Search

Knowledge Components (KCs) linked to assessments enhance the measurement of student learning, enrich analytics, and facilitate adaptivity. However, generating and linking KCs to assessment items requires significant effort and domain-specific knowledge. To streamline this process for higher-education courses, we employed GPT-4 to generate KCs for multiple-choice questions (MCQs) in Chemistry and E-Learning. We analyzed discrepancies between the KCs generated by the Large Language Model (LLM) and those made by humans through evaluation from three domain experts in each subject area. This evaluation aimed to determine whether, in instances of non-matching KCs, evaluators showed a preference for the LLM-generated KCs over their human-created counterparts. We also developed an ontology induction algorithm to cluster questions that assess similar KCs based on their content. Our most effective LLM strategy accurately matched KCs for 56% of Chemistry and 35% of E-Learning MCQs, with even higher success when considering the top five KC suggestions. Human evaluators favored LLM-generated KCs, choosing them over human-assigned ones approximately two-thirds of the time, a preference that was statistically significant across both domains. Our clustering algorithm successfully grouped questions by their underlying KCs without needing explicit labels or contextual information. This research advances the automation of KC generation and classification for assessment items, alleviating the need for student data or predefined KC labels., Comment: Learning @ Scale 2024

Conversational tutoring systems (CTSs) offer learning experiences through interactions based on natural language. They are recognized for promoting cognitive engagement and improving learning outcomes, especially in reasoning tasks. Nonetheless, the cost associated with authoring CTS content is a major obstacle to widespread adoption and to research on effective instructional design. In this paper, we discuss and evaluate a novel type of CTS that leverages recent advances in large language models (LLMs) in two ways: First, the system enables AI-assisted content authoring by inducing an easily editable tutoring script automatically from a lesson text. Second, the system automates the script orchestration in a learning-by-teaching format via two LLM-based agents (Ruffle&Riley) acting as a student and a professor. The system allows for free-form conversations that follow the ITS-typical inner and outer loop structure. We evaluate Ruffle&Riley's ability to support biology lessons in two between-subject online user studies (N = 200) comparing the system to simpler QA chatbots and reading activity. Analyzing system usage patterns, pre/post-test scores and user experience surveys, we find that Ruffle&Riley users report high levels of engagement, understanding and perceive the offered support as helpful. Even though Ruffle&Riley users require more time to complete the activity, we did not find significant differences in short-term learning gains over the reading activity. Our system architecture and user study provide various insights for designers of future CTSs. We further open-source our system to support ongoing research on effective instructional design of LLM-based learning technologies., Comment: arXiv admin note: substantial text overlap with arXiv:2310.01420

We propose an intuitive LLM prompting framework (AgentKit) for multifunctional agents. AgentKit offers a unified framework for explicitly constructing a complex "thought process" from simple natural language prompts. The basic building block in AgentKit is a node, containing a natural language prompt for a specific subtask. The user then puts together chains of nodes, like stacking LEGO pieces. The chains of nodes can be designed to explicitly enforce a naturally structured "thought process". For example, for the task of writing a paper, one may start with the thought process of 1) identify a core message, 2) identify prior research gaps, etc. The nodes in AgentKit can be designed and combined in different ways to implement multiple advanced capabilities including on-the-fly hierarchical planning, reflection, and learning from interactions. In addition, due to the modular nature and the intuitive design to simulate explicit human thought process, a basic agent could be implemented as simple as a list of prompts for the subtasks and therefore could be designed and tuned by someone without any programming experience. Quantitatively, we show that agents designed through AgentKit achieve SOTA performance on WebShop and Crafter. These advances underscore AgentKit's potential in making LLM agents effective and accessible for a wider range of applications. https://github.com/holmeswww/AgentKit

Remarkable performance of large language models (LLMs) in a variety of tasks brings forth many opportunities as well as challenges of utilizing them in production settings. Towards practical adoption of LLMs, multi-agent systems hold great promise to augment, integrate, and orchestrate LLMs in the larger context of enterprise platforms that use existing proprietary data and models to tackle complex real-world tasks. Despite the tremendous success of these systems, current approaches rely on narrow, single-focus objectives for optimization and evaluation, often overlooking potential constraints in real-world scenarios, including restricted budgets, resources and time. Furthermore, interpreting, analyzing, and debugging these systems requires different components to be evaluated in relation to one another. This demand is currently not feasible with existing methodologies. In this postion paper, we introduce the concept of reasoning capacity as a unifying criterion to enable integration of constraints during optimization and establish connections among different components within the system, which also enable a more holistic and comprehensive approach to evaluation. We present a formal definition of reasoning capacity and illustrate its utility in identifying limitations within each component of the system. We then argue how these limitations can be addressed with a self-reflective process wherein human-feedback is used to alleviate shortcomings in reasoning and enhance overall consistency of the system.

