel conversations with every interaction. Here we describe Athena's system design and performance in the Alexa Prize during the 20/21 competition. A live demo of Athena as well as video recordings will provoke discussion on the state of the art in conversational AI.

Marilyn Walker, Colin Harmon, James Graupera et al.

There has recently been an explosion of work on spoken dialogue systems, along with an increased interest in open-domain systems that engage in casual conversations on popular topics such as movies, books and music. These systems aim to socially engage, entertain, and even empathize with their users. Since the achievement of such social goals is hard to measure, recent research has used dialogue length or human ratings as evaluation metrics, and developed methods for automatically calculating novel metrics, such as coherence, consistency, relevance and engagement. Here we develop a PARADISE model for predicting the performance of Athena, a dialogue system that has participated in thousands of conversations with real users, while competing as a finalist in the Alexa Prize. We use both user ratings and dialogue length as metrics for dialogue quality, and experiment with predicting these metrics using automatic features that are both system dependent and independent. Our goal is to learn a general objective function that can be used to optimize the dialogue choices of any Alexa Prize system in real time and evaluate its performance. Our best model for predicting user ratings gets an R$^2$ of .136 with a DistilBert model, and the best model for predicting length with system independent features gets an R$^2$ of .865, suggesting that conversation length may be a more reliable measure for automatic training of dialogue systems.

Lena Reed, Cecilia Li, Angela Ramirez et al.

One challenge with open-domain dialogue systems is the need to produce truthful, high-quality responses on any topic. We aim to improve the quality and coverage of Athena, an Alexa Prize dialogue system. We experiment with few-shot prompt-based learning, comparing GPT-Neo to Jurassic-1, for the movies, music, TV, sports, and video game domains, both within and cross-domain, with different prompt set sizes (2, 3, 10), formats, and meaning representations consisting of either sets of WikiData KG triples, or dialogue acts. Our evaluation uses BLEURT and human metrics, and shows that with 10-shot prompting, Athena-Jurassic's performance is significantly better for coherence and semantic accuracy. Experiments with 2-shot cross-domain prompts results in a huge performance drop for Athena-GPT-Neo, whose semantic accuracy falls to 0.41, and whose untrue hallucination rate increases to 12%. Experiments with dialogue acts for video games show that with 10-shot prompting, both models learn to control dialogue acts, but Athena-Jurassic has significantly higher coherence, and only 4% untrue hallucinations. Our results suggest that Athena-Jurassic produces high enough quality outputs to be useful in live systems with real users. To our knowledge, these are the first results demonstrating that few-shot semantic prompt-based learning can create NLGs that generalize to new domains, and produce high-quality, semantically-controlled, conversational responses directly from meaning representations.

Juraj Juraska, Marilyn Walker

Ever since neural models were adopted in data-to-text language generation, they have invariably been reliant on extrinsic components to improve their semantic accuracy, because the models normally do not exhibit the ability to generate text that reliably mentions all of the information provided in the input. In this paper, we propose a novel decoding method that extracts interpretable information from encoder-decoder models' cross-attention, and uses it to infer which attributes are mentioned in the generated text, which is subsequently used to rescore beam hypotheses. Using this decoding method with T5 and BART, we show on three datasets its ability to dramatically reduce semantic errors in the generated outputs, while maintaining their state-of-the-art quality.

Vrindavan Harrison, Juraj Juraska, Wen Cui et al.

This report describes Athena, a dialogue system for spoken conversation on popular topics and current events. We develop a flexible topic-agnostic approach to dialogue management that dynamically configures dialogue based on general principles of entity and topic coherence. Athena's dialogue manager uses a contract-based method where discourse constraints are dispatched to clusters of response generators. This allows Athena to procure responses from dynamic sources, such as knowledge graph traversals and feature-based on-the-fly response retrieval methods. After describing the dialogue system architecture, we perform an analysis of conversations that Athena participated in during the 2019 Alexa Prize Competition. We conclude with a report on several user studies we carried out to better understand how individual user characteristics affect system ratings.

Lena Reed, Vrindavan Harrison, Shereen Oraby et al.

