Kartik Talamadupula

Zijun Cui, Tian Gao, Kartik Talamadupula et al. · ibm-research

Deep learning models, though having achieved great success in many different fields over the past years, are usually data hungry, fail to perform well on unseen samples, and lack of interpretability. Various prior knowledge often exists in the target domain and their use can alleviate the deficiencies with deep learning. To better mimic the behavior of human brains, different advanced methods have been proposed to identify domain knowledge and integrate it into deep models for data-efficient, generalizable, and interpretable deep learning, which we refer to as knowledge-augmented deep learning (KADL). In this survey, we define the concept of KADL, and introduce its three major tasks, i.e., knowledge identification, knowledge representation, and knowledge integration. Different from existing surveys that are focused on a specific type of knowledge, we provide a broad and complete taxonomy of domain knowledge and its representations. Based on our taxonomy, we provide a systematic review of existing techniques, different from existing works that survey integration approaches agnostic to taxonomy of knowledge. This survey subsumes existing works and offers a bird's-eye view of research in the general area of knowledge-augmented deep learning. The thorough and critical reviews of numerous papers help not only understand current progresses but also identify future directions for the research on knowledge-augmented deep learning.

Benjamin Reichman, Adar Avsian, Kartik Talamadupula et al. · ibm-research

Queries to large language models (LLMs) can be divided into two parts: the instruction/question and the accompanying context. The context for retrieval-augmented generation (RAG) systems in most benchmarks comes from Wikipedia-like texts written in a neutral and factual tone. However, real-world RAG applications often retrieve internet-based text with diverse tones and linguistic styles, posing challenges for downstream tasks. This paper introduces (a) a dataset that transforms RAG-retrieved passages into emotionally inflected and sarcastic text, (b) an emotion translation model for adapting text to different tones, and (c) a prompt-based method to improve LLMs' pragmatic interpretation of retrieved text.

Mehdi Fatemi, Banafsheh Rafiee, Mingjie Tang et al. · ibm-research

Despite significant advancements in large language models (LLMs), a major drawback of reasoning models is their enormous token usage, which increases computational cost, resource requirements, and response time. In this work, we revisit the core principles of reinforcement learning (RL) and, through mathematical analysis, demonstrate that the tendency to generate lengthy responses arises inherently from RL-based optimization during training. This finding questions the prevailing assumption that longer responses inherently improve reasoning accuracy. Instead, we uncover a natural correlation between conciseness and accuracy that has been largely overlooked. We show that introducing a secondary phase of RL training, using a very small set of problems, can significantly reduce chains of thought while maintaining or even enhancing accuracy. Additionally, we demonstrate that, while GRPO shares some interesting properties of PPO, it suffers from collapse modes, which limit its reliability for concise reasoning. Finally, we validate our conclusions through extensive experimental results.

Kinjal Basu, Keerthiram Murugesan, Subhajit Chaudhury et al. · ibm-research

Text-based games (TBGs) have emerged as an important collection of NLP tasks, requiring reinforcement learning (RL) agents to combine natural language understanding with reasoning. A key challenge for agents attempting to solve such tasks is to generalize across multiple games and demonstrate good performance on both seen and unseen objects. Purely deep-RL-based approaches may perform well on seen objects; however, they fail to showcase the same performance on unseen objects. Commonsense-infused deep-RL agents may work better on unseen data; unfortunately, their policies are often not interpretable or easily transferable. To tackle these issues, in this paper, we present EXPLORER which is an exploration-guided reasoning agent for textual reinforcement learning. EXPLORER is neurosymbolic in nature, as it relies on a neural module for exploration and a symbolic module for exploitation. It can also learn generalized symbolic policies and perform well over unseen data. Our experiments show that EXPLORER outperforms the baseline agents on Text-World cooking (TW-Cooking) and Text-World Commonsense (TWC) games.

Toshish Jawale, Chaitanya Animesh, Sekhar Vallath et al. · ibm-research

This study analyzes changes in the attention mechanisms of large language models (LLMs) when used to understand natural conversations between humans (human-human). We analyze three use cases of LLMs: interactions over web content, code, and mathematical texts. By analyzing attention distance, dispersion, and interdependency across these domains, we highlight the unique challenges posed by conversational data. Notably, conversations require nuanced handling of long-term contextual relationships and exhibit higher complexity through their attention patterns. Our findings reveal that while language models exhibit domain-specific attention behaviors, there is a significant gap in their ability to specialize in human conversations. Through detailed attention entropy analysis and t-SNE visualizations, we demonstrate the need for models trained with a diverse array of high-quality conversational data to enhance understanding and generation of human-like dialogue. This research highlights the importance of domain specialization in language models and suggests pathways for future advancement in modeling human conversational nuances.