Recent large language models (LLMs) have demonstrated great potential toward intelligent agents and next-gen automation, but there currently lacks a systematic benchmark for evaluating LLMs' abilities as agents. We introduce SmartPlay: both a challenging benchmark and a methodology for evaluating LLMs as agents. SmartPlay consists of 6 different games, including Rock-Paper-Scissors, Tower of Hanoi, Minecraft. Each game features a unique setting, providing up to 20 evaluation settings and infinite environment variations. Each game in SmartPlay uniquely challenges a subset of 9 important capabilities of an intelligent LLM agent, including reasoning with object dependencies, planning ahead, spatial reasoning, learning from history, and understanding randomness. The distinction between the set of capabilities each game test allows us to analyze each capability separately. SmartPlay serves not only as a rigorous testing ground for evaluating the overall performance of LLM agents but also as a road-map for identifying gaps in current methodologies. We release our benchmark at github.com/Microsoft/SmartPlay

Conversational tutoring systems (CTSs) offer learning experiences driven by natural language interaction. They are known to promote high levels of cognitive engagement and benefit learning outcomes, particularly in reasoning tasks. Nonetheless, the time and cost required to author CTS content is a major obstacle to widespread adoption. In this paper, we introduce a novel type of CTS that leverages the recent advances in large language models (LLMs) in two ways: First, the system induces a tutoring script automatically from a lesson text. Second, the system automates the script orchestration via two LLM-based agents (Ruffle&Riley) with the roles of a student and a professor in a learning-by-teaching format. The system allows a free-form conversation that follows the ITS-typical inner and outer loop structure. In an initial between-subject online user study (N = 100) comparing Ruffle&Riley to simpler QA chatbots and reading activity, we found no significant differences in post-test scores. Nonetheless, in the learning experience survey, Ruffle&Riley users expressed higher ratings of understanding and remembering and further perceived the offered support as more helpful and the conversation as coherent. Our study provides insights for a new generation of scalable CTS technologies., Comment: NeurIPS'23 GAIED, Camera-ready

Open-world survival games pose significant challenges for AI algorithms due to their multi-tasking, deep exploration, and goal prioritization requirements. Despite reinforcement learning (RL) being popular for solving games, its high sample complexity limits its effectiveness in complex open-world games like Crafter or Minecraft. We propose a novel approach, SPRING, to read the game's original academic paper and use the knowledge learned to reason and play the game through a large language model (LLM). Prompted with the LaTeX source as game context and a description of the agent's current observation, our SPRING framework employs a directed acyclic graph (DAG) with game-related questions as nodes and dependencies as edges. We identify the optimal action to take in the environment by traversing the DAG and calculating LLM responses for each node in topological order, with the LLM's answer to final node directly translating to environment actions. In our experiments, we study the quality of in-context "reasoning" induced by different forms of prompts under the setting of the Crafter open-world environment. Our experiments suggest that LLMs, when prompted with consistent chain-of-thought, have great potential in completing sophisticated high-level trajectories. Quantitatively, SPRING with GPT-4 outperforms all state-of-the-art RL baselines, trained for 1M steps, without any training. Finally, we show the potential of games as a test bed for LLMs.

Pre-trained large language models (LLMs) capture procedural knowledge about the world. Recent work has leveraged LLM's ability to generate abstract plans to simplify challenging control tasks, either by action scoring, or action modeling (fine-tuning). However, the transformer architecture inherits several constraints that make it difficult for the LLM to directly serve as the agent: e.g. limited input lengths, fine-tuning inefficiency, bias from pre-training, and incompatibility with non-text environments. To maintain compatibility with a low-level trainable actor, we propose to instead use the knowledge in LLMs to simplify the control problem, rather than solving it. We propose the Plan, Eliminate, and Track (PET) framework. The Plan module translates a task description into a list of high-level sub-tasks. The Eliminate module masks out irrelevant objects and receptacles from the observation for the current sub-task. Finally, the Track module determines whether the agent has accomplished each sub-task. On the AlfWorld instruction following benchmark, the PET framework leads to a significant 15% improvement over SOTA for generalization to human goal specifications.

We propose that symbols are first and foremost external communication tools used between intelligent agents that allow knowledge to be transferred in a more efficient and effective manner than having to experience the world directly. But, they are also used internally within an agent through a form of self-communication to help formulate, describe and justify subsymbolic patterns of neural activity that truly implement thinking. Symbols, and our languages that make use of them, not only allow us to explain our thinking to others and ourselves, but also provide beneficial constraints (inductive bias) on learning about the world. In this paper we present relevant insights from neuroscience and cognitive science, about how the human brain represents symbols and the concepts they refer to, and how today's artificial neural networks can do the same. We then present a novel neuro-symbolic hypothesis and a plausible architecture for intelligent agents that combines subsymbolic representations for symbols and concepts for learning and reasoning. Our hypothesis and associated architecture imply that symbols will remain critical to the future of intelligent systems NOT because they are the fundamental building blocks of thought, but because they are characterizations of subsymbolic processes that constitute thought., Comment: 28 pages

While Large Language Models (LLMs) have shown exceptional performance in various tasks, one of their most prominent drawbacks is generating inaccurate or false information with a confident tone. In this paper, we provide evidence that the LLM's internal state can be used to reveal the truthfulness of statements. This includes both statements provided to the LLM, and statements that the LLM itself generates. Our approach is to train a classifier that outputs the probability that a statement is truthful, based on the hidden layer activations of the LLM as it reads or generates the statement. Experiments demonstrate that given a set of test sentences, of which half are true and half false, our trained classifier achieves an average of 71\% to 83\% accuracy labeling which sentences are true versus false, depending on the LLM base model. Furthermore, we explore the relationship between our classifier's performance and approaches based on the probability assigned to the sentence by the LLM. We show that while LLM-assigned sentence probability is related to sentence truthfulness, this probability is also dependent on sentence length and the frequencies of words in the sentence, resulting in our trained classifier providing a more reliable approach to detecting truthfulness, highlighting its potential to enhance the reliability of LLM-generated content and its practical applicability in real-world scenarios.