Natural language generators (NLGs) for task-oriented dialogue typically take a meaning representation (MR) as input. They are trained end-to-end with a corpus of MR/utterance pairs, where the MRs cover a specific set of dialogue acts and domain attributes. Creation of such datasets is labor-intensive and time-consuming. Therefore, dialogue systems for new domain ontologies would benefit from using data for pre-existing ontologies. Here we explore, for the first time, whether it is possible to train an NLG for a new larger ontology using existing training sets for the restaurant domain, where each set is based on a different ontology. We create a new, larger combined ontology, and then train an NLG to produce utterances covering it. For example, if one dataset has attributes for family-friendly and rating information, and the other has attributes for decor and service, our aim is an NLG for the combined ontology that can produce utterances that realize values for family-friendly, rating, decor and service. Initial experiments with a baseline neural sequence-to-sequence model show that this task is surprisingly challenging. We then develop a novel self-training method that identifies (errorful) model outputs, automatically constructs a corrected MR input to form a new (MR, utterance) training pair, and then repeatedly adds these new instances back into the training data. We then test the resulting model on a new test set. The result is a self-trained model whose performance is an absolute 75.4% improvement over the baseline model. We also report a human qualitative evaluation of the final model showing that it achieves high naturalness, semantic coherence and grammaticality

Juraj Juraska, Kevin K. Bowden, Marilyn Walker

The uptake of deep learning in natural language generation (NLG) led to the release of both small and relatively large parallel corpora for training neural models. The existing data-to-text datasets are, however, aimed at task-oriented dialogue systems, and often thus limited in diversity and versatility. They are typically crowdsourced, with much of the noise left in them. Moreover, current neural NLG models do not take full advantage of large training data, and due to their strong generalizing properties produce sentences that look template-like regardless. We therefore present a new corpus of 7K samples, which (1) is clean despite being crowdsourced, (2) has utterances of 9 generalizable and conversational dialogue act types, making it more suitable for open-domain dialogue systems, and (3) explores the domain of video games, which is new to dialogue systems despite having excellent potential for supporting rich conversations.

Kevin K. Bowden, Jiaqi Wu, Wen Cui et al.

Conversational systems typically focus on functional tasks such as scheduling appointments or creating todo lists. Instead we design and evaluate SlugBot (SB), one of 8 semifinalists in the 2018 AlexaPrize, whose goal is to support casual open-domain social inter-action. This novel application requires both broad topic coverage and engaging interactive skills. We developed a new technical approach to meet this demanding situation by crowd-sourcing novel content and introducing playful conversational strategies based on storytelling and games. We collected over 10,000 conversations during August 2018 as part of the Alexa Prize competition. We also conducted an in-lab follow-up qualitative evaluation. Over-all users found SB moderately engaging; conversations averaged 3.6 minutes and involved 26 user turns. However, users reacted very differently to different conversation subtypes. Storytelling and games were evaluated positively; these were seen as entertaining with predictable interactive structure. They also led users to impute personality and intelligence to SB. In contrast, search and general Chit-Chat induced coverage problems; here users found it hard to infer what topics SB could understand, with these conversations seen as being too system-driven. Theoretical and design implications suggest a move away from conversational systems that simply provide factual information. Future systems should be designed to have their own opinions with personal stories to share, and SB provides an example of how we might achieve this.

Kevin K. Bowden, Jiaqi Wu, Wen Cui et al.

One of the most interesting aspects of the Amazon Alexa Prize competition is that the framing of the competition requires the development of new computational models of dialogue and its structure. Traditional computational models of dialogue are of two types: (1) task-oriented dialogue, supported by AI planning models,or simplified planning models consisting of frames with slots to be filled; or (2)search-oriented dialogue where every user turn is treated as a search query that may elaborate and extend current search results. Alexa Prize dialogue systems such as SlugBot must support conversational capabilities that go beyond what these traditional models can do. Moreover, while traditional dialogue systems rely on theoretical computational models, there are no existing computational theories that circumscribe the expected system and user behaviors in the intended conversational genre of the Alexa Prize Bots. This paper describes how UCSC's SlugBot team has combined the development of a novel computational theoretical model, Discourse Relation Dialogue Model, with its implementation in a modular system in order to test and refine it. We highlight how our novel dialogue model has led us to create a novel ontological resource, UniSlug, and how the structure of UniSlug determine show we curate and structure content so that our dialogue manager implements and tests our novel computational dialogue model.

Vrindavan Harrison, Lena Reed, Shereen Oraby et al.

