Alessandro Oltramari

Kaixin Ma, Filip Ilievski, Jonathan Francis et al. · cmu

Procedural text understanding is a challenging language reasoning task that requires models to track entity states across the development of a narrative. A complete procedural understanding solution should combine three core aspects: local and global views of the inputs, and global view of outputs. Prior methods considered a subset of these aspects, resulting in either low precision or low recall. In this paper, we propose Coalescing Global and Local Information (CGLI), a new model that builds entity- and timestep-aware input representations (local input) considering the whole context (global input), and we jointly model the entity states with a structured prediction objective (global output). Thus, CGLI simultaneously optimizes for both precision and recall. We extend CGLI with additional output layers and integrate it into a story reasoning framework. Extensive experiments on a popular procedural text understanding dataset show that our model achieves state-of-the-art results; experiments on a story reasoning benchmark show the positive impact of our model on downstream reasoning.

Jiarui Zhang, Filip Ilievski, Kaixin Ma et al. · cmu

Intelligent Traffic Monitoring (ITMo) technologies hold the potential for improving road safety/security and for enabling smart city infrastructure. Understanding traffic situations requires a complex fusion of perceptual information with domain-specific and causal commonsense knowledge. Whereas prior work has provided benchmarks and methods for traffic monitoring, it remains unclear whether models can effectively align these information sources and reason in novel scenarios. To address this assessment gap, we devise three novel text-based tasks for situational reasoning in the traffic domain: i) BDD-QA, which evaluates the ability of Language Models (LMs) to perform situational decision-making, ii) TV-QA, which assesses LMs' abilities to reason about complex event causality, and iii) HDT-QA, which evaluates the ability of models to solve human driving exams. We adopt four knowledge-enhanced methods that have shown generalization capability across language reasoning tasks in prior work, based on natural language inference, commonsense knowledge-graph self-supervision, multi-QA joint training, and dense retrieval of domain information. We associate each method with a relevant knowledge source, including knowledge graphs, relevant benchmarks, and driving manuals. In extensive experiments, we benchmark various knowledge-aware methods against the three datasets, under zero-shot evaluation; we provide in-depth analyses of model performance on data partitions and examine model predictions categorically, to yield useful insights on traffic understanding, given different background knowledge and reasoning strategies.

Ehsan Qasemi, Jonathan M. Francis, Alessandro Oltramari · cmu

Video Question Answering (VidQA) exhibits remarkable potential in facilitating advanced machine reasoning capabilities within the domains of Intelligent Traffic Monitoring and Intelligent Transportation Systems. Nevertheless, the integration of urban traffic scene knowledge into VidQA systems has received limited attention in previous research endeavors. In this work, we present a novel approach termed Traffic-domain Video Question Answering with Automatic Captioning (TRIVIA), which serves as a weak-supervision technique for infusing traffic-domain knowledge into large video-language models. Empirical findings obtained from the SUTD-TrafficQA task highlight the substantial enhancements achieved by TRIVIA, elevating the accuracy of representative video-language models by a remarkable 6.5 points (19.88%) compared to baseline settings. This pioneering methodology holds great promise for driving advancements in the field, inspiring researchers and practitioners alike to unlock the full potential of emerging video-language models in traffic-related applications.

Jiarui Zhang, Filip Ilievski, Kaixin Ma et al. · cmu

Self-supervision based on the information extracted from large knowledge graphs has been shown to improve the generalization of language models, in zero-shot evaluation on various downstream language reasoning tasks. Since these improvements are reported in aggregate, however, little is known about (i) how to select the appropriate knowledge for solid performance across tasks, (ii) how to combine this knowledge with neural language models, and (iii) how these pairings affect granular task performance. In this paper, we study the effect of knowledge sampling strategies and sizes that can be used to generate synthetic data for adapting language models. We study the effect of different synthetic datasets on language models with various architectures and sizes. The resulting models are evaluated against four task properties: domain overlap, answer similarity, vocabulary overlap, and answer length. Our experiments show that encoder-decoder models benefit from more data to learn from, whereas sampling strategies that balance across different aspects yield best performance. Most of the improvement occurs on questions with short answers and dissimilar answer candidates, which corresponds to the characteristics of the data used for pre-training.

