Shikib Mehri, Jinho Choi, Luis Fernando D'Haro et al.
This is a report on the NSF Future Directions Workshop on Automatic Evaluation of Dialog. The workshop explored the current state of the art along with its limitations and suggested promising directions for future work in this important and very rapidly changing area of research.
Erfan Al-Hossami, Samira Shaikh
In this survey paper, we overview major deep learning methods used in Natural Language Processing (NLP) and source code over the last 35 years. Next, we present a survey of the applications of Artificial Intelligence (AI) for source code, also known as Code Intelligence (CI) and Programming Language Processing (PLP). We survey over 287 publications and present a software-engineering centered taxonomy for CI placing each of the works into one category describing how it best assists the software development cycle. Then, we overview the field of conversational assistants and their applications in software engineering and education. Lastly, we highlight research opportunities at the intersection of AI for code and conversational assistants and provide future directions for researching conversational assistants with CI capabilities.
Pietro Liguori, Erfan Al-Hossami, Domenico Cotroneo et al.
Writing software exploits is an important practice for offensive security analysts to investigate and prevent attacks. In particular, shellcodes are especially time-consuming and a technical challenge, as they are written in assembly language. In this work, we address the task of automatically generating shellcodes, starting purely from descriptions in natural language, by proposing an approach based on Neural Machine Translation (NMT). We then present an empirical study using a novel dataset (Shellcode_IA32), which consists of 3,200 assembly code snippets of real Linux/x86 shellcodes from public databases, annotated using natural language. Moreover, we propose novel metrics to evaluate the accuracy of NMT at generating shellcodes. The empirical analysis shows that NMT can generate assembly code snippets from the natural language with high accuracy and that in many cases can generate entire shellcodes with no errors.
Pietro Liguori, Erfan Al-Hossami, Vittorio Orbinato et al.
Writing exploits for security assessment is a challenging task. The writer needs to master programming and obfuscation techniques to develop a successful exploit. To make the task easier, we propose an approach (EVIL) to automatically generate exploits in assembly/Python language from descriptions in natural language. The approach leverages Neural Machine Translation (NMT) techniques and a dataset that we developed for this work. We present an extensive experimental study to evaluate the feasibility of EVIL, using both automatic and manual analysis, and both at generating individual statements and entire exploits. The generated code achieved high accuracy in terms of syntactic and semantic correctness.
Pietro Liguori, Erfan Al-Hossami, Domenico Cotroneo et al.
We take the first step to address the task of automatically generating shellcodes, i.e., small pieces of code used as a payload in the exploitation of a software vulnerability, starting from natural language comments. We assemble and release a novel dataset (Shellcode_IA32), consisting of challenging but common assembly instructions with their natural language descriptions. We experiment with standard methods in neural machine translation (NMT) to establish baseline performance levels on this task.
Sebastian Gehrmann, Tosin Adewumi, Karmanya Aggarwal et al.
We introduce GEM, a living benchmark for natural language Generation (NLG), its Evaluation, and Metrics. Measuring progress in NLG relies on a constantly evolving ecosystem of automated metrics, datasets, and human evaluation standards. Due to this moving target, new models often still evaluate on divergent anglo-centric corpora with well-established, but flawed, metrics. This disconnect makes it challenging to identify the limitations of current models and opportunities for progress. Addressing this limitation, GEM provides an environment in which models can easily be applied to a wide set of tasks and in which evaluation strategies can be tested. Regular updates to the benchmark will help NLG research become more multilingual and evolve the challenge alongside models. This paper serves as the description of the data for which we are organizing a shared task at our ACL 2021 Workshop and to which we invite the entire NLG community to participate.
Sashank Santhanam, Zhuo Cheng, Brodie Mather et al.
Achieving true human-like ability to conduct a conversation remains an elusive goal for open-ended dialogue systems. We posit this is because extant approaches towards natural language generation (NLG) are typically construed as end-to-end architectures that do not adequately model human generation processes. To investigate, we decouple generation into two separate phases: planning and realization. In the planning phase, we train two planners to generate plans for response utterances. The realization phase uses response plans to produce an appropriate response. Through rigorous evaluations, both automated and human, we demonstrate that decoupling the process into planning and realization performs better than an end-to-end approach.
Adam Dalton, Ehsan Aghaei, Ehab Al-Shaer et al.
