Dr. Kristen B Venable

The expansion of the Internet of Things (IoT) in the battlefield, Internet of Battlefield Things (IoBT), gives rise to new opportunities for enhancing situational awareness. To increase the potential of IoBT for situational awareness in critical decision making, the data from these devices must be processed into consumer-ready information objects and made available to consumers on demand. To address this challenge we propose a workflow that makes use of natural language processing (NLP) to query a database and return a response in natural language. Our solution utilizes Large Language Models (LLMs) sized for edge devices to perform NLP, as well as a graph database. These types of databases are well suited for the dynamic, connected networks pervasive in the IoBT. Our architecture employs LLMs for both mapping questions in natural language to Cypher database queries as well as to summarize the database output back to the user in natural language. We evaluated several medium-sized LLMs for both of these tasks on a database representing publicly available data from the US Army's Multi-purpose Sensing Area Multi-Purpose Sensing Area (MSA) at the Jornada Range in Las Cruces, NM. We observe that Llama 3.1 (8 billion parameters) outperforms the other models across all considered metrics. Most importantly, we note that, unlike current methods, our two step approach allows the relaxation of the Exact Match (EM) requirement of the produced Cypher queries with ground truth code and, in this way, it achieves a 19.4% increase in accuracy. Our workflow lays the groundwork for deploying LLMs on edge devices to enable natural language interactions with databases containing information objects for critical decision making.

Modern foundational models are often compressed via a combination of structured pruning and re-training to meet the strict compute, memory, and connectivity constraints of edge deployments. While state-of-the-art (SoTA) pruning schemes target the entire Transformer, we adopt a simple, layer-wise L 2 -norm pruning on only the multi-layer perceptron (MLP) blocks as a fixed baseline. Our focus is not on achieving maximal compression, but on isolating the impact of the re-training loss function: (i) L2-norm Pruning with Cross-Entropy Fine-Tuning (L2PFT), which relies on labeled data, versus (ii) L2-norm Pruning with KL-Divergence Self-Distillation (L2PSD), which utilizes only teacher logits without requiring labeled data. We evaluate both pipelines on the OLMo2-7B-SFT model for CommonsenseQA, suitable for intermittent or denied connectivity scenarios typical of edge networks. Under identical pruning schedules, L2PSD achieves comparable or superior test accuracy to L2PFT, indicating that the choice of loss function has a significant impact on compressed model recovery in resource-constrained environments.

The peer review process is strained by increasing submission volumes, reviewer fatigue, and inconsistent standards. While Large Language Models (LLMs) can aid in reviews, they are often overly optimistic and lack technical depth. We developed an innovative prompting strategy that, when applied to ChatGPT-4 on ICLR 2025 papers, reduced score inflation and generated reviews more closely aligned with human reviewer median scores.

This paper considers the challenges Large Language Models (LLMs) face when reasoning over text that includes information involving uncertainty explicitly quantified via probability values. This type of reasoning is relevant to a variety of contexts ranging from everyday conversations to medical decision-making. Despite improvements in the mathematical reasoning capabilities of LLMs, they still exhibit significant difficulties when it comes to probabilistic reasoning. To deal with this problem, we introduce the Bayesian Linguistic Inference Dataset (BLInD), a new dataset specifically designed to test the probabilistic reasoning capabilities of LLMs. We use BLInD to find out the limitations of LLMs for tasks involving probabilistic reasoning. In addition, we present several prompting strategies that map the problem to different formal representations, including Python code, probabilistic algorithms, and probabilistic logical programming. We conclude by providing an evaluation of our methods on BLInD and an adaptation of a causal reasoning question-answering dataset. Our empirical results highlight the effectiveness of our proposed strategies for multiple LLMs.
Code and Dataset - https://github.com/HLR/BLInD
Extended Version - https://arxiv.org/abs/2402.09614

Efficient information dissemination is crucial for supporting critical operations across domains like disaster response, autonomous vehicles, and sensor networks. This paper introduces a Multi-Agent Reinforcement Learning (MARL) approach as a significant step forward in achieving more decentralized, efficient, and collaborative information dissemination. We propose a Partially Observable Stochastic Game (POSG) formulation for information dissemination empowering each agent to decide on message forwarding independently, based on the observation of their one-hop neighborhood. This constitutes a significant paradigm shift from heuristics currently employed in real-world broadcast protocols. Our novel approach harnesses Graph Convolutional Reinforcement Learning and Graph Attention Networks (GATs) with dynamic attention to capture essential network features. We propose two approaches to accomplish cooperative information dissemination, L-DyAN and HL-DyAN, differing in terms of the information exchanged among agents. Our experimental results show that our trained policies outperform existing methods, including the state-of-the-art heuristic, in terms of network coverage and communication overhead on dynamic networks of varying density and behavior.