Soufiane Amini, Yassine Benajiba, Cesare Bernardis et al. · ibm-research

The proliferation of agent frameworks has led to fragmentation in how agents are defined, executed, and evaluated. Existing systems differ in their abstractions, data flow semantics, and tool integrations, making it difficult to share or reproduce workflows. We introduce Open Agent Specification (Agent Spec), a declarative language that defines AI agents and agentic workflows in a way that is compatible across frameworks, promoting reusability, portability and interoperability of AI agents. Agent Spec defines a common set of components, control and data flow semantics, and schemas that allow an agent to be defined once and executed across different runtimes. Agent Spec also introduces a standardized Evaluation harness to assess agent behavior and agentic workflows across runtimes - analogous to how HELM and related harnesses standardized LLM evaluation - so that performance, robustness, and efficiency can be compared consistently across frameworks. We demonstrate this using four distinct runtimes (LangGraph, CrewAI, AutoGen, and WayFlow) evaluated over three different benchmarks (SimpleQA Verified, $τ^2$-Bench and BIRD-SQL). We provide accompanying toolsets: a Python SDK (PyAgentSpec), a reference runtime (WayFlow), and adapters for popular frameworks (e.g., LangGraph, AutoGen, CrewAI). Agent Spec bridges the gap between model-centric and agent-centric standardization & evaluation, laying the groundwork for reliable, reusable, and portable agentic systems.

Benjamin Reichman, Kartik Talamadupula · ibm-research

Human communication is a multifaceted and multimodal skill. Communication requires an understanding of both the surface-level textual content and the connotative intent of a piece of communication. In humans, learning to go beyond the surface level starts by learning communicative intent in speech. Once humans acquire these skills in spoken communication, they transfer those skills to written communication. In this paper, we assess the ability of speech+text models and text models trained with special emphasis on human-to-human conversations to make this multimodal transfer of skill. We specifically test these models on their ability to detect covert deceptive communication. We find that with no special prompting speech+text LLMs have an advantage over unimodal LLMs in performing this task. Likewise, we find that human-to-human conversation-trained LLMs are also advantaged in this skill.

Justin D. Weisz, Michael Muller, Steven I. Ross et al.

Generative machine learning models have recently been applied to source code, for use cases including translating code between programming languages, creating documentation from code, and auto-completing methods. Yet, state-of-the-art models often produce code that is erroneous or incomplete. In a controlled study with 32 software engineers, we examined whether such imperfect outputs are helpful in the context of Java-to-Python code translation. When aided by the outputs of a code translation model, participants produced code with fewer errors than when working alone. We also examined how the quality and quantity of AI translations affected the work process and quality of outcomes, and observed that providing multiple translations had a larger impact on the translation process than varying the quality of provided translations. Our results tell a complex, nuanced story about the benefits of generative code models and the challenges software engineers face when working with their outputs. Our work motivates the need for intelligent user interfaces that help software engineers effectively work with generative code models in order to understand and evaluate their outputs and achieve superior outcomes to working alone.

Jiao Sun, Q. Vera Liao, Michael Muller et al.

What does it mean for a generative AI model to be explainable? The emergent discipline of explainable AI (XAI) has made great strides in helping people understand discriminative models. Less attention has been paid to generative models that produce artifacts, rather than decisions, as output. Meanwhile, generative AI (GenAI) technologies are maturing and being applied to application domains such as software engineering. Using scenario-based design and question-driven XAI design approaches, we explore users' explainability needs for GenAI in three software engineering use cases: natural language to code, code translation, and code auto-completion. We conducted 9 workshops with 43 software engineers in which real examples from state-of-the-art generative AI models were used to elicit users' explainability needs. Drawing from prior work, we also propose 4 types of XAI features for GenAI for code and gathered additional design ideas from participants. Our work explores explainability needs for GenAI for code and demonstrates how human-centered approaches can drive the technical development of XAI in novel domains.

Edmond Awad, Sydney Levine, Andrea Loreggia et al.