High sample complexity has long been a challenge for RL. On the other hand, humans learn to perform tasks not only from interaction or demonstrations, but also by reading unstructured text documents, e.g., instruction manuals. Instruction manuals and wiki pages are among the most abundant data that could inform agents of valuable features and policies or task-specific environmental dynamics and reward structures. Therefore, we hypothesize that the ability to utilize human-written instruction manuals to assist learning policies for specific tasks should lead to a more efficient and better-performing agent. We propose the Read and Reward framework. Read and Reward speeds up RL algorithms on Atari games by reading manuals released by the Atari game developers. Our framework consists of a QA Extraction module that extracts and summarizes relevant information from the manual and a Reasoning module that evaluates object-agent interactions based on information from the manual. An auxiliary reward is then provided to a standard A2C RL agent, when interaction is detected. Experimentally, various RL algorithms obtain significant improvement in performance and training speed when assisted by our design.

The task of triplet extraction aims to extract pairs of entities and their corresponding relations from unstructured text. Most existing methods train an extraction model on training data involving specific target relations, and are incapable of extracting new relations that were not observed at training time. Generalizing the model to unseen relations typically requires fine-tuning on synthetic training data which is often noisy and unreliable. We show that by reducing triplet extraction to a template infilling task over a pre-trained language model (LM), we can equip the extraction model with zero-shot learning capabilities and eliminate the need for additional training data. We propose a novel framework, ZETT (ZEro-shot Triplet extraction by Template infilling), that aligns the task objective to the pre-training objective of generative transformers to generalize to unseen relations. Experiments on FewRel and Wiki-ZSL datasets demonstrate that ZETT shows consistent and stable performance, outperforming previous state-of-the-art methods, even when using automatically generated templates. https://github.com/megagonlabs/zett/, Comment: IJCNLP-AACL 2023 (main)

Millions of learners worldwide are now using intelligent tutoring systems (ITSs). At their core, ITSs rely on machine learning algorithms to track each user's changing performance level over time to provide personalized instruction. Crucially, student performance models are trained using interaction sequence data of previous learners to analyse data generated by future learners. This induces a cold-start problem when a new course is introduced for which no training data is available. Here, we consider transfer learning techniques as a way to provide accurate performance predictions for new courses by leveraging log data from existing courses. We study two settings: (i) In the naive transfer setting, we propose course-agnostic performance models that can be applied to any course. (ii) In the inductive transfer setting, we tune pre-trained course-agnostic performance models to new courses using small-scale target course data (e.g., collected during a pilot study). We evaluate the proposed techniques using student interaction sequence data from 5 different mathematics courses containing data from over 47,000 students in a real world large-scale ITS. The course-agnostic models that use additional features provided by human domain experts (e.g, difficulty ratings for questions in the new course) but no student interaction training data for the new course, achieve prediction accuracy on par with standard BKT and PFA models that use training data from thousands of students in the new course. In the inductive setting our transfer learning approach yields more accurate predictions than conventional performance models when only limited student interaction training data (<100 students) is available to both.

One of the challenges faced by conversational agents is their inability to identify unstated presumptions of their users' commands, a task trivial for humans due to their common sense. In this paper, we propose a zero-shot commonsense reasoning system for conversational agents in an attempt to achieve this. Our reasoner uncovers unstated presumptions from user commands satisfying a general template of if-(state), then-(action), because-(goal). Our reasoner uses a state-of-the-art transformer-based generative commonsense knowledge base (KB) as its source of background knowledge for reasoning. We propose a novel and iterative knowledge query mechanism to extract multi-hop reasoning chains from the neural KB which uses symbolic logic rules to significantly reduce the search space. Similar to any KBs gathered to date, our commonsense KB is prone to missing knowledge. Therefore, we propose to conversationally elicit the missing knowledge from human users with our novel dynamic question generation strategy, which generates and presents contextualized queries to human users. We evaluate the model with a user study with human users that achieves a 35% higher success rate compared to SOTA., Comment: Appearing in the 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP)

We consider the problem of assessing the changing performance levels of individual students as they go through online courses. This student performance (SP) modeling problem is a critical step for building adaptive online teaching systems. Specifically, we conduct a study of how to utilize various types and large amounts of student log data to train accurate machine learning (ML) models that predict the performance of future students. This study is the first to use four very large sets of student data made available recently from four distinct intelligent tutoring systems. Our results include a new ML approach that defines a new state of the art for logistic regression based SP modeling, improving over earlier methods in several ways: First, we achieve improved accuracy by introducing new features that can be easily computed from conventional question-response logs (e.g., the pattern in the student 's most recent answers). Second, we take advantage of features of the student history that go beyond question-response pairs (e.g., features such as which video segments the student watched, or skipped) as well as information about prerequisite structure in the curriculum. Third, we train multiple specialized SP models for different aspects of the curriculum (e.g., specializing in early versus later segments of the student history), then combine these specialized models to create a group prediction of the SP. Taken together, these innovations yield an average AUC score across these four datasets of 0.808 compared to the previous best logistic regression approach score of 0.767, and also outperforming state-of-the-art deep neural net approaches. Importantly, we observe consistent improvements from each of our three methodological innovations, in each dataset, suggesting that our methods are of general utility and likely to produce improvements for other online tutoring systems as well.

When faced with learning challenging new tasks, humans often follow sequences of steps that allow them to incrementally build up the necessary skills for performing these new tasks. However, in machine learning, models are most often trained to solve the target tasks directly.Inspired by human learning, we propose a novel curriculum learning approach which decomposes challenging tasks into sequences of easier intermediate goals that are used to pre-train a model before tackling the target task. We focus on classification tasks, and design the intermediate tasks using an automatically constructed label hierarchy. We train the model at each level of the hierarchy, from coarse labels to fine labels, transferring acquired knowledge across these levels. For instance, the model will first learn to distinguish animals from objects, and then use this acquired knowledge when learning to classify among more fine-grained classes such as cat, dog, car, and truck. Most existing curriculum learning algorithms for supervised learning consist of scheduling the order in which the training examples are presented to the model. In contrast, our approach focuses on the output space of the model. We evaluate our method on several established datasets and show significant performance gains especially on classification problems with many labels. We also evaluate on a new synthetic dataset which allows us to study multiple aspects of our method.