Jiarui Zhang, Filip Ilievski, Aravinda Kollaa et al. · cmu

Understanding novel situations in the traffic domain requires an intricate combination of domain-specific and causal commonsense knowledge. Prior work has provided sufficient perception-based modalities for traffic monitoring, in this paper, we focus on a complementary research aspect of Intelligent Transportation: traffic understanding. We scope our study to text-based methods and datasets given the abundant commonsense knowledge that can be extracted using language models from large corpus and knowledge graphs. We adopt three knowledge-driven approaches for zero-shot QA over traffic situations, based on prior natural language inference methods, commonsense models with knowledge graph self-supervision, and dense retriever-based models. We constructed two text-based multiple-choice question answering sets: BDD-QA for evaluating causal reasoning in the traffic domain and HDT-QA for measuring the possession of domain knowledge akin to human driving license tests. Among the methods, Unified-QA reaches the best performance on the BDD-QA dataset with the adaptation of multiple formats of question answers. Language models trained with inference information and commonsense knowledge are also good at predicting the cause and effect in the traffic domain but perform badly at answering human-driving QA sets. For such sets, DPR+Unified-QA performs the best due to its efficient knowledge extraction.

Aaron Lohner, Francesco Compagno, Jonathan Francis et al.

Recognizing a traffic accident is an essential part of any autonomous driving or road monitoring system. An accident can appear in a wide variety of forms, and understanding what type of accident is taking place may be useful to prevent it from recurring. This work focuses on classifying traffic scenes into specific accident types. We approach the problem by representing a traffic scene as a graph, where objects such as cars can be represented as nodes, and relative distances and directions between them as edges. This representation of a traffic scene is referred to as a scene graph, and can be used as input for an accident classifier. Better results are obtained with a classifier that fuses the scene graph input with visual and textual representations. This work introduces a multi-stage, multimodal pipeline that pre-processes videos of traffic accidents, encodes them as scene graphs, and aligns this representation with vision and language modalities before executing the classification task. When trained on 4 classes, our method achieves a balanced accuracy score of 57.77% on an (unbalanced) subset of the popular Detection of Traffic Anomaly (DoTA) benchmark, representing an increase of close to 5 percentage points from the case where scene graph information is not taken into account.

Alessandro Oltramari

High-level reasoning can be defined as the capability to generalize over knowledge acquired via experience, and to exhibit robust behavior in novel situations. Such form of reasoning is a basic skill in humans, who seamlessly use it in a broad spectrum of tasks, from language communication to decision making in complex situations. When it manifests itself in understanding and manipulating the everyday world of objects and their interactions, we talk about common sense or commonsense reasoning. State-of-the-art AI systems don't possess such capability: for instance, Large Language Models have recently become popular by demonstrating remarkable fluency in conversing with humans, but they still make trivial mistakes when probed for commonsense competence; on a different level, performance degradation outside training data prevents self-driving vehicles to safely adapt to unseen scenarios, a serious and unsolved problem that limits the adoption of such technology. In this paper we propose to enable high-level reasoning in AI systems by integrating cognitive architectures with external neuro-symbolic components. We illustrate a hybrid framework centered on ACT-R and we discuss the role of generative models in recent and future applications.

Siyu Wu, Alessandro Oltramari, Jonathan Francis et al.

Resolving the dichotomy between the human-like yet constrained reasoning processes of Cognitive Architectures and the broad but often noisy inference behavior of Large Language Models (LLMs) remains a challenging but exciting pursuit, for enabling reliable machine reasoning capabilities in production systems. Because Cognitive Architectures are famously developed for the purpose of modeling the internal mechanisms of human cognitive decision-making at a computational level, new investigations consider the goal of informing LLMs with the knowledge necessary for replicating such processes, e.g., guided perception, memory, goal-setting, and action. Previous approaches that use LLMs for grounded decision-making struggle with complex reasoning tasks that require slower, deliberate cognition over fast and intuitive inference -- reporting issues related to the lack of sufficient grounding, as in hallucination. To resolve these challenges, we introduce LLM-ACTR, a novel neuro-symbolic architecture that provides human-aligned and versatile decision-making by integrating the ACT-R Cognitive Architecture with LLMs. Our framework extracts and embeds knowledge of ACT-R's internal decision-making process as latent neural representations, injects this information into trainable LLM adapter layers, and fine-tunes the LLMs for downstream prediction. Our experiments on novel Design for Manufacturing tasks show both improved task performance as well as improved grounded decision-making capability of our approach, compared to LLM-only baselines that leverage chain-of-thought reasoning strategies.

Ehsan Qasemi, Alessandro Oltramari

Challenges in Intelligent Traffic Monitoring (ITMo) are exacerbated by the large quantity and modalities of data and the need for the utilization of state-of-the-art (SOTA) reasoners. We formulate the problem of ITMo and introduce HANS, a neuro-symbolic architecture for multi-modal context understanding, and its application to ITMo. HANS utilizes knowledge graph technology to serve as a backbone for SOTA reasoning in the traffic domain. Through case studies, we show how HANS addresses the challenges associated with traffic monitoring while being able to integrate with a wide range of reasoning methods

Filip Ilievski, Barbara Hammer, Frank van Harmelen et al.