One of the most remarkable things about the human moral mind is its flexibility. We can make moral judgments about cases we have never seen before. We can decide that pre-established rules should be broken. We can invent novel rules on the fly. Capturing this flexibility is one of the central challenges in developing AI systems that can interpret and produce human-like moral judgment. This paper details the results of a study of real-world decision makers who judge whether it is acceptable to break a well-established norm: ``no cutting in line.'' We gather data on how human participants judge the acceptability of line-cutting in a range of scenarios. Then, in order to effectively embed these reasoning capabilities into a machine, we propose a method for modeling them using a preference-based structure, which captures a novel modification to standard ``dual process'' theories of moral judgment.

Mayank Agarwal, Kartik Talamadupula, Fernando Martinez et al.

Translating source code from one programming language to another is a critical, time-consuming task in modernizing legacy applications and codebases. Recent work in this space has drawn inspiration from the software naturalness hypothesis by applying natural language processing techniques towards automating the code translation task. However, due to the paucity of parallel data in this domain, supervised techniques have only been applied to a limited set of popular programming languages. To bypass this limitation, unsupervised neural machine translation techniques have been proposed to learn code translation using only monolingual corpora. In this work, we propose to use document similarity methods to create noisy parallel datasets of code, thus enabling supervised techniques to be applied for automated code translation without having to rely on the availability or expensive curation of parallel code datasets. We explore the noise tolerance of models trained on such automatically-created datasets and show that these models perform comparably to models trained on ground truth for reasonable levels of noise. Finally, we exhibit the practical utility of the proposed method by creating parallel datasets for languages beyond the ones explored in prior work, thus expanding the set of programming languages for automated code translation.

Keerthiram Murugesan, Subhajit Chaudhury, Kartik Talamadupula

Text-based games (TBGs) have become a popular proving ground for the demonstration of learning-based agents that make decisions in quasi real-world settings. The crux of the problem for a reinforcement learning agent in such TBGs is identifying the objects in the world, and those objects' relations with that world. While the recent use of text-based resources for increasing an agent's knowledge and improving its generalization have shown promise, we posit in this paper that there is much yet to be learned from visual representations of these same worlds. Specifically, we propose to retrieve images that represent specific instances of text observations from the world and train our agents on such images. This improves the agent's overall understanding of the game 'scene' and objects' relationships to the world around them, and the variety of visual representations on offer allow the agent to generate a better generalization of a relationship. We show that incorporating such images improves the performance of agents in various TBG settings.

Justin D. Weisz, Michael Muller, Stephanie Houde et al.

Generative models have become adept at producing artifacts such as images, videos, and prose at human-like levels of proficiency. New generative techniques, such as unsupervised neural machine translation (NMT), have recently been applied to the task of generating source code, translating it from one programming language to another. The artifacts produced in this way may contain imperfections, such as compilation or logical errors. We examine the extent to which software engineers would tolerate such imperfections and explore ways to aid the detection and correction of those errors. Using a design scenario approach, we interviewed 11 software engineers to understand their reactions to the use of an NMT model in the context of application modernization, focusing on the task of translating source code from one language to another. Our three-stage scenario sparked discussions about the utility and desirability of working with an imperfect AI system, how acceptance of that system's outputs would be established, and future opportunities for generative AI in application modernization. Our study highlights how UI features such as confidence highlighting and alternate translations help software engineers work with and better understand generative NMT models.

Mayank Agarwal, Tathagata Chakraborti, Quchen Fu et al.

The NLC2CMD Competition hosted at NeurIPS 2020 aimed to bring the power of natural language processing to the command line. Participants were tasked with building models that can transform descriptions of command line tasks in English to their Bash syntax. This is a report on the competition with details of the task, metrics, data, attempted solutions, and lessons learned.

Mayank Agarwal, Kartik Talamadupula, Stephanie Houde et al.

Recently, the automated translation of source code from one programming language to another by using automatic approaches inspired by Neural Machine Translation (NMT) methods for natural languages has come under study. However, such approaches suffer from the same problem as previous NMT approaches on natural languages, viz. the lack of an ability to estimate and evaluate the quality of the translations; and consequently ascribe some measure of interpretability to the model's choices. In this paper, we attempt to estimate the quality of source code translations built on top of the TransCoder model. We consider the code translation task as an analog of machine translation (MT) for natural languages, with some added caveats. We present our main motivation from a user study built around code translation; and present a technique that correlates the confidences generated by that model to lint errors in the translated code. We conclude with some observations on these correlations, and some ideas for future work.