Neural generation methods for task-oriented dialogue typically generate from a meaning representation that is populated using a database of domain information, such as a table of data describing a restaurant. While earlier work focused solely on the semantic fidelity of outputs, recent work has started to explore methods for controlling the style of the generated text while simultaneously achieving semantic accuracy. Here we experiment with two stylistic benchmark tasks, generating language that exhibits variation in personality, and generating discourse contrast. We report a huge performance improvement in both stylistic control and semantic accuracy over the state of the art on both of these benchmarks. We test several different models and show that putting stylistic conditioning in the decoder and eliminating the semantic re-ranker used in earlier models results in more than 15 points higher BLEU for Personality, with a reduction of semantic error to near zero. We also report an improvement from .75 to .81 in controlling contrast and a reduction in semantic error from 16% to 2%.

Mingyu Derek Ma, Kevin K. Bowden, Jiaqi Wu et al.

Discourse relation identification has been an active area of research for many years, and the challenge of identifying implicit relations remains largely an unsolved task, especially in the context of an open-domain dialogue system. Previous work primarily relies on a corpora of formal text which is inherently non-dialogic, i.e., news and journals. This data however is not suitable to handle the nuances of informal dialogue nor is it capable of navigating the plethora of valid topics present in open-domain dialogue. In this paper, we designed a novel discourse relation identification pipeline specifically tuned for open-domain dialogue systems. We firstly propose a method to automatically extract the implicit discourse relation argument pairs and labels from a dataset of dialogic turns, resulting in a novel corpus of discourse relation pairs; the first of its kind to attempt to identify the discourse relations connecting the dialogic turns in open-domain discourse. Moreover, we have taken the first steps to leverage the dialogue features unique to our task to further improve the identification of such relations by performing feature ablation and incorporating dialogue features to enhance the state-of-the-art model.

Shereen Oraby, Vrindavan Harrison, Abteen Ebrahimi et al.

Neural natural language generation (NNLG) from structured meaning representations has become increasingly popular in recent years. While we have seen progress with generating syntactically correct utterances that preserve semantics, various shortcomings of NNLG systems are clear: new tasks require new training data which is not available or straightforward to acquire, and model outputs are simple and may be dull and repetitive. This paper addresses these two critical challenges in NNLG by: (1) scalably (and at no cost) creating training datasets of parallel meaning representations and reference texts with rich style markup by using data from freely available and naturally descriptive user reviews, and (2) systematically exploring how the style markup enables joint control of semantic and stylistic aspects of neural model output. We present YelpNLG, a corpus of 300,000 rich, parallel meaning representations and highly stylistically varied reference texts spanning different restaurant attributes, and describe a novel methodology that can be scalably reused to generate NLG datasets for other domains. The experiments show that the models control important aspects, including lexical choice of adjectives, output length, and sentiment, allowing the models to successfully hit multiple style targets without sacrificing semantics.

Jiaqi Wu, Ryan Compton, Geetanjali Rakshit et al.

We present our system, CruzAffect, for the CL-Aff Shared Task 2019. CruzAffect consists of several types of robust and efficient models for affective classification tasks. We utilize both traditional classifiers, such as XGBoosted Forest, as well as a deep learning Convolutional Neural Networks (CNN) classifier. We explore rich feature sets such as syntactic features, emotional features, and profile features, and utilize several sentiment lexicons, to discover essential indicators of social involvement and control that a subject might exercise in their happy moments, as described in textual snippets from the HappyDB database. The data comes with a labeled set (10K), and a larger unlabeled set (70K). We therefore use supervised methods on the 10K dataset, and a bootstrapped semi-supervised approach for the 70K. We evaluate these models for binary classification of agency and social labels (Task 1), as well as multi-class prediction for concepts labels (Task 2). We obtain promising results on the held-out data, suggesting that the proposed feature sets effectively represent the data for affective classification tasks. We also build concepts models that discover general themes recurring in happy moments. Our results indicate that generic characteristics are shared between the classes of agency, social and concepts, suggesting it should be possible to build general models for affective classification tasks.