We introduce the Probabilistic Worldbuilding Model (PWM), a new fully-symbolic Bayesian model of semantic parsing and reasoning, as a first step in a research program toward more domain- and task-general NLU and AI. Humans create internal mental models of their observations which greatly aid in their ability to understand and reason about a large variety of problems. In PWM, the meanings of sentences, acquired facts about the world, and intermediate steps in reasoning are all expressed in a human-readable formal language, with the design goal of interpretability. PWM is Bayesian, designed specifically to be able to generalize to new domains and new tasks. We derive and implement an inference algorithm that reads sentences by parsing and abducing updates to its latent world model that capture the semantics of those sentences, and evaluate it on two out-of-domain question-answering datasets: (1) ProofWriter and (2) a new dataset we call FictionalGeoQA, designed to be more representative of real language but still simple enough to focus on evaluating reasoning ability, while being robust against heuristics. Our method outperforms baselines on both, thereby demonstrating its value as a proof-of-concept., Comment: Accepted to TACL; pre-MIT Press publication version

With an outreach in more than 90 countries, a market share of 2.1 billion dollars and a target audience base of at least 1.2 billion people, Bollywood, aka the Mumbai film industry, is a formidable entertainment force. While the number of lives Bollywood can potentially touch is massive, no comprehensive NLP study on the evolution of social and gender biases in Bollywood dialogues exists. Via a substantial corpus of movie dialogues spanning a time horizon of 70 years, we seek to understand the portrayal of women, in a broader context studying subtle social signals, and analyze the evolving trends in geographic and religious representation in India. Our argument is simple -- popular movie content reflects social norms and beliefs in some form or shape. In this project, we propose to analyze such trends over 70 years of Bollywood movies contrasting them with their Hollywood counterpart and critically acclaimed world movies.

The growing political polarization of the American electorate over the last several decades has been widely studied and documented. During the administration of President Donald Trump, charges of "fake news" made social and news media not only the means but, to an unprecedented extent, the topic of political communication. Using data from before the November 3rd, 2020 US Presidential election, recent work has demonstrated the viability of using YouTube's social media ecosystem to obtain insights into the extent of US political polarization as well as the relationship between this polarization and the nature of the content and commentary provided by different US news networks. With that work as background, this paper looks at the sharp transformation of the relationship between news consumers and here-to-fore "fringe" news media channels in the 64 days between the US presidential election and the violence that took place at US Capitol on January 6th. This paper makes two distinct types of contributions. The first is to introduce a novel methodology to analyze large social media data to study the dynamics of social political news networks and their viewers. The second is to provide insights into what actually happened regarding US political social media channels and their viewerships during this volatile 64 day period.

Representing the semantics of GUI screens and components is crucial to data-driven computational methods for modeling user-GUI interactions and mining GUI designs. Existing GUI semantic representations are limited to encoding either the textual content, the visual design and layout patterns, or the app contexts. Many representation techniques also require significant manual data annotation efforts. This paper presents Screen2Vec, a new self-supervised technique for generating representations in embedding vectors of GUI screens and components that encode all of the above GUI features without requiring manual annotation using the context of user interaction traces. Screen2Vec is inspired by the word embedding method Word2Vec, but uses a new two-layer pipeline informed by the structure of GUIs and interaction traces and incorporates screen- and app-specific metadata. Through several sample downstream tasks, we demonstrate Screen2Vec's key useful properties: representing between-screen similarity through nearest neighbors, composability, and capability to represent user tasks., Comment: Accepted to CHI Conference on Human Factors in Computing Systems (CHI 2021)

Polarization among US political parties, media and elites is a widely studied topic. Prominent lines of prior research across multiple disciplines have observed and analyzed growing polarization in social media. In this paper, we present a new methodology that offers a fresh perspective on interpreting polarization through the lens of machine translation. With a novel proposition that two sub-communities are speaking in two different \emph{languages}, we demonstrate that modern machine translation methods can provide a simple yet powerful and interpretable framework to understand the differences between two (or more) large-scale social media discussion data sets at the granularity of words. Via a substantial corpus of 86.6 million comments by 6.5 million users on over 200,000 news videos hosted by YouTube channels of four prominent US news networks, we demonstrate that simple word-level and phrase-level translation pairs can reveal deep insights into the current political divide -- what is \emph{black lives matter} to one can be \emph{all lives matter} to the other.

How meaning is represented in the brain is still one of the big open questions in neuroscience. Does a word (e.g., bird) always have the same representation, or does the task under which the word is processed alter its representation (answering "can you eat it?" versus "can it fly?")? The brain activity of subjects who read the same word while performing different semantic tasks has been shown to differ across tasks. However, it is still not understood how the task itself contributes to this difference. In the current work, we study Magnetoencephalography (MEG) brain recordings of participants tasked with answering questions about concrete nouns. We investigate the effect of the task (i.e. the question being asked) on the processing of the concrete noun by predicting the millisecond-resolution MEG recordings as a function of both the semantics of the noun and the task. Using this approach, we test several hypotheses about the task-stimulus interactions by comparing the zero-shot predictions made by these hypotheses for novel tasks and nouns not seen during training. We find that incorporating the task semantics significantly improves the prediction of MEG recordings, across participants. The improvement occurs 475-550ms after the participants first see the word, which corresponds to what is considered to be the ending time of semantic processing for a word. These results suggest that only the end of semantic processing of a word is task-dependent, and pose a challenge for future research to formulate new hypotheses for earlier task effects as a function of the task and stimuli., Comment: accepted at NeurIPS 2020