Thomas Carta, Subhajit Chaudhury, Kartik Talamadupula et al.

We present VisualHints, a novel environment for multimodal reinforcement learning (RL) involving text-based interactions along with visual hints (obtained from the environment). Real-life problems often demand that agents interact with the environment using both natural language information and visual perception towards solving a goal. However, most traditional RL environments either solve pure vision-based tasks like Atari games or video-based robotic manipulation; or entirely use natural language as a mode of interaction, like Text-based games and dialog systems. In this work, we aim to bridge this gap and unify these two approaches in a single environment for multimodal RL. We introduce an extension of the TextWorld cooking environment with the addition of visual clues interspersed throughout the environment. The goal is to force an RL agent to use both text and visual features to predict natural language action commands for solving the final task of cooking a meal. We enable variations and difficulties in our environment to emulate various interactive real-world scenarios. We present a baseline multimodal agent for solving such problems using CNN-based feature extraction from visual hints and LSTMs for textual feature extraction. We believe that our proposed visual-lingual environment will facilitate novel problem settings for the RL community.

Keerthiram Murugesan, Mattia Atzeni, Pavan Kapanipathi et al.

Text-based games have emerged as an important test-bed for Reinforcement Learning (RL) research, requiring RL agents to combine grounded language understanding with sequential decision making. In this paper, we examine the problem of infusing RL agents with commonsense knowledge. Such knowledge would allow agents to efficiently act in the world by pruning out implausible actions, and to perform look-ahead planning to determine how current actions might affect future world states. We design a new text-based gaming environment called TextWorld Commonsense (TWC) for training and evaluating RL agents with a specific kind of commonsense knowledge about objects, their attributes, and affordances. We also introduce several baseline RL agents which track the sequential context and dynamically retrieve the relevant commonsense knowledge from ConceptNet. We show that agents which incorporate commonsense knowledge in TWC perform better, while acting more efficiently. We conduct user-studies to estimate human performance on TWC and show that there is ample room for future improvement.

Anshuman Mishra, Dhruvesh Patel, Aparna Vijayakumar et al.

In the recent past, Natural language Inference (NLI) has gained significant attention, particularly given its promise for downstream NLP tasks. However, its true impact is limited and has not been well studied. Therefore, in this paper, we explore the utility of NLI for one of the most prominent downstream tasks, viz. Question Answering (QA). We transform the one of the largest available MRC dataset (RACE) to an NLI form, and compare the performances of a state-of-the-art model (RoBERTa) on both these forms. We propose new characterizations of questions, and evaluate the performance of QA and NLI models on these categories. We highlight clear categories for which the model is able to perform better when the data is presented in a coherent entailment form, and a structured question-answer concatenation form, respectively.

Subhajit Chaudhury, Daiki Kimura, Kartik Talamadupula et al.

We show that Reinforcement Learning (RL) methods for solving Text-Based Games (TBGs) often fail to generalize on unseen games, especially in small data regimes. To address this issue, we propose Context Relevant Episodic State Truncation (CREST) for irrelevant token removal in observation text for improved generalization. Our method first trains a base model using Q-learning, which typically overfits the training games. The base model's action token distribution is used to perform observation pruning that removes irrelevant tokens. A second bootstrapped model is then retrained on the pruned observation text. Our bootstrapped agent shows improved generalization in solving unseen TextWorld games, using 10x-20x fewer training games compared to previous state-of-the-art methods despite requiring less number of training episodes.

Anshuman Mishra, Dhruvesh Patel, Aparna Vijayakumar et al.

In recent years, the Natural Language Inference (NLI) task has garnered significant attention, with new datasets and models achieving near human-level performance on it. However, the full promise of NLI -- particularly that it learns knowledge that should be generalizable to other downstream NLP tasks -- has not been realized. In this paper, we study this unfulfilled promise from the lens of two downstream tasks: question answering (QA), and text summarization. We conjecture that a key difference between the NLI datasets and these downstream tasks concerns the length of the premise; and that creating new long premise NLI datasets out of existing QA datasets is a promising avenue for training a truly generalizable NLI model. We validate our conjecture by showing competitive results on the task of QA and obtaining the best reported results on the task of Checking Factual Correctness of Summaries.

Zijun Cui, Pavan Kapanipathi, Kartik Talamadupula et al.