Juraj Juraska, Marilyn Walker

One of the biggest challenges of end-to-end language generation from meaning representations in dialogue systems is making the outputs more natural and varied. Here we take a large corpus of 50K crowd-sourced utterances in the restaurant domain and develop text analysis methods that systematically characterize types of sentences in the training data. We then automatically label the training data to allow us to conduct two kinds of experiments with a neural generator. First, we test the effect of training the system with different stylistic partitions and quantify the effect of smaller, but more stylistically controlled training data. Second, we propose a method of labeling the style variants during training, and show that we can modify the style of the generated utterances using our stylistic labels. We contrast and compare these methods that can be used with any existing large corpus, showing how they vary in terms of semantic quality and stylistic control.

Lena Reed, Shereen Oraby, Marilyn Walker

Responses in task-oriented dialogue systems often realize multiple propositions whose ultimate form depends on the use of sentence planning and discourse structuring operations. For example a recommendation may consist of an explicitly evaluative utterance e.g. Chanpen Thai is the best option, along with content related by the justification discourse relation, e.g. It has great food and service, that combines multiple propositions into a single phrase. While neural generation methods integrate sentence planning and surface realization in one end-to-end learning framework, previous work has not shown that neural generators can: (1) perform common sentence planning and discourse structuring operations; (2) make decisions as to whether to realize content in a single sentence or over multiple sentences; (3) generalize sentence planning and discourse relation operations beyond what was seen in training. We systematically create large training corpora that exhibit particular sentence planning operations and then test neural models to see what they learn. We compare models without explicit latent variables for sentence planning with ones that provide explicit supervision during training. We show that only the models with additional supervision can reproduce sentence planing and discourse operations and generalize to situations unseen in training.

Vrindavan Harrison, Marilyn Walker

Question Generation is the task of automatically creating questions from textual input. In this work we present a new Attentional Encoder--Decoder Recurrent Neural Network model for automatic question generation. Our model incorporates linguistic features and an additional sentence embedding to capture meaning at both sentence and word levels. The linguistic features are designed to capture information related to named entity recognition, word case, and entity coreference resolution. In addition our model uses a copying mechanism and a special answer signal that enables generation of numerous diverse questions on a given sentence. Our model achieves state of the art results of 19.98 Bleu_4 on a benchmark Question Generation dataset, outperforming all previously published results by a significant margin. A human evaluation also shows that these added features improve the quality of the generated questions.

Shereen Oraby, Lena Reed, Sharath TS et al.

Natural language generators for task-oriented dialog should be able to vary the style of the output utterance while still effectively realizing the system dialog actions and their associated semantics. While the use of neural generation for training the response generation component of conversational agents promises to simplify the process of producing high quality responses in new domains, to our knowledge, there has been very little investigation of neural generators for task-oriented dialog that can vary their response style, and we know of no experiments on models that can generate responses that are different in style from those seen during training, while still maintain- ing semantic fidelity to the input meaning representation. Here, we show that a model that is trained to achieve a single stylis- tic personality target can produce outputs that combine stylistic targets. We carefully evaluate the multivoice outputs for both semantic fidelity and for similarities to and differences from the linguistic features that characterize the original training style. We show that contrary to our predictions, the learned models do not always simply interpolate model parameters, but rather produce styles that are distinct, and novel from the personalities they were trained on.

Shereen Oraby, Lena Reed, Shubhangi Tandon et al.

Natural language generators for task-oriented dialogue must effectively realize system dialogue actions and their associated semantics. In many applications, it is also desirable for generators to control the style of an utterance. To date, work on task-oriented neural generation has primarily focused on semantic fidelity rather than achieving stylistic goals, while work on style has been done in contexts where it is difficult to measure content preservation. Here we present three different sequence-to-sequence models and carefully test how well they disentangle content and style. We use a statistical generator, Personage, to synthesize a new corpus of over 88,000 restaurant domain utterances whose style varies according to models of personality, giving us total control over both the semantic content and the stylistic variation in the training data. We then vary the amount of explicit stylistic supervision given to the three models. We show that our most explicit model can simultaneously achieve high fidelity to both semantic and stylistic goals: this model adds a context vector of 36 stylistic parameters as input to the hidden state of the encoder at each time step, showing the benefits of explicit stylistic supervision, even when the amount of training data is large.

Kevin K. Bowden, Jiaqi Wu, Shereen Oraby et al.