Bilingual lexicons and phrase tables are critical resources for modern Machine Translation systems. Although recent results show that without any seed lexicon or parallel data, highly accurate bilingual lexicons can be learned using unsupervised methods, such methods rely on the existence of large, clean monolingual corpora. In this work, we utilize a single Skip-gram model trained on a multilingual corpus yielding polyglot word embeddings, and present a novel finding that a surprisingly simple constrained nearest-neighbor sampling technique in this embedding space can retrieve bilingual lexicons, even in harsh social media data sets predominantly written in English and Romanized Hindi and often exhibiting code switching. Our method does not require monolingual corpora, seed lexicons, or any other such resources. Additionally, across three European language pairs, we observe that polyglot word embeddings indeed learn a rich semantic representation of words and substantial bilingual lexicons can be retrieved using our constrained nearest neighbor sampling. We investigate potential reasons and downstream applications in settings spanning both clean texts and noisy social media data sets, and in both resource-rich and under-resourced language pairs.

A confluence of advances in the computer and mathematical sciences has unleashed unprecedented capabilities for enabling true evidence-based decision making. These capabilities are making possible the large-scale capture of data and the transformation of that data into insights and recommendations in support of decisions about challenging problems in science, society, and government. Key advances include jumps in the availability of rich streams of data, precipitous drops in the cost of storing and retrieving massive amounts of data, exponential increases in computing power and memory, and jumps in the prowess of methods for performing machine learning and reasoning. These advances have come together to create an inflection point in our ability to harness large amounts of data for generating insights and guiding decision making. The shift of commerce, science, education, art, and entertainment to the web makes available unprecedented quantities of structured and unstructured databases about human activities - much of it available to anyone who wishes to mine it for insights. In the sciences, new evidential paradigms and sensing technologies are making available great quantities of data, via use of fundamentally new kinds of low-cost sensors (e.g., genomic microarrays) or through viewers that provide unprecedented scope and resolution. The data pose a huge opportunity for data-centric analyses. To date, we have only scratched the surface of the potential for learning from these large-scale data sets. Opportunities abound for tapping our new capabilities more broadly to provide insights to decision makers and to enhance the quality of their actions and policies., Comment: A Computing Community Consortium (CCC) white paper, 8 pages

In order for conversational AI systems to hold more natural and broad-ranging conversations, they will require much more commonsense, including the ability to identify unstated presumptions of their conversational partners. For example, in the command "If it snows at night then wake me up early because I don't want to be late for work" the speaker relies on commonsense reasoning of the listener to infer the implicit presumption that they wish to be woken only if it snows enough to cause traffic slowdowns. We consider here the problem of understanding such imprecisely stated natural language commands given in the form of "if-(state), then-(action), because-(goal)" statements. More precisely, we consider the problem of identifying the unstated presumptions of the speaker that allow the requested action to achieve the desired goal from the given state (perhaps elaborated by making the implicit presumptions explicit). We release a benchmark data set for this task, collected from humans and annotated with commonsense presumptions. We present a neuro-symbolic theorem prover that extracts multi-hop reasoning chains, and apply it to this problem. Furthermore, to accommodate the reality that current AI commonsense systems lack full coverage, we also present an interactive conversational framework built on our neuro-symbolic system, that conversationally evokes commonsense knowledge from humans to complete its reasoning chains., Comment: Appearing in the 35th AAAI international Conference on Artificial Intelligence, 2021

Many machine learning systems today are trained on large amounts of human-annotated data. Data annotation tasks that require a high level of competency make data acquisition expensive, while the resulting labels are often subjective, inconsistent, and may contain a variety of human biases. To improve the data quality, practitioners often need to collect multiple annotations per example and aggregate them before training models. Such a multi-stage approach results in redundant annotations and may often produce imperfect "ground truth" that may limit the potential of training accurate machine learning models. We propose a new end-to-end framework that enables us to: (i) merge the aggregation step with model training, thus allowing deep learning systems to learn to predict ground truth estimates directly from the available data, and (ii) model difficulties of examples and learn representations of the annotators that allow us to estimate and take into account their competencies. Our approach is general and has many applications, including training more accurate models on crowdsourced data, ensemble learning, as well as classifier accuracy estimation from unlabeled data. We conduct an extensive experimental evaluation of our method on 5 crowdsourcing datasets of varied difficulty and show accuracy gains of up to 25% over the current state-of-the-art approaches for aggregating annotations, as well as significant reductions in the required annotation redundancy.

We argue that a key challenge in enabling usable and useful interactive task learning for intelligent agents is to facilitate effective Human-AI collaboration. We reflect on our past 5 years of efforts on designing, developing and studying the SUGILITE system, discuss the issues on incorporating recent advances in AI with HCI principles in mixed-initiative interactions and multi-modal interactions, and summarize the lessons we learned. Lastly, we identify several challenges and opportunities, and describe our ongoing work

Machine learning has shown growing success in recent years. However, current machine learning systems are highly specialized, trained for particular problems or domains, and typically on a single narrow dataset. Human learning, on the other hand, is highly general and adaptable. Never-ending learning is a machine learning paradigm that aims to bridge this gap, with the goal of encouraging researchers to design machine learning systems that can learn to perform a wider variety of inter-related tasks in more complex environments. To date, there is no environment or testbed to facilitate the development and evaluation of never-ending learning systems. To this end, we propose the Jelly Bean World testbed. The Jelly Bean World allows experimentation over two-dimensional grid worlds which are filled with items and in which agents can navigate. This testbed provides environments that are sufficiently complex and where more generally intelligent algorithms ought to perform better than current state-of-the-art reinforcement learning approaches. It does so by producing non-stationary environments and facilitating experimentation with multi-task, multi-agent, multi-modal, and curriculum learning settings. We hope that this new freely-available software will prompt new research and interest in the development and evaluation of never-ending learning systems and more broadly, general intelligence systems., Comment: Published as a conference paper at ICLR 2020