We describe Panacea, a system that supports natural language processing (NLP) components for active defenses against social engineering attacks. We deploy a pipeline of human language technology, including Ask and Framing Detection, Named Entity Recognition, Dialogue Engineering, and Stylometry. Panacea processes modern message formats through a plug-in architecture to accommodate innovative approaches for message analysis, knowledge representation and dialogue generation. The novelty of the Panacea system is that uses NLP for cyber defense and engages the attacker using bots to elicit evidence to attribute to the attacker and to waste the attacker's time and resources.
Archna Bhatia, Adam Dalton, Brodie Mather et al.
We present a paradigm for extensible lexicon development based on Lexical Conceptual Structure to support social engineering detection and response generation. We leverage the central notions of ask (elicitation of behaviors such as providing access to money) and framing (risk/reward implied by the ask). We demonstrate improvements in ask/framing detection through refinements to our lexical organization and show that response generation qualitatively improves as ask/framing detection performance improves. The paradigm presents a systematic and efficient approach to resource adaptation for improved task-specific performance.
Bonnie J. Dorr, Archna Bhatia, Adam Dalton et al.
Social engineers attempt to manipulate users into undertaking actions such as downloading malware by clicking links or providing access to money or sensitive information. Natural language processing, computational sociolinguistics, and media-specific structural clues provide a means for detecting both the ask (e.g., buy gift card) and the risk/reward implied by the ask, which we call framing (e.g., lose your job, get a raise). We apply linguistic resources such as Lexical Conceptual Structure to tackle ask detection and also leverage structural clues such as links and their proximity to identified asks to improve confidence in our results. Our experiments indicate that the performance of ask detection, framing detection, and identification of the top ask is improved by linguistically motivated classes coupled with structural clues such as links. Our approach is implemented in a system that informs users about social engineering risk situations.
Sashank Santhanam, Alireza Karduni, Samira Shaikh
Humans quite frequently interact with conversational agents. The rapid advancement in generative language modeling through neural networks has helped advance the creation of intelligent conversational agents. Researchers typically evaluate the output of their models through crowdsourced judgments, but there are no established best practices for conducting such studies. Moreover, it is unclear if cognitive biases in decision-making are affecting crowdsourced workers' judgments when they undertake these tasks. To investigate, we conducted a between-subjects study with 77 crowdsourced workers to understand the role of cognitive biases, specifically anchoring bias, when humans are asked to evaluate the output of conversational agents. Our results provide insight into how best to evaluate conversational agents. We find increased consistency in ratings across two experimental conditions may be a result of anchoring bias. We also determine that external factors such as time and prior experience in similar tasks have effects on inter-rater consistency.
Vidhushini Srinivasan, Sashank Santhanam, Samira Shaikh
We propose an approach towards natural language generation using a bidirectional encoder-decoder which incorporates external rewards through reinforcement learning (RL). We use attention mechanism and maximum mutual information as an initial objective function using RL. Using a two-part training scheme, we train an external reward analyzer to predict the external rewards and then use the predicted rewards to maximize the expected rewards (both internal and external). We evaluate the system on two standard dialogue corpora - Cornell Movie Dialog Corpus and Yelp Restaurant Review Corpus. We report standard evaluation metrics including BLEU, ROUGE-L, and perplexity as well as human evaluation to validate our approach.
Sashank Santhanam, Samira Shaikh
Emotional language generation is one of the keys to human-like artificial intelligence. Humans use different type of emotions depending on the situation of the conversation. Emotions also play an important role in mediating the engagement level with conversational partners. However, current conversational agents do not effectively account for emotional content in the language generation process. To address this problem, we develop a language modeling approach that generates affective content when the dialogue is situated in a given context. We use the recently released Empathetic-Dialogues corpus to build our models. Through detailed experiments, we find that our approach outperforms the state-of-the-art method on the perplexity metric by about 5 points and achieves a higher BLEU metric score.