Knowledge graphs (KGs) are of great importance to many real world applications, but they generally suffer from incomplete information in the form of missing relations between entities. Knowledge graph completion (also known as relation prediction) is the task of inferring missing facts given existing ones. Most of the existing work is proposed by maximizing the likelihood of observed instance-level triples. Not much attention, however, is paid to the ontological information, such as type information of entities and relations. In this work, we propose a type-augmented relation prediction (TaRP) method, where we apply both the type information and instance-level information for relation prediction. In particular, type information and instance-level information are encoded as prior probabilities and likelihoods of relations respectively, and are combined by following Bayes' rule. Our proposed TaRP method achieves significantly better performance than state-of-the-art methods on four benchmark datasets: FB15K, FB15K-237, YAGO26K-906, and DB111K-174. In addition, we show that TaRP achieves significantly improved data efficiency. More importantly, the type information extracted from a specific dataset can generalize well to other datasets through the proposed TaRP model.

Anurag Acharya, Kartik Talamadupula, Mark A Finlayson

Existing commonsense reasoning datasets for AI and NLP tasks fail to address an important aspect of human life: cultural differences. We introduce an approach that extends prior work on crowdsourcing commonsense knowledge by incorporating differences in knowledge that are attributable to cultural or national groups. We demonstrate the technique by collecting commonsense knowledge that surrounds six fairly universal rituals -- birth, coming-of-age, marriage, funerals, new year, and birthdays -- across two national groups: the United States and India. Our study expands the different types of relationships identified by existing work in the field of commonsense reasoning for commonplace events, and uses these new types to gather information that distinguish the identity of the groups providing the knowledge. It also moves us a step closer towards building a machine that doesn't assume a rigid framework of universal (and likely Western-biased) commonsense knowledge, but rather has the ability to reason in a contextually and culturally sensitive way. Our hope is that cultural knowledge of this sort will lead to more human-like performance in NLP tasks such as question answering (QA) and text understanding and generation.

Keerthiram Murugesan, Mattia Atzeni, Pushkar Shukla et al.

In this paper, we consider the recent trend of evaluating progress on reinforcement learning technology by using text-based environments and games as evaluation environments. This reliance on text brings advances in natural language processing into the ambit of these agents, with a recurring thread being the use of external knowledge to mimic and better human-level performance. We present one such instantiation of agents that use commonsense knowledge from ConceptNet to show promising performance on two text-based environments.

Mayank Agarwal, Jorge J. Barroso, Tathagata Chakraborti et al.

This whitepaper reports on Project CLAI (Command Line AI), which aims to bring the power of AI to the command line interface (CLI). The CLAI platform sets up the CLI as a new environment for AI researchers to conquer by surfacing the command line as a generic environment that researchers can interface to using a simple sense-act API, much like the traditional AI agent architecture. In this paper, we discuss the design and implementation of the platform in detail, through illustrative use cases of new end user interaction patterns enabled by this design, and through quantitative evaluation of the system footprint of a CLAI-enabled terminal. We also report on some early user feedback on CLAI's features from an internal survey.

Kshitij Fadnis, Kartik Talamadupula, Pavan Kapanipathi et al.

In this paper, we introduce the problem of knowledge graph contextualization -- that is, given a specific NLP task, the problem of extracting meaningful and relevant sub-graphs from a given knowledge graph. The task in the case of this paper is the textual entailment problem, and the context is a relevant sub-graph for an instance of the textual entailment problem -- where given two sentences p and h, the entailment relationship between them has to be predicted automatically. We base our methodology on finding paths in a cost-customized external knowledge graph, and building the most relevant sub-graph that connects p and h. We show that our path selection mechanism to generate sub-graphs not only reduces noise, but also retrieves meaningful information from large knowledge graphs. Our evaluation shows that using information on entities as well as the relationships between them improves on the performance of purely text-based systems.

Pavan Kapanipathi, Veronika Thost, Siva Sankalp Patel et al.

