Generative AI has a problem with trust. Knowledge-based AI does not, but faces the knowledge bottleneck: the cost of manually building the lexicons and ontologies that enable reasoning, explanation, and targeted correction. We present a neurosymbolic approach toward overcoming this bottleneck. This approach uses a symbolic semantic language interpreter and static knowledge resources (ontology, lexicon, and a corpus of 1,780 validated text meaning representations) of OntoAgent, a cognitive architecture, in coordination with a language model guided by narrative descriptions of decision-making algorithms that encode principles from decades of knowledge acquisition research. The acquisition process: a) is triggered when the semantic anlyzer encounters knowledge gaps while constructing meanng representations; b) proposes new lexicon entries and ontological concepts; and c) automatically validates them against ontological constraints. We report the results of an initial experiment using this approach as the first step toward establishing what an LLM can achieve when boot-strapped from deep knowledge resources and integrated with a knowledge-based analyzer. This work describes a step in an ongoing R&D program whose core objective is to demonstrate that, just as GPUs enabled neural methods to scale, the use of language models as knowledge acquisition tools will enable knowledge-based AI to scale.

Deploying embodied AI agents in the physical world demands cognitive capabilities for long-horizon planning that execute reliably, deterministically, and transparently. We present HARMONIC, a cognitive-robotic architecture that pairs OntoAgent, a content-centric cognitive architecture providing metacognitive self-monitoring, domain-grounded diagnosis, and consequence-based action selection over ontologically structured knowledge, with a modular reactive tactical layer. HARMONIC's modular design enables a functional evaluation of whether LLMs can replicate OntoAgent's cognitive capabilities, evaluated within the same robotic system under identical conditions. Six LLMs spanning frontier and efficient tiers replace OntoAgent in a collaborative maintenance scenario under native and knowledge-equalized conditions. Results reveal that LLMs do not consistently assess their own knowledge state before acting, causing downstream failures in diagnostic reasoning and action selection. These deficits persist even with equivalent procedural knowledge, indicating the issues are architectural rather than knowledge-based. These findings support the design of physically embodied systems in which cognitive architectures retain primary authority for reasoning, owing to their deterministic and transparent characteristics.

Robots operating alongside humans must recognize what they do not know before acting, diagnose problems from domain knowledge, and reason about action consequences. These capabilities are operational requirements, not optimization targets, and their absence produces silent and unrecoverable failures. We present a first-of-its-kind controlled comparison between OntoAgent, our content-centric cognitive architecture, and six LLMs spanning frontier and efficient tiers as drop-in replacements at the strategic layer of the same robotic system in HARMONIC. LLMs fail to verify their knowledge state before acting, even when given equivalent procedural knowledge. The deficit is architectural, not knowledge-based. Knowledge-grounded architectures must retain decision authority; LLMs contribute where their strengths apply.

Shapes of cognition is a new conceptual paradigm for the computational cognitive modeling of Language-Endowed Intelligent Agents (LEIAs). Shapes are remembered constellations of sensory, linguistic, conceptual, episodic, and procedural knowledge that allow agents to cut through the complexity of real life the same way as people do: by expecting things to be typical, recognizing patterns, acting by habit, reasoning by analogy, satisficing, and generally minimizing cognitive load to the degree situations permit. Atypical outcomes are treated using shapes-based recovery methods, such as learning on the fly, asking a human partner for help, or seeking an actionable, even if imperfect, situational understanding. Although shapes is an umbrella term, it is not vague: shapes-based modeling involves particular objectives, hypotheses, modeling strategies, knowledge bases, and actual models of wide-ranging phenomena, all implemented within a particular cognitive architecture. Such specificity is needed both to vet our hypotheses and to achieve our practical aims of building useful agent systems that are explainable, extensible, and worthy of our trust, even in critical domains. However, although the LEIA example of shapes-based modeling is specific, the principles can be applied more broadly, giving new life to knowledge-based and hybrid AI.

We report a neurosymbolic approach to authorship anonymization that combines knowledge-based paraphrasing, grounded in cognitive modeling, with support functions provided by a large language model (LLM). The cognitive model accounts for four things: what it means to faithfully retain meaning and discourse coherence in a paraphrase, how do deal with polysemy given that full semantic analysis of open text is beyond the state of the art, how to define and characterize an author’s style, and how to leverage human linguistic capabilities when preparing systems to automatically anonymize texts. LLMs augment the knowledge-based paraphrases in three ways: by filtering out atypical formulations, by selecting the best from multiple candidate paraphrases, and by offering additional paraphrases in case the knowledge-based paraphrasing fails to adequately anonymize the text. This neurosymbolic architecture favors knowledge-based processing for being reliable and explainable, while exploiting LLMs for what they do best: manipulate regularities in the surface form of language.

This paper introduces HARMONIC, a cognitive-robotic architecture designed for robots in human-robotic teams. HARMONIC supports semantic perception interpretation, human-like decision-making, and intentional language communication. It addresses the issues of safety and quality of results; aims to solve problems of data scarcity, explainability, and safety; and promotes transparency and trust. Two proof-of-concept HARMONIC-based robotic systems are demonstrated, each implemented in both a high-fidelity simulation environment and on physical robotic platforms.

For AI agents to emulate human behavior, they must be able to perceive, meaningfully interpret, store, and use large amounts of information about the world, themselves, and other agents. Metacognition is a necessary component of all of these processes. In this paper, we briefly a) introduce content-centric computational cognitive (C4) modeling for next-generation AI agents; b) review the long history of developing C4 agents at RPI's LEIA (Language-Endowed Intelligent Agents) Lab; c) discuss our current work on extending LEIAs' cognitive capabilities to cognitive robotic applications developed using a neuro symbolic processing model; and d) sketch plans for future developments in this paradigm that aim to overcome underappreciated limitations of currently popular, LLM-driven methods in AI.