Sashank Santhanam, Samira Shaikh
To overcome the limitations of automated metrics (e.g. BLEU, METEOR) for evaluating dialogue systems, researchers typically use human judgments to provide convergent evidence. While it has been demonstrated that human judgments can suffer from the inconsistency of ratings, extant research has also found that the design of the evaluation task affects the consistency and quality of human judgments. We conduct a between-subjects study to understand the impact of four experiment conditions on human ratings of dialogue system output. In addition to discrete and continuous scale ratings, we also experiment with a novel application of Best-Worst scaling to dialogue evaluation. Through our systematic study with 40 crowdsourced workers in each task, we find that using continuous scales achieves more consistent ratings than Likert scale or ranking-based experiment design. Additionally, we find that factors such as time taken to complete the task and no prior experience of participating in similar studies of rating dialogue system output positively impact consistency and agreement amongst raters
Sashank Santhanam, Vidhushini Srinivasan, Shaina Glass et al.
We study how emojis are used to express solidarity in social media in the context of two major crisis events - a natural disaster, Hurricane Irma in 2017 and terrorist attacks that occurred on November 2015 in Paris. Using annotated corpora, we first train a recurrent neural network model to classify expressions of solidarity in text. Next, we use these expressions of solidarity to characterize human behavior in online social networks, through the temporal and geospatial diffusion of emojis. Our analysis reveals that emojis are a powerful indicator of sociolinguistic behaviors (solidarity) that are exhibited on social media as the crisis events unfold.
Sashank Santhanam, Samira Shaikh
One of the hardest problems in the area of Natural Language Processing and Artificial Intelligence is automatically generating language that is coherent and understandable to humans. Teaching machines how to converse as humans do falls under the broad umbrella of Natural Language Generation. Recent years have seen unprecedented growth in the number of research articles published on this subject in conferences and journals both by academic and industry researchers. There have also been several workshops organized alongside top-tier NLP conferences dedicated specifically to this problem. All this activity makes it hard to clearly define the state of the field and reason about its future directions. In this work, we provide an overview of this important and thriving area, covering traditional approaches, statistical approaches and also approaches that use deep neural networks. We provide a comprehensive review towards building open domain dialogue systems, an important application of natural language generation. We find that, predominantly, the approaches for building dialogue systems use seq2seq or language models architecture. Notably, we identify three important areas of further research towards building more effective dialogue systems: 1) incorporating larger context, including conversation context and world knowledge; 2) adding personae or personality in the NLG system; and 3) overcoming dull and generic responses that affect the quality of system-produced responses. We provide pointers on how to tackle these open problems through the use of cognitive architectures that mimic human language understanding and generation capabilities.
Alireza Karduni, Isaac Cho, Ryan Wesslen et al.
We present Verifi2, a visual analytic system to support the investigation of misinformation on social media. On the one hand, social media platforms empower individuals and organizations by democratizing the sharing of information. On the other hand, even well-informed and experienced social media users are vulnerable to misinformation. To address the issue, various models and studies have emerged from multiple disciplines to detect and understand the effects of misinformation. However, there is still a lack of intuitive and accessible tools that help social media users distinguish misinformation from verified news. In this paper, we present Verifi2, a visual analytic system that uses state-of-the-art computational methods to highlight salient features from text, social network, and images. By exploring news on a source level through multiple coordinated views in Verifi2, users can interact with the complex dimensions that characterize misinformation and contrast how real and suspicious news outlets differ on these dimensions. To evaluate Verifi2, we conduct interviews with experts in digital media, journalism, education, psychology, and computing who study misinformation. Our interviews show promising potential for Verifi2 to serve as an educational tool on misinformation. Furthermore, our interview results highlight the complexity of the problem of combating misinformation and call for more work from the visualization community.
Ryan Wesslen, Sashank Santhanam, Alireza Karduni et al.
Cognitive biases have been shown to lead to faulty decision-making. Recent research has demonstrated that the effect of cognitive biases, anchoring bias in particular, transfers to information visualization and visual analytics. However, it is still unclear how users of visual interfaces can be anchored and the impact of anchoring on user performance and decision-making process. To investigate, we performed two rounds of between-subjects, in-laboratory experiments with 94 participants to analyze the effect of visual anchors and strategy cues in decision-making with a visual analytic system that employs coordinated multiple view design. The decision-making task is identifying misinformation from Twitter news accounts. Participants were randomly assigned one of three treatment groups (including control) in which participant training processes were modified. Our findings reveal that strategy cues and visual anchors (scenario videos) can significantly affect user activity, speed, confidence, and, under certain circumstances, accuracy. We discuss the implications of our experiment results on training users how to use a newly developed visual interface. We call for more careful consideration into how visualization designers and researchers train users to avoid unintentionally anchoring users and thus affecting the end result.