Textual entailment is a fundamental task in natural language processing. Most approaches for solving the problem use only the textual content present in training data. A few approaches have shown that information from external knowledge sources like knowledge graphs (KGs) can add value, in addition to the textual content, by providing background knowledge that may be critical for a task. However, the proposed models do not fully exploit the information in the usually large and noisy KGs, and it is not clear how it can be effectively encoded to be useful for entailment. We present an approach that complements text-based entailment models with information from KGs by (1) using Personalized PageR- ank to generate contextual subgraphs with reduced noise and (2) encoding these subgraphs using graph convolutional networks to capture KG structure. Our technique extends the capability of text models exploiting structural and semantic information found in KGs. We evaluate our approach on multiple textual entailment datasets and show that the use of external knowledge helps improve prediction accuracy. This is particularly evident in the challenging BreakingNLI dataset, where we see an absolute improvement of 5-20% over multiple text-based entailment models.

Selmer Bringsjord, Naveen Sundar Govindarajulu, Atriya Sen et al.

We briefly introduce herein a new form of distributed, multi-agent artificial intelligence, which we refer to as "tentacular." Tentacular AI is distinguished by six attributes, which among other things entail a capacity for reasoning and planning based in highly expressive calculi (logics), and which enlists subsidiary agents across distances circumscribed only by the reach of one or more given networks.

Ryan Musa, Xiaoyan Wang, Achille Fokoue et al.

Open-domain question answering (QA) is an important problem in AI and NLP that is emerging as a bellwether for progress on the generalizability of AI methods and techniques. Much of the progress in open-domain QA systems has been realized through advances in information retrieval methods and corpus construction. In this paper, we focus on the recently introduced ARC Challenge dataset, which contains 2,590 multiple choice questions authored for grade-school science exams. These questions are selected to be the most challenging for current QA systems, and current state of the art performance is only slightly better than random chance. We present a system that rewrites a given question into queries that are used to retrieve supporting text from a large corpus of science-related text. Our rewriter is able to incorporate background knowledge from ConceptNet and -- in tandem with a generic textual entailment system trained on SciTail that identifies support in the retrieved results -- outperforms several strong baselines on the end-to-end QA task despite only being trained to identify essential terms in the original source question. We use a generalizable decision methodology over the retrieved evidence and answer candidates to select the best answer. By combining query rewriting, background knowledge, and textual entailment our system is able to outperform several strong baselines on the ARC dataset.

Xiaoyan Wang, Pavan Kapanipathi, Ryan Musa et al.

Natural Language Inference (NLI) is fundamental to many Natural Language Processing (NLP) applications including semantic search and question answering. The NLI problem has gained significant attention thanks to the release of large scale, challenging datasets. Present approaches to the problem largely focus on learning-based methods that use only textual information in order to classify whether a given premise entails, contradicts, or is neutral with respect to a given hypothesis. Surprisingly, the use of methods based on structured knowledge -- a central topic in artificial intelligence -- has not received much attention vis-a-vis the NLI problem. While there are many open knowledge bases that contain various types of reasoning information, their use for NLI has not been well explored. To address this, we present a combination of techniques that harness knowledge graphs to improve performance on the NLI problem in the science questions domain. We present the results of applying our techniques on text, graph, and text-to-graph based models, and discuss implications for the use of external knowledge in solving the NLI problem. Our model achieves the new state-of-the-art performance on the NLI problem over the SciTail science questions dataset.

Michael Boratko, Harshit Padigela, Divyendra Mikkilineni et al.

The recent work of Clark et al. introduces the AI2 Reasoning Challenge (ARC) and the associated ARC dataset that partitions open domain, complex science questions into an Easy Set and a Challenge Set. That paper includes an analysis of 100 questions with respect to the types of knowledge and reasoning required to answer them; however, it does not include clear definitions of these types, nor does it offer information about the quality of the labels. We propose a comprehensive set of definitions of knowledge and reasoning types necessary for answering the questions in the ARC dataset. Using ten annotators and a sophisticated annotation interface, we analyze the distribution of labels across the Challenge Set and statistics related to them. Additionally, we demonstrate that although naive information retrieval methods return sentences that are irrelevant to answering the query, sufficient supporting text is often present in the (ARC) corpus. Evaluating with human-selected relevant sentences improves the performance of a neural machine comprehension model by 42 points.

Matthew Peveler, Naveen Sundar Govindarajulu, Selmer Bringsjord et al.