Metacognition is the concept of reasoning about an agent's own internal processes and was originally introduced in the field of developmental psychology. In this position paper, we examine the concept of applying metacognition to artificial intelligence. We introduce a framework for understanding metacognitive artificial intelligence (AI) that we call TRAP: transparency, reasoning, adaptation, and perception. We discuss each of these aspects in-turn and explore how neurosymbolic AI (NSAI) can be leveraged to address challenges of metacognition.

Explanation is key to people having confidence in high-stakes AI systems. However, machine-learning-based systems - which account for almost all current AI - can't explain because they are usually black boxes. The explainable AI (XAI) movement hedges this problem by redefining "explanation". The human-centered explainable AI (HCXAI) movement identifies the explanation-oriented needs of users but can't fulfill them because of its commitment to machine learning. In order to achieve the kinds of explanations needed by real people operating in critical domains, we must rethink how to approach AI. We describe a hybrid approach to developing cognitive agents that uses a knowledge-based infrastructure supplemented by data obtained through machine learning when applicable. These agents will serve as assistants to humans who will bear ultimate responsibility for the decisions and actions of the human-robot team. We illustrate the explanatory potential of such agents using the under-the-hood panels of a demonstration system in which a team of simulated robots collaborates on a search task assigned by a human.

We present HARMONIC, a framework for implementing cognitive robots that transforms general-purpose robots into trusted teammates capable of complex decision-making, natural communication and human-level explanation. The framework supports interoperability between a strategic (cognitive) layer for high-level decision-making and a tactical (robot) layer for low-level control and execution. We describe the core features of the framework and our initial implementation, in which HARMONIC was deployed on a simulated UGV and drone involved in a multi-robot search and retrieval task.

This paper describes HARMONIC, a cognitive-robotic architecture that integrates the OntoAgent cognitive framework with general-purpose robot control systems applied to human-robot teaming (HRT). HARMONIC incorporates metacognition, meaningful natural language communication, and explainability capabilities required for developing mutual trust in HRT. Through simulation experiments involving a joint search task performed by a heterogeneous team of two HARMONIC-based robots and a human operator, we demonstrate heterogeneous robots that coordinate their actions, adapt to complex scenarios, and engage in natural human-robot communication. Evaluation results show that HARMONIC-based robots can reason about plans, goals, and team member attitudes while providing clear explanations for their decisions, which are essential requirements for realistic human-robot teaming.

A novel approach to hybrid AI aimed at developing trustworthy agent collaborators. The vast majority of current AI relies wholly on machine learning (ML). However, the past thirty years of effort in this paradigm have shown that, despite the many things that ML can achieve, it is not an all-purpose solution to building human-like intelligent systems. One hope for overcoming this limitation is hybrid AI: that is, AI that combines ML with knowledge-based processing. In Agents in the Long Game of AI, Marjorie McShane, Sergei Nirenburg, and Jesse English present recent advances in hybrid AI with special emphases on content-centric computational cognitive modeling, explainability, and development methodologies. At present, hybridization typically involves sprinkling knowledge into an ML black box. The authors, by contrast, argue that hybridization will be best achieved in the opposite way: by building agents within a cognitive architecture and then integrating judiciously selected ML results. This approach leverages the power of ML without sacrificing the kind of explainability that will foster society's trust in AI. This book shows how we can develop trustworthy agent collaborators of a type not being addressed by the “ML alone” or “ML sprinkled by knowledge” paradigms—and why it is imperative to do so.

Neurosymbolic approaches to AI typically involve attempts to reincorporate the structure and speed of symbolic reasoning into the flexible representations of deep learning. “Knowledge,” in this understanding, is typically represented in a structured ontology or knowledge base that relies on human expertise and effort to construct. In this paper, we present a vision for “language-endowed intelligent agents,” a type of lifelong learner which begins with a hand-crafted knowledge base and a deep language understander and increases it over the course of its life through dialogue and interaction with both humans and other AI systems—generative large language learning models in particular. We discuss the requirements for such a system, present evidence toward the feasibility of the approach, and conclude with future challenges and research directions.

The paper describes a system that uses large language model (LLM) technology to support the automatic learning of new entries in an intelligent agent's semantic lexicon. The process is bootstrapped by an existing non-toy lexicon and a natural language generator that converts formal, ontologically-grounded representations of meaning into natural language sentences. The learning method involves a sequence of LLM requests and includes an automatic quality control step. To date, this learning method has been applied to learning multiword expressions whose meanings are equivalent to those of transitive verbs in the agent's lexicon. The experiment demonstrates the benefits of a hybrid learning architecture that integrates knowledge-based methods and resources with both traditional data analytics and LLMs.

In the framework of computational cognitive modeling, natural language understanding and generation must be integrated with other cognitive capabilities, such as reasoning and learning. The language understanding component of an intelligent agent extracts and formally represents the meaning of texts and dialog turns. The output of language understanding must reflect the speaker’s intended meaning and be sufficiently detailed to serve as input to reasoning and action in artificial intelligent agents. One kind of agent action is verbal, so agents must include a language generation capability. This chapter describes a particular language understanding system that meets the requirements for the above language capabilities and also puts forward methodological arguments about the interplay between theories, models, and computational systems.