Recent advances in AI -- including generative approaches -- have resulted in technology that can support humans in scientific discovery and forming decisions, but may also disrupt democracies and target individuals. The responsible use of AI and its participation in human-AI teams increasingly shows the need for AI alignment, that is, to make AI systems act according to our preferences. A crucial yet often overlooked aspect of these interactions is the different ways in which humans and machines generalise. In cognitive science, human generalisation commonly involves abstraction and concept learning. In contrast, AI generalisation encompasses out-of-domain generalisation in machine learning, rule-based reasoning in symbolic AI, and abstraction in neurosymbolic AI. In this perspective paper, we combine insights from AI and cognitive science to identify key commonalities and differences across three dimensions: notions of, methods for, and evaluation of generalisation. We map the different conceptualisations of generalisation in AI and cognitive science along these three dimensions and consider their role for alignment in human-AI teaming. This results in interdisciplinary challenges across AI and cognitive science that must be tackled to provide a foundation for effective and cognitively supported alignment in human-AI teaming scenarios.

Vinay K Chaudhri, Chaitan Baru, Brandon Bennett et al.

The long-standing goal of creating a comprehensive, multi-purpose knowledge resource, reminiscent of the 1984 Cyc project, still persists in AI. Despite the success of knowledge resources like WordNet, ConceptNet, Wolfram|Alpha and other commercial knowledge graphs, verifiable, general-purpose widely available sources of knowledge remain a critical deficiency in AI infrastructure. Large language models struggle due to knowledge gaps; robotic planning lacks necessary world knowledge; and the detection of factually false information relies heavily on human expertise. What kind of knowledge resource is most needed in AI today? How can modern technology shape its development and evaluation? A recent AAAI workshop gathered over 50 researchers to explore these questions. This paper synthesizes our findings and outlines a community-driven vision for a new knowledge infrastructure. In addition to leveraging contemporary advances in knowledge representation and reasoning, one promising idea is to build an open engineering framework to exploit knowledge modules effectively within the context of practical applications. Such a framework should include sets of conventions and social structures that are adopted by contributors.

Alessandro Oltramari, Jonathan Francis, Filip Ilievski et al.

This chapter illustrates how suitable neuro-symbolic models for language understanding can enable domain generalizability and robustness in downstream tasks. Different methods for integrating neural language models and knowledge graphs are discussed. The situations in which this combination is most appropriate are characterized, including quantitative evaluation and qualitative error analysis on a variety of commonsense question answering benchmark datasets.

Kaixin Ma, Filip Ilievski, Jonathan Francis et al.

Commonsense reasoning benchmarks have been largely solved by fine-tuning language models. The downside is that fine-tuning may cause models to overfit to task-specific data and thereby forget their knowledge gained during pre-training. Recent works only propose lightweight model updates as models may already possess useful knowledge from past experience, but a challenge remains in understanding what parts and to what extent models should be refined for a given task. In this paper, we investigate what models learn from commonsense reasoning datasets. We measure the impact of three different adaptation methods on the generalization and accuracy of models. Our experiments with two models show that fine-tuning performs best, by learning both the content and the structure of the task, but suffers from overfitting and limited generalization to novel answers. We observe that alternative adaptation methods like prefix-tuning have comparable accuracy, but generalize better to unseen answers and are more robust to adversarial splits.

Filip Ilievski, Alessandro Oltramari, Kaixin Ma et al.

Commonsense knowledge is essential for many AI applications, including those in natural language processing, visual processing, and planning. Consequently, many sources that include commonsense knowledge have been designed and constructed over the past decades. Recently, the focus has been on large text-based sources, which facilitate easier integration with neural (language) models and application to textual tasks, typically at the expense of the semantics of the sources and their harmonization. Efforts to consolidate commonsense knowledge have yielded partial success, with no clear path towards a comprehensive solution. We aim to organize these sources around a common set of dimensions of commonsense knowledge. We survey a wide range of popular commonsense sources with a special focus on their relations. We consolidate these relations into 13 knowledge dimensions. This consolidation allows us to unify the separate sources and to compute indications of their coverage, overlap, and gaps with respect to the knowledge dimensions. Moreover, we analyze the impact of each dimension on downstream reasoning tasks that require commonsense knowledge, observing that the temporal and desire/goal dimensions are very beneficial for reasoning on current downstream tasks, while distinctness and lexical knowledge have little impact. These results reveal preferences for some dimensions in current evaluation, and potential neglect of others.

Yikang Li, Pulkit Goel, Varsha Kuppur Rajendra et al.