As computational power has continued to increase, and sensors have become more accurate, the corresponding advent of systems that are at once cognitive and immersive has arrived. These \textit{cognitive and immersive systems} (CAISs) fall squarely into the intersection of AI with HCI/HRI: such systems interact with and assist the human agents that enter them, in no small part because such systems are infused with AI able to understand and reason about these humans and their knowledge, beliefs, goals, communications, plans, etc. We herein explain our approach to engineering CAISs. We emphasize the capacity of a CAIS to develop and reason over a `theory of the mind' of its human partners. This capacity entails that the AI in question has a sophisticated model of the beliefs, knowledge, goals, desires, emotions, etc.\ of these humans. To accomplish this engineering, a formal framework of very high expressivity is needed. In our case, this framework is a \textit{cognitive event calculus}, a particular kind of quantified multi-operator modal logic, and a matching high-expressivity automated reasoner and planner. To explain, advance, and to a degree validate our approach, we show that a calculus of this type satisfies a set of formal requirements, and can enable a CAIS to understand a psychologically tricky scenario couched in what we call the \textit{cognitive polysolid framework} (CPF). We also formally show that a room that satisfies these requirements can have a useful property we term \emph{expectation of usefulness}. CPF, a sub-class of \textit{cognitive microworlds}, includes machinery able to represent and plan over not merely blocks and actions (such as seen in the primitive `blocks worlds' of old), but also over agents and their mental attitudes about both other agents and inanimate objects.

Kartik Talamadupula, Biplav Srivastava, Jeffrey O. Kephart

In this paper, we introduce the problem of denoting and deriving the complexity of workflows (plans, schedules) in collaborative, planner-assisted settings where humans and agents are trying to jointly solve a task. The interactions -- and hence the workflows that connect the human and the agents -- may differ according to the domain and the kind of agents. We adapt insights from prior work in human-agent teaming and workflow analysis to suggest metrics for workflow complexity. The main motivation behind this work is to highlight metrics for human comprehensibility of plans and schedules. The planning community has seen its fair share of work on the synthesis of plans that take diversity into account -- what value do such plans hold if their generation is not guided at least in part by metrics that reflect the ease of engaging with and using those plans?

Tathagata Chakraborti, Kshitij P. Fadnis, Kartik Talamadupula et al.

In this paper, we report on the visualization capabilities of an Explainable AI Planning (XAIP) agent that can support human in the loop decision making. Imposing transparency and explainability requirements on such agents is especially important in order to establish trust and common ground with the end-to-end automated planning system. Visualizing the agent's internal decision-making processes is a crucial step towards achieving this. This may include externalizing the "brain" of the agent -- starting from its sensory inputs, to progressively higher order decisions made by it in order to drive its planning components. We also show how the planner can bootstrap on the latest techniques in explainable planning to cast plan visualization as a plan explanation problem, and thus provide concise model-based visualization of its plans. We demonstrate these functionalities in the context of the automated planning components of a smart assistant in an instrumented meeting space.

Tathagata Chakraborti, Kartik Talamadupula, Kshitij P. Fadnis et al.

In this paper, we present UbuntuWorld 1.0 LTS - a platform for developing automated technical support agents in the Ubuntu operating system. Specifically, we propose to use the Bash terminal as a simulator of the Ubuntu environment for a learning-based agent and demonstrate the usefulness of adopting reinforcement learning (RL) techniques for basic problem solving and troubleshooting in this environment. We provide a plug-and-play interface to the simulator as a python package where different types of agents can be plugged in and evaluated, and provide pathways for integrating data from online support forums like AskUbuntu into an automated agent's learning process. Finally, we show that the use of this data significantly improves the agent's learning efficiency. We believe that this platform can be adopted as a real-world test bed for research on automated technical support.

Iulian Vlad Serban, Tim Klinger, Gerald Tesauro et al.

We introduce the multiresolution recurrent neural network, which extends the sequence-to-sequence framework to model natural language generation as two parallel discrete stochastic processes: a sequence of high-level coarse tokens, and a sequence of natural language tokens. There are many ways to estimate or learn the high-level coarse tokens, but we argue that a simple extraction procedure is sufficient to capture a wealth of high-level discourse semantics. Such procedure allows training the multiresolution recurrent neural network by maximizing the exact joint log-likelihood over both sequences. In contrast to the standard log- likelihood objective w.r.t. natural language tokens (word perplexity), optimizing the joint log-likelihood biases the model towards modeling high-level abstractions. We apply the proposed model to the task of dialogue response generation in two challenging domains: the Ubuntu technical support domain, and Twitter conversations. On Ubuntu, the model outperforms competing approaches by a substantial margin, achieving state-of-the-art results according to both automatic evaluation metrics and a human evaluation study. On Twitter, the model appears to generate more relevant and on-topic responses according to automatic evaluation metrics. Finally, our experiments demonstrate that the proposed model is more adept at overcoming the sparsity of natural language and is better able to capture long-term structure.