It is well known that successful human-AI collaboration depends on the perceived trustworthiness of the AI. We argue that a key to securing trust in such collaborations is ensuring that the AI competently addresses ethics' foundational role in engagements. Specifically, developers need to identify, address, and implement mechanisms for accommodating ethical components of AI choices. We propose an approach that instantiates ethics semantically as ontology-based moral policies. To accommodate the wide variation and interpretation of ethics, we capture such variations into ethics sets, which are situationally specific aggregations of relevant moral policies. We are extending our ontology-based policy management systems with new representations and capabilities to allow trustworthy AI-human ethical collaborative behavior. Moreover, we believe that such AI-human ethical encounters demand that trustworthiness is bi-directional – humans need to be able to assess and calibrate their actions to be consistent with the trustworthiness of AI in a given context, and AIs need to be able to do the same with respect to humans.

Effective operation and seamless cooperation of robotic systems are a fundamental component of next-generation technologies and applications. In contexts such as disaster response, swarm operations require coordinated behavior and mobility control to be handled in a distributed manner, with the quality of the agents' actions heavily relying on the communication between them and the underlying network. In this paper, we formulate the problem of dynamic network bridging in a novel Decentralized Partially Observable Markov Decision Process (Dec-POMDP), where a swarm of agents cooperates to form a link between two distant moving targets. Furthermore, we propose a Multi-Agent Reinforcement Learning (MARL) approach for the problem based on Graph Convolutional Reinforcement Learning (DGN) which naturally applies to the networked, distributed nature of the task. The proposed method is evaluated in a simulated environment and compared to a centralized heuristic baseline showing promising results. Moreover, a further step in the direction of sim-to-real transfer is presented, by additionally evaluating the proposed approach in a near Live Virtual Constructive (LVC) UAV framework.

In this paper, we evaluate the capability of transformer-based language models in making inferences over uncertain text that includes uncertain rules of reasoning. We cover both Pretrained Language Models (PLMs) and generative Large Language Models (LLMs). Our evaluation results show that both generations of language models struggle with reasoning over uncertain text. We propose a novel end-to-end fine-tuning approach, Probabilistic Constraint Training (PCT), that utilizes probabilistic logical rules as constraints in the fine-tuning phase without relying on these rules in the inference stage. To assess the effectiveness of PCT, we utilize the related corpora and, additionally, create a new and more challenging benchmark that, unlike the previous ones, uses instance-specific rules. Our study demonstrates that PCT improves the transformer-based language model's intrinsic reasoning and makes their probabilistic logical reasoning process more explicit and explainable. Furthermore, PCT equips these models to effectively handle novel situations, including higher reasoning depth, new domains, and complex probabilistic structures.

We consider a scenario where a user must make a set of correlated decisions and we propose a computational cognitive model of the deliberation process. We assume the user compactly expresses her preferences via soft constraints and we study how a psychology-based model of human decision-making, namely Multi-Alternative Decision Field Theory (MDFT), can be applied in this context. We design and study sequential and synchronous procedures which combine local decision-making on each variable, with constraint propagation, as well as a one-shot approach. Our experimental results, which focus on tree-shaped Fuzzy Constraint Satisfaction Problems, suggest that decomposing the decision process along the preference structure allows to find solutions of high quality in terms of preferences, maintains MDFT's ability to replicate behavioral effects and is more efficient in terms of computational cost.

If we want people to trust AI systems, we need to provide the systems we create with the ability to discriminate between what humans would consider good and bad decisions. The quality of a decision should not be based only on the preferences or optimization criteria of the decision makers, but also on other properties related to the impact of the decision, such as whether it is ethical, or if it complies to constraints and priorities given by feasibility constraints or safety regulations. The CP-net formalism [2] is a convenient and expressive way to model preferences, providing an effective compact way to qualitatively model preferences over outcomes, i.e., decisions, with a combinatorial structure [3, 7]. If we wish to incorporate ethical, moral, or norms based constraints to a decision context, it means that the subjective preferences of the decision makers are not the only source of information we should consider [1, 8]. Indeed, depending on the context, we may have to consider specific ethical principles derived from an appropriate ethical theory or various laws and norms. While preferences are important, when preferences and ethical principles are in conflict, the principles should override the subjective preferences of the decision maker. Therefore, it is essential to have well founded techniques to evaluate whether preferences are compatible with a set of ethical principles, and to measure how much these preferences deviate from the ethical principles.