In dialogue systems, the tasks of named entity recognition (NER) and named entity linking (NEL) are vital preprocessing steps for understanding user intent, especially in open domain interaction where we cannot rely on domain-specific inference. UCSC's effort as one of the funded teams in the 2017 Amazon Alexa Prize Contest has yielded Slugbot, an open domain social bot, aimed at casual conversation. We discovered several challenges specifically associated with both NER and NEL when building Slugbot, such as that the NE labels are too coarse-grained or the entity types are not linked to a useful ontology. Moreover, we have discovered that traditional approaches do not perform well in our context: even systems designed to operate on tweets or other social media data do not work well in dialogue systems. In this paper, we introduce Slugbot's Named Entity Recognition for dialogue Systems (SlugNERDS), a NER and NEL tool which is optimized to address these issues. We describe two new resources that we are building as part of this work: SlugEntityDB and SchemaActuator. We believe these resources will be useful for the research community.

Kevin K. Bowden, Jiaqi Wu, Shereen Oraby et al.

In this paper we introduce a novel, open domain socialbot for the Amazon Alexa Prize competition, aimed at carrying on friendly conversations with users on a variety of topics. We present our modular system, highlighting our different data sources and how we use the human mind as a model for data management. Additionally we build and employ natural language understanding and information retrieval tools and APIs to expand our knowledge bases. We describe our semistructured, scalable framework for crafting topic-specific dialogue flows, and give details on our dialogue management schemes and scoring mechanisms. Finally we briefly evaluate the performance of our system and observe the challenges that an open domain socialbot faces.

Amita Misra, Shereen Oraby, Shubhangi Tandon et al.

Online argumentative dialog is a rich source of information on popular beliefs and opinions that could be useful to companies as well as governmental or public policy agencies. Compact, easy to read, summaries of these dialogues would thus be highly valuable. A priori, it is not even clear what form such a summary should take. Previous work on summarization has primarily focused on summarizing written texts, where the notion of an abstract of the text is well defined. We collect gold standard training data consisting of five human summaries for each of 161 dialogues on the topics of Gay Marriage, Gun Control and Abortion. We present several different computational models aimed at identifying segments of the dialogues whose content should be used for the summary, using linguistic features and Word2vec features with both SVMs and Bidirectional LSTMs. We show that we can identify the most important arguments by using the dialog context with a best F-measure of 0.74 for gun control, 0.71 for gay marriage, and 0.67 for abortion.

Kevin K. Bowden, Shereen Oraby, Jiaqi Wu et al.

The greatest challenges in building sophisticated open-domain conversational agents arise directly from the potential for ongoing mixed-initiative multi-turn dialogues, which do not follow a particular plan or pursue a particular fixed information need. In order to make coherent conversational contributions in this context, a conversational agent must be able to track the types and attributes of the entities under discussion in the conversation and know how they are related. In some cases, the agent can rely on structured information sources to help identify the relevant semantic relations and produce a turn, but in other cases, the only content available comes from search, and it may be unclear which semantic relations hold between the search results and the discourse context. A further constraint is that the system must produce its contribution to the ongoing conversation in real-time. This paper describes our experience building SlugBot for the 2017 Alexa Prize, and discusses how we leveraged search and structured data from different sources to help SlugBot produce dialogic turns and carry on conversations whose length over the semi-finals user evaluation period averaged 8:17 minutes.

Shereen Oraby, Vrindavan Harrison, Lena Reed et al.

The use of irony and sarcasm in social media allows us to study them at scale for the first time. However, their diversity has made it difficult to construct a high-quality corpus of sarcasm in dialogue. Here, we describe the process of creating a large- scale, highly-diverse corpus of online debate forums dialogue, and our novel methods for operationalizing classes of sarcasm in the form of rhetorical questions and hyperbole. We show that we can use lexico-syntactic cues to reliably retrieve sarcastic utterances with high accuracy. To demonstrate the properties and quality of our corpus, we conduct supervised learning experiments with simple features, and show that we achieve both higher precision and F than previous work on sarcasm in debate forums dialogue. We apply a weakly-supervised linguistic pattern learner and qualitatively analyze the linguistic differences in each class.