Conditional text generation has been a challenging task that is yet to see human-level performance from state-of-the-art models. In this work, we specifically focus on the Commongen benchmark, wherein the aim is to generate a plausible sentence for a given set of input concepts. Despite advances in other tasks, large pre-trained language models that are fine-tuned on this dataset often produce sentences that are syntactically correct but qualitatively deviate from a human understanding of common sense. Furthermore, generated sequences are unable to fulfill such lexical requirements as matching part-of-speech and full concept coverage. In this paper, we explore how commonsense knowledge graphs can enhance model performance, with respect to commonsense reasoning and lexically-constrained decoding. We propose strategies for enhancing the semantic correctness of the generated text, which we accomplish through: extracting commonsense relations from Conceptnet, injecting these relations into the Unified Language Model (UniLM) through attention mechanisms, and enforcing the aforementioned lexical requirements through output constraints. By performing several ablations, we find that commonsense injection enables the generation of sentences that are more aligned with human understanding, while remaining compliant with lexical requirements.

Kaixin Ma, Filip Ilievski, Jonathan Francis et al.

Recent developments in pre-trained neural language modeling have led to leaps in accuracy on commonsense question-answering benchmarks. However, there is increasing concern that models overfit to specific tasks, without learning to utilize external knowledge or perform general semantic reasoning. In contrast, zero-shot evaluations have shown promise as a more robust measure of a model's general reasoning abilities. In this paper, we propose a novel neuro-symbolic framework for zero-shot question answering across commonsense tasks. Guided by a set of hypotheses, the framework studies how to transform various pre-existing knowledge resources into a form that is most effective for pre-training models. We vary the set of language models, training regimes, knowledge sources, and data generation strategies, and measure their impact across tasks. Extending on prior work, we devise and compare four constrained distractor-sampling strategies. We provide empirical results across five commonsense question-answering tasks with data generated from five external knowledge resources. We show that, while an individual knowledge graph is better suited for specific tasks, a global knowledge graph brings consistent gains across different tasks. In addition, both preserving the structure of the task as well as generating fair and informative questions help language models learn more effectively.

Alessandro Oltramari, Jonathan Francis, Cory Henson et al.

Computational context understanding refers to an agent's ability to fuse disparate sources of information for decision-making and is, therefore, generally regarded as a prerequisite for sophisticated machine reasoning capabilities, such as in artificial intelligence (AI). Data-driven and knowledge-driven methods are two classical techniques in the pursuit of such machine sense-making capability. However, while data-driven methods seek to model the statistical regularities of events by making observations in the real-world, they remain difficult to interpret and they lack mechanisms for naturally incorporating external knowledge. Conversely, knowledge-driven methods, combine structured knowledge bases, perform symbolic reasoning based on axiomatic principles, and are more interpretable in their inferential processing; however, they often lack the ability to estimate the statistical salience of an inference. To combat these issues, we propose the use of hybrid AI methodology as a general framework for combining the strengths of both approaches. Specifically, we inherit the concept of neuro-symbolism as a way of using knowledge-bases to guide the learning progress of deep neural networks. We further ground our discussion in two applications of neuro-symbolism and, in both cases, show that our systems maintain interpretability while achieving comparable performance, relative to the state-of-the-art.

Kaixin Ma, Jonathan Francis, Quanyang Lu et al.

Non-extractive commonsense QA remains a challenging AI task, as it requires systems to reason about, synthesize, and gather disparate pieces of information, in order to generate responses to queries. Recent approaches on such tasks show increased performance, only when models are either pre-trained with additional information or when domain-specific heuristics are used, without any special consideration regarding the knowledge resource type. In this paper, we perform a survey of recent commonsense QA methods and we provide a systematic analysis of popular knowledge resources and knowledge-integration methods, across benchmarks from multiple commonsense datasets. Our results and analysis show that attention-based injection seems to be a preferable choice for knowledge integration and that the degree of domain overlap, between knowledge bases and datasets, plays a crucial role in determining model success.

Alessandro Oltramari, Alexander Kott

The prominence and use of the concept of cyber risk has been rising in recent years. This paper presents empirical investigations focused on two important and distinct groups within the broad community of cyber-defense professionals and researchers: (1) cyber practitioners and (2) developers of cyber ontologies. The key finding of this work is that the ways the concept of cyber risk is treated by practitioners of cybersecurity is largely inconsistent with definitions of cyber risk commonly offered in the literature. Contrary to commonly cited definitions of cyber risk, concepts such as the likelihood of an event and the extent of its impact are not used by cybersecurity practitioners. This is also the case for use of these concepts in the current generation of cybersecurity ontologies. Instead, terms and concepts reflective of the adversarial nature of cyber defense appear to take the most prominent roles. This research offers the first quantitative empirical evidence that rejection of traditional concepts of cyber risk by cybersecurity professionals is indeed observed in real-world practice.