The ability to inferring latent psychological traits from human behavior is key to developing personalized human-interacting machine learning systems. Approaches to infer such traits range from surveys to manually-constructed experiments and games. However, these traditional games are limited because they are typically designed based on heuristics. In this paper, we formulate the task of designing \emph{behavior diagnostic games} that elicit distinguishable behavior as a mutual information maximization problem, which can be solved by optimizing a variational lower bound. Our framework is instantiated by using prospect theory to model varying player traits, and Markov Decision Processes to parameterize the games. We validate our approach empirically, showing that our designed games can successfully distinguish among players with different traits, outperforming manually-designed ones by a large margin., Comment: 33rd Conference on Neural Information Processing Systems (NeurIPS 2019)

Manipulating data, such as weighting data examples or augmenting with new instances, has been increasingly used to improve model training. Previous work has studied various rule- or learning-based approaches designed for specific types of data manipulation. In this work, we propose a new method that supports learning different manipulation schemes with the same gradient-based algorithm. Our approach builds upon a recent connection of supervised learning and reinforcement learning (RL), and adapts an off-the-shelf reward learning algorithm from RL for joint data manipulation learning and model training. Different parameterization of the "data reward" function instantiates different manipulation schemes. We showcase data augmentation that learns a text transformation network, and data weighting that dynamically adapts the data sample importance. Experiments show the resulting algorithms significantly improve the image and text classification performance in low data regime and class-imbalance problems., Comment: NeurIPS 2019

Computing devices have recently become capable of interacting with their end users via natural language. However, they can only operate within a limited "supported" domain of discourse and fail drastically when faced with an out-of-domain utterance, mainly due to the limitations of their semantic parser. In this paper, we propose a semantic parser that generalizes to out-of-domain examples by learning a general strategy for parsing an unseen utterance through adapting the logical forms of seen utterances, instead of learning to generate a logical form from scratch. Our parser maintains a memory consisting of a representative subset of the seen utterances paired with their logical forms. Given an unseen utterance, our parser works by looking up a similar utterance from the memory and adapting its logical form until it fits the unseen utterance. Moreover, we present a data generation strategy for constructing utterance-logical form pairs from different domains. Our results show an improvement of up to 68.8% on one-shot parsing under two different evaluation settings compared to the baselines., Comment: 2019 Conference on Empirical Methods in Natural Language Processing (EMNLP)

Natural language programming is a promising approach to enable end users to instruct new tasks for intelligent agents. However, our formative study found that end users would often use unclear, ambiguous or vague concepts when naturally instructing tasks in natural language, especially when specifying conditionals. Existing systems have limited support for letting the user teach agents new concepts or explaining unclear concepts. In this paper, we describe a new multi-modal domain-independent approach that combines natural language programming and programming-by-demonstration to allow users to first naturally describe tasks and associated conditions at a high level, and then collaborate with the agent to recursively resolve any ambiguities or vagueness through conversations and demonstrations. Users can also define new procedures and concepts by demonstrating and referring to contents within GUIs of existing mobile apps. We demonstrate this approach in PUMICE, an end-user programmable agent that implements this approach. A lab study with 10 users showed its usability., Comment: The AAAI-20 Workshop on Intelligent Process Automation (IPA-20)

What is the relationship between sentence representations learned by deep recurrent models against those encoded by the brain? Is there any correspondence between hidden layers of these recurrent models and brain regions when processing sentences? Can these deep models be used to synthesize brain data which can then be utilized in other extrinsic tasks? We investigate these questions using sentences with simple syntax and semantics (e.g., The bone was eaten by the dog.). We consider multiple neural network architectures, including recently proposed ELMo and BERT. We use magnetoencephalography (MEG) brain recording data collected from human subjects when they were reading these simple sentences. Overall, we find that BERT's activations correlate the best with MEG brain data. We also find that the deep network representation can be used to generate brain data from new sentences to augment existing brain data. To the best of our knowledge, this is the first work showing that the MEG brain recording when reading a word in a sentence can be used to distinguish earlier words in the sentence. Our exploration is also the first to use deep neural network representations to generate synthetic brain data and to show that it helps in improving subsequent stimuli decoding task accuracy., Comment: Association for Computational Linguistics (ACL) 2019

Electroencephalography (EEG) recordings of brain activity taken while participants read or listen to language are widely used within the cognitive neuroscience and psycholinguistics communities as a tool to study language comprehension. Several time-locked stereotyped EEG responses to word-presentations -- known collectively as event-related potentials (ERPs) -- are thought to be markers for semantic or syntactic processes that take place during comprehension. However, the characterization of each individual ERP in terms of what features of a stream of language trigger the response remains controversial. Improving this characterization would make ERPs a more useful tool for studying language comprehension. We take a step towards better understanding the ERPs by fine-tuning a language model to predict them. This new approach to analysis shows for the first time that all of the ERPs are predictable from embeddings of a stream of language. Prior work has only found two of the ERPs to be predictable. In addition to this analysis, we examine which ERPs benefit from sharing parameters during joint training. We find that two pairs of ERPs previously identified in the literature as being related to each other benefit from joint training, while several other pairs of ERPs that benefit from joint training are suggestive of potential relationships. Extensions of this analysis that further examine what kinds of information in the model embeddings relate to each ERP have the potential to elucidate the processes involved in human language comprehension., Comment: To appear in Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics

Current state-of-the-art NMT systems use large neural networks that are not only slow to train, but also often require many heuristics and optimization tricks, such as specialized learning rate schedules and large batch sizes. This is undesirable as it requires extensive hyperparameter tuning. In this paper, we propose a curriculum learning framework for NMT that reduces training time, reduces the need for specialized heuristics or large batch sizes, and results in overall better performance. Our framework consists of a principled way of deciding which training samples are shown to the model at different times during training, based on the estimated difficulty of a sample and the current competence of the model. Filtering training samples in this manner prevents the model from getting stuck in bad local optima, making it converge faster and reach a better solution than the common approach of uniformly sampling training examples. Furthermore, the proposed method can be easily applied to existing NMT models by simply modifying their input data pipelines. We show that our framework can help improve the training time and the performance of both recurrent neural network models and Transformers, achieving up to a 70% decrease in training time, while at the same time obtaining accuracy improvements of up to 2.2 BLEU.

This paper focuses on how to take advantage of external knowledge bases (KBs) to improve recurrent neural networks for machine reading. Traditional methods that exploit knowledge from KBs encode knowledge as discrete indicator features. Not only do these features generalize poorly, but they require task-specific feature engineering to achieve good performance. We propose KBLSTM, a novel neural model that leverages continuous representations of KBs to enhance the learning of recurrent neural networks for machine reading. To effectively integrate background knowledge with information from the currently processed text, our model employs an attention mechanism with a sentinel to adaptively decide whether to attend to background knowledge and which information from KBs is useful. Experimental results show that our model achieves accuracies that surpass the previous state-of-the-art results for both entity extraction and event extraction on the widely used ACE2005 dataset., Comment: published at ACL 2017

As humans, we often rely on language to learn language. For example, when corrected in a conversation, we may learn from that correction, over time improving our language fluency. Inspired by this observation, we propose a learning algorithm for training semantic parsers from supervision (feedback) expressed in natural language. Our algorithm learns a semantic parser from users' corrections such as "no, what I really meant was before his job, not after", by also simultaneously learning to parse this natural language feedback in order to leverage it as a form of supervision. Unlike supervision with gold-standard logical forms, our method does not require the user to be familiar with the underlying logical formalism, and unlike supervision from denotation, it does not require the user to know the correct answer to their query. This makes our learning algorithm naturally scalable in settings where existing conversational logs are available and can be leveraged as training data. We construct a novel dataset of natural language feedback in a conversational setting, and show that our method is effective at learning a semantic parser from such natural language supervision., Comment: published at EMNLP 2018

To ensure readability, text is often written and presented with due formatting. These text formatting devices help the writer to effectively convey the narrative. At the same time, these help the readers pick up the structure of the discourse and comprehend the conveyed information. There have been a number of linguistic theories on discourse structure of text. However, these theories only consider unformatted text. Multimedia text contains rich formatting features which can be leveraged for various NLP tasks. In this paper, we study some of these discourse features in multimedia text and what communicative function they fulfil in the context. We examine how these multimedia discourse features can be used to improve an information extraction system. We show that the discourse and text layout features provide information that is complementary to lexical semantic information commonly used for information extraction. As a case study, we use these features to harvest structured subject knowledge of geometry from textbooks. We show that the harvested structured knowledge can be used to improve an existing solver for geometry problems, making it more accurate as well as more explainable.

We propose a simple modification to existing neural machine translation (NMT) models that enables using a single universal model to translate between multiple languages while allowing for language specific parameterization, and that can also be used for domain adaptation. Our approach requires no changes to the model architecture of a standard NMT system, but instead introduces a new component, the contextual parameter generator (CPG), that generates the parameters of the system (e.g., weights in a neural network). This parameter generator accepts source and target language embeddings as input, and generates the parameters for the encoder and the decoder, respectively. The rest of the model remains unchanged and is shared across all languages. We show how this simple modification enables the system to use monolingual data for training and also perform zero-shot translation. We further show it is able to surpass state-of-the-art performance for both the IWSLT-15 and IWSLT-17 datasets and that the learned language embeddings are able to uncover interesting relationships between languages., Comment: Published in the proceedings of Empirical Methods in Natural Language Processing (EMNLP), 2018

Translating the complex, multi-dimensional data from simulations of biomolecules to intuitive knowledge is a major challenge in computational chemistry and biology. The so-called "free energy landscape" is amongst the most fundamental concepts used by scientists to understand both static and dynamic properties of biomolecular systems. In this paper we use Markov models to design a strategy for mapping features of this landscape to sonic parameters, for use in conjunction with visual display techniques such as structural animations and free energy diagrams.

We propose an efficient method to estimate the accuracy of classifiers using only unlabeled data. We consider a setting with multiple classification problems where the target classes may be tied together through logical constraints. For example, a set of classes may be mutually exclusive, meaning that a data instance can belong to at most one of them. The proposed method is based on the intuition that: (i) when classifiers agree, they are more likely to be correct, and (ii) when the classifiers make a prediction that violates the constraints, at least one classifier must be making an error. Experiments on four real-world data sets produce accuracy estimates within a few percent of the true accuracy, using solely unlabeled data. Our models also outperform existing state-of-the-art solutions in both estimating accuracies, and combining multiple classifier outputs. The results emphasize the utility of logical constraints in estimating accuracy, thus validating our intuition.