Shereen Oraby, Sheideh Homayon, Marilyn Walker

Many of the creative and figurative elements that make language exciting are lost in translation in current natural language generation engines. In this paper, we explore a method to harvest templates from positive and negative reviews in the restaurant domain, with the goal of vastly expanding the types of stylistic variation available to the natural language generator. We learn hyperbolic adjective patterns that are representative of the strongly-valenced expressive language commonly used in either positive or negative reviews. We then identify and delexicalize entities, and use heuristics to extract generation templates from review sentences. We evaluate the learned templates against more traditional review templates, using subjective measures of "convincingness", "interestingness", and "naturalness". Our results show that the learned templates score highly on these measures. Finally, we analyze the linguistic categories that characterize the learned positive and negative templates. We plan to use the learned templates to improve the conversational style of dialogue systems in the restaurant domain.

Shereen Oraby, Vrindavan Harrison, Amita Misra et al.

Effective models of social dialog must understand a broad range of rhetorical and figurative devices. Rhetorical questions (RQs) are a type of figurative language whose aim is to achieve a pragmatic goal, such as structuring an argument, being persuasive, emphasizing a point, or being ironic. While there are computational models for other forms of figurative language, rhetorical questions have received little attention to date. We expand a small dataset from previous work, presenting a corpus of 10,270 RQs from debate forums and Twitter that represent different discourse functions. We show that we can clearly distinguish between RQs and sincere questions (0.76 F1). We then show that RQs can be used both sarcastically and non-sarcastically, observing that non-sarcastic (other) uses of RQs are frequently argumentative in forums, and persuasive in tweets. We present experiments to distinguish between these uses of RQs using SVM and LSTM models that represent linguistic features and post-level context, achieving results as high as 0.76 F1 for "sarcastic" and 0.77 F1 for "other" in forums, and 0.83 F1 for both "sarcastic" and "other" in tweets. We supplement our quantitative experiments with an in-depth characterization of the linguistic variation in RQs.

Shereen Oraby, Lena Reed, Ryan Compton et al.

We investigate the characteristics of factual and emotional argumentation styles observed in online debates. Using an annotated set of "factual" and "feeling" debate forum posts, we extract patterns that are highly correlated with factual and emotional arguments, and then apply a bootstrapping methodology to find new patterns in a larger pool of unannotated forum posts. This process automatically produces a large set of patterns representing linguistic expressions that are highly correlated with factual and emotional language. Finally, we analyze the most discriminating patterns to better understand the defining characteristics of factual and emotional arguments.

Geetanjali Rakshit, Kevin K. Bowden, Lena Reed et al.

Chatbots are a rapidly expanding application of dialogue systems with companies switching to bot services for customer support, and new applications for users interested in casual conversation. One style of casual conversation is argument, many people love nothing more than a good argument. Moreover, there are a number of existing corpora of argumentative dialogues, annotated for agreement and disagreement, stance, sarcasm and argument quality. This paper introduces Debbie, a novel arguing bot, that selects arguments from conversational corpora, and aims to use them appropriately in context. We present an initial working prototype of Debbie, with some preliminary evaluation and describe future work.

Amita Misra, Brian Ecker, Theodore Handleman et al.

Stance classification aims to identify, for a particular issue under discussion, whether the speaker or author of a conversational turn has Pro (Favor) or Con (Against) stance on the issue. Detecting stance in tweets is a new task proposed for SemEval-2016 Task6, involving predicting stance for a dataset of tweets on the topics of abortion, atheism, climate change, feminism and Hillary Clinton. Given the small size of the dataset, our team created our own topic-specific training corpus by developing a set of high precision hashtags for each topic that were used to query the twitter API, with the aim of developing a large training corpus without additional human labeling of tweets for stance. The hashtags selected for each topic were predicted to be stance-bearing on their own. Experimental results demonstrate good performance for our features for opinion-target pairs based on generalizing dependency features using sentiment lexicons.

Amita Misra, Pranav Anand, Jean E Fox Tree et al.

More and more of the information available on the web is dialogic, and a significant portion of it takes place in online forum conversations about current social and political topics. We aim to develop tools to summarize what these conversations are about. What are the CENTRAL PROPOSITIONS associated with different stances on an issue, what are the abstract objects under discussion that are central to a speaker's argument? How can we recognize that two CENTRAL PROPOSITIONS realize the same FACET of the argument? We hypothesize that the CENTRAL PROPOSITIONS are exactly those arguments that people find most salient, and use human summarization as a probe for discovering them. We describe our corpus of human summaries of opinionated dialogs, then show how we can identify similar repeated arguments, and group them into FACETS across many discussions of a topic. We define a new task, ARGUMENT FACET SIMILARITY (AFS), and show that we can predict AFS with a .54 correlation score, versus an ngram system baseline of .39 and a semantic textual similarity system baseline of .45.