Kartik Talamadupula, David E. Smith, Subbarao Kambhampati

When autonomous agents are executing in the real world, the state of the world as well as the objectives of the agent may change from the agent's original model. In such cases, the agent's planning process must modify the plan under execution to make it amenable to the new conditions, and to resume execution. This brings up the replanning problem, and the various techniques that have been proposed to solve it. In all, three main techniques -- based on three different metrics -- have been proposed in prior automated planning work. An open question is whether these metrics are interchangeable; answering this requires a normalized comparison of the various replanning quality metrics. In this paper, we show that it is possible to support such a comparison by compiling all the respective techniques into a single substrate. Using this novel compilation, we demonstrate that these different metrics are not interchangeable, and that they are not good surrogates for each other. Thus we focus attention on the incompatibility of the various replanning flavors with each other, founded in the differences between the metrics that they respectively seek to optimize.

Srijith Ravikumar, Kartik Talamadupula, Raju Balakrishnan et al.

The increasing popularity of Twitter renders improved trustworthiness and relevance assessment of tweets much more important for search. However, given the limitations on the size of tweets, it is hard to extract measures for ranking from the tweets' content alone. We present a novel ranking method, called RAProp, which combines two orthogonal measures of relevance and trustworthiness of a tweet. The first, called Feature Score, measures the trustworthiness of the source of the tweet. This is done by extracting features from a 3-layer twitter ecosystem, consisting of users, tweets and the pages referred to in the tweets. The second measure, called agreement analysis, estimates the trustworthiness of the content of the tweet, by analyzing how and whether the content is independently corroborated by other tweets. We view the candidate result set of tweets as the vertices of a graph, with the edges measuring the estimated agreement between each pair of tweets. The feature score is propagated over this agreement graph to compute the top-k tweets that have both trustworthy sources and independent corroboration. The evaluation of our method on 16 million tweets from the TREC 2011 Microblog Dataset shows that for top-30 precision we achieve 53% higher than current best performing method on the Dataset and over 300% over current Twitter Search. We also present a detailed internal empirical evaluation of RAProp in comparison to several alternative approaches proposed by us.

Kartik Talamadupula, Subbarao Kambhampati

There has been significant interest in crowdsourcing and human computation. One subclass of human computation applications are those directed at tasks that involve planning (e.g. travel planning) and scheduling (e.g. conference scheduling). Much of this work appears outside the traditional automated planning forums, and at the outset it is not clear whether automated planning has much of a role to play in these human computation systems. Interestingly however, work on these systems shows that even primitive forms of automated oversight of the human planner does help in significantly improving the effectiveness of the humans/crowd. In this paper, we will argue that the automated oversight used in these systems can be viewed as a primitive automated planner, and that there are several opportunities for more sophisticated automated planning in effectively steering crowdsourced planning. Straightforward adaptation of current planning technology is however hampered by the mismatch between the capabilities of human workers and automated planners. We identify two important challenges that need to be overcome before such adaptation of planning technology can occur: (i) interpreting the inputs of the human workers (and the requester) and (ii) steering or critiquing the plans being produced by the human workers armed only with incomplete domain and preference models. In this paper, we discuss approaches for handling these challenges, and characterize existing human computation systems in terms of the specific choices they make in handling these challenges.

Kartik Talamadupula, Octavian Udrea, Anton Riabov et al.

As network traffic monitoring software for cybersecurity, malware detection, and other critical tasks becomes increasingly automated, the rate of alerts and supporting data gathered, as well as the complexity of the underlying model, regularly exceed human processing capabilities. Many of these applications require complex models and constituent rules in order to come up with decisions that influence the operation of entire systems. In this paper, we motivate the novel "strategic planning" problem -- one of gathering data from the world and applying the underlying model of the domain in order to come up with decisions that will monitor the system in an automated manner. We describe our use of automated planning methods to this problem, including the technique that we used to solve it in a manner that would scale to the demands of a real-time, real world scenario. We then present a PDDL model of one such application scenario related to network administration and monitoring, followed by a description of a novel integrated system that was built to accept generated plans and to continue the execution process. Finally, we present evaluations of two different automated planners and their different capabilities with our integrated system, both on a six-month window of network data, and using a simulator.