Intelligent systems capable of automatically understanding natural language text are important for many artificial intelligence applications including mobile phone voice assistants, computer vision, and robotics. Understanding language often constitutes fitting new information into a previously acquired view of the world. However, many machine reading systems rely on the text alone to infer its meaning. In this paper, we pursue a different approach; machine reading methods that make use of background knowledge to facilitate language understanding. To this end, we have developed two methods: The first method addresses prepositional phrase attachment ambiguity. It uses background knowledge within a semi-supervised machine learning algorithm that learns from both labeled and unlabeled data. This approach yields state-of-the-art results on two datasets against strong baselines; The second method extracts relationships from compound nouns. Our knowledge-aware method for compound noun analysis accurately extracts relationships and significantly outperforms a baseline that does not make use of background knowledge., Comment: 28 pages

Events and entities are closely related; entities are often actors or participants in events and events without entities are uncommon. The interpretation of events and entities is highly contextually dependent. Existing work in information extraction typically models events separately from entities, and performs inference at the sentence level, ignoring the rest of the document. In this paper, we propose a novel approach that models the dependencies among variables of events, entities, and their relations, and performs joint inference of these variables across a document. The goal is to enable access to document-level contextual information and facilitate context-aware predictions. We demonstrate that our approach substantially outperforms the state-of-the-art methods for event extraction as well as a strong baseline for entity extraction., Comment: 11 pages, 2 figures, published at NAACL 2016

Characterizing relationships between people is fundamental for the understanding of narratives. In this work, we address the problem of inferring the polarity of relationships between people in narrative summaries. We formulate the problem as a joint structured prediction for each narrative, and present a model that combines evidence from linguistic and semantic features, as well as features based on the structure of the social community in the text. We also provide a clustering-based approach that can exploit regularities in narrative types. e.g., learn an affinity for love-triangles in romantic stories. On a dataset of movie summaries from Wikipedia, our structured models provide more than a 30% error-reduction over a competitive baseline that considers pairs of characters in isolation.

One important challenge for probabilistic logics is reasoning with very large knowledge bases (KBs) of imperfect information, such as those produced by modern web-scale information extraction systems. One scalability problem shared by many probabilistic logics is that answering queries involves "grounding" the query---i.e., mapping it to a propositional representation---and the size of a "grounding" grows with database size. To address this bottleneck, we present a first-order probabilistic language called ProPPR in which that approximate "local groundings" can be constructed in time independent of database size. Technically, ProPPR is an extension to stochastic logic programs (SLPs) that is biased towards short derivations; it is also closely related to an earlier relational learning algorithm called the path ranking algorithm (PRA). We show that the problem of constructing proofs for this logic is related to computation of personalized PageRank (PPR) on a linearized version of the proof space, and using on this connection, we develop a proveably-correct approximate grounding scheme, based on the PageRank-Nibble algorithm. Building on this, we develop a fast and easily-parallelized weight-learning algorithm for ProPPR. In experiments, we show that learning for ProPPR is orders magnitude faster than learning for Markov logic networks; that allowing mutual recursion (joint learning) in KB inference leads to improvements in performance; and that ProPPR can learn weights for a mutually recursive program with hundreds of clauses, which define scores of interrelated predicates, over a KB containing one million entities., Comment: arXiv admin note: substantial text overlap with arXiv:1305.2254

How can we correlate neural activity in the human brain as it responds to words, with behavioral data expressed as answers to questions about these same words? In short, we want to find latent variables, that explain both the brain activity, as well as the behavioral responses. We show that this is an instance of the Coupled Matrix-Tensor Factorization (CMTF) problem. We propose Scoup-SMT, a novel, fast, and parallel algorithm that solves the CMTF problem and produces a sparse latent low-rank subspace of the data. In our experiments, we find that Scoup-SMT is 50-100 times faster than a state-of-the-art algorithm for CMTF, along with a 5 fold increase in sparsity. Moreover, we extend Scoup-SMT to handle missing data without degradation of performance. We apply Scoup-SMT to BrainQ, a dataset consisting of a (nouns, brain voxels, human subjects) tensor and a (nouns, properties) matrix, with coupling along the nouns dimension. Scoup-SMT is able to find meaningful latent variables, as well as to predict brain activity with competitive accuracy. Finally, we demonstrate the generality of Scoup-SMT, by applying it on a Facebook dataset (users, friends, wall-postings); there, Scoup-SMT spots spammer-like anomalies., Comment: 9 pages

In recognition of the need to increase academic standards in its freshman composition course and expand its remedial English offerings, Laredo Junior College (LJC) in Texas developed a three-level remedial English program, with the lowest level focusing on sentence writing, the middle level on paragraph writing, and the upper level on essay writing. After mandatory placement, students should progress from one course to another based on standardized departmental exams and satisfactory performance on writing assignments. Once the curricular decisions had been made and placement criteria determined, several severe potential problems emerged: (1) the staff was worried that the prospect of taking as many as three remedial English courses before being allowed to take the college-level English course would be so demoralizing that prospective students would be driven away; (2) since many other disciplines also planned to impose entry-level writing requirements, there was great concern that enrollments would drop in the other disciplines for at least a year while students worked their way to the college level; and (3) there was concern that students attending college under full-time Pell Grants would be unable to schedule of full load of 12 hours given their scheduling limitations. Some of these problems were ameliorated by offering students the chance to take 6-hour courses made up of the first and second levels or second and third levels of the remedial program. Under the new standards, 77% of LJC's freshmen are required to take remedial English (30% at the lowest level, 30% at the middle level, and 40% at the upper level), and every discipline developed minimum reading requirements for courses. With these changes, LJC is now ready to face the anticipated impact of a state-mandated basic skills test for college freshmen. (UCM)