Amita Misra, Marilyn Walker

Research on the structure of dialogue has been hampered for years because large dialogue corpora have not been available. This has impacted the dialogue research community's ability to develop better theories, as well as good off the shelf tools for dialogue processing. Happily, an increasing amount of information and opinion exchange occur in natural dialogue in online forums, where people share their opinions about a vast range of topics. In particular we are interested in rejection in dialogue, also called disagreement and denial, where the size of available dialogue corpora, for the first time, offers an opportunity to empirically test theoretical accounts of the expression and inference of rejection in dialogue. In this paper, we test whether topic-independent features motivated by theoretical predictions can be used to recognize rejection in online forums in a topic independent way. Our results show that our theoretically motivated features achieve 66% accuracy, an improvement over a unigram baseline of an absolute 6%.

Jiaqi Wu, Marilyn Walker, Pranav Anand et al.

Different theories posit different sources for feelings of well-being and happiness. Appraisal theory grounds our emotional responses in our goals and desires and their fulfillment, or lack of fulfillment. Self Determination theory posits that the basis for well-being rests on our assessment of our competence, autonomy, and social connection. And surveys that measure happiness empirically note that people require their basic needs to be met for food and shelter, but beyond that tend to be happiest when socializing, eating or having sex. We analyze a corpus of private microblogs from a well-being application called ECHO, where users label each written post about daily events with a happiness score between 1 and 9. Our goal is to ground the linguistic descriptions of events that users experience in theories of well-being and happiness, and then examine the extent to which different theoretical accounts can explain the variance in the happiness scores. We show that recurrent event types, such as OBLIGATION and INCOMPETENCE, which affect people's feelings of well-being are not captured in current lexical or semantic resources.

Lena Reed, Jiaqi Wu, Shereen Oraby et al.

Informal first-person narratives are a unique resource for computational models of everyday events and people's affective reactions to them. People blogging about their day tend not to explicitly say I am happy. Instead they describe situations from which other humans can readily infer their affective reactions. However current sentiment dictionaries are missing much of the information needed to make similar inferences. We build on recent work that models affect in terms of lexical predicate functions and affect on the predicate's arguments. We present a method to learn proxies for these functions from first-person narratives. We construct a novel fine-grained test set, and show that the patterns we learn improve our ability to predict first-person affective reactions to everyday events, from a Stanford sentiment baseline of .67F to .75F.

Stephanie M. Lukin, Pranav Anand, Marilyn Walker et al.

Americans spend about a third of their time online, with many participating in online conversations on social and political issues. We hypothesize that social media arguments on such issues may be more engaging and persuasive than traditional media summaries, and that particular types of people may be more or less convinced by particular styles of argument, e.g. emotional arguments may resonate with some personalities while factual arguments resonate with others. We report a set of experiments testing at large scale how audience variables interact with argument style to affect the persuasiveness of an argument, an under-researched topic within natural language processing. We show that belief change is affected by personality factors, with conscientious, open and agreeable people being more convinced by emotional arguments.

Stephanie Lukin, Marilyn Walker

More and more of the information on the web is dialogic, from Facebook newsfeeds, to forum conversations, to comment threads on news articles. In contrast to traditional, monologic Natural Language Processing resources such as news, highly social dialogue is frequent in social media, making it a challenging context for NLP. This paper tests a bootstrapping method, originally proposed in a monologic domain, to train classifiers to identify two different types of subjective language in dialogue: sarcasm and nastiness. We explore two methods of developing linguistic indicators to be used in a first level classifier aimed at maximizing precision at the expense of recall. The best performing classifier for the first phase achieves 54% precision and 38% recall for sarcastic utterances. We then use general syntactic patterns from previous work to create more general sarcasm indicators, improving precision to 62% and recall to 52%. To further test the generality of the method, we then apply it to bootstrapping a classifier for nastiness dialogic acts. Our first phase, using crowdsourced nasty indicators, achieves 58% precision and 49% recall, which increases to 75% precision and 62% recall when we bootstrap over the first level with generalized syntactic patterns.