Despite the significant strides made by generative AI in just a few short years, its future progress is constrained by the challenge of building modular and robust systems. This capability has been a cornerstone of past technological revolutions, which relied on combining components to create increasingly sophisticated and reliable systems. Cars, airplanes, computers, and software consist of components-such as engines, wheels, CPUs, and libraries-that can be assembled, debugged, and replaced. A key tool for building such reliable and modular systems is specification: the precise description of the expected behavior, inputs, and outputs of each component. However, the generality of LLMs and the inherent ambiguity of natural language make defining specifications for LLM-based components (e.g., agents) both a challenging and urgent problem. In this paper, we discuss the progress the field has made so far-through advances like structured outputs, process supervision, and test-time compute-and outline several future directions for research to enable the development of modular and reliable LLM-based systems through improved specifications.

With the growing interest in underwater exploration and monitoring, Autonomous Underwater Vehicles (AUVs) have become essential. The recent interest in onboard Deep Learning (DL) has advanced real-time environmental interaction capabilities relying on efficient and accurate vision-based DL models. However, the predominant use of sonar in underwater environments, characterized by limited training data and inherent noise, poses challenges to model robustness. This autonomy improvement raises safety concerns for deploying such models during underwater operations, potentially leading to hazardous situations. This paper aims to provide the first comprehensive overview of sonar-based DL under the scope of robustness. It studies sonar-based DL perception task models, such as classification, object detection, segmentation, and SLAM. Furthermore, the paper systematizes sonar-based state-of-the-art datasets, simulators, and robustness methods such as neural network verification, out-of-distribution, and adversarial attacks. This paper highlights the lack of robustness in sonar-based DL research and suggests future research pathways, notably establishing a baseline sonar-based dataset and bridging the simulation-to-reality gap.

Gravitational-wave observatories like LIGO are large-scale, terrestrial instruments housed in infrastructure that spans a multi-kilometer geographic area and which must be actively controlled to maintain operational stability for long observation periods. Despite exquisite seismic isolation, they remain susceptible to seismic noise and other terrestrial disturbances that can couple undesirable vibrations into the instrumental infrastructure, potentially leading to control instabilities or noise artifacts in the detector output. It is, therefore, critical to characterize the seismic state of these observatories to identify a set of temporal patterns that can inform the detector operators in day-to-day monitoring and diagnostics. On a day-to-day basis, the operators monitor several seismically relevant data streams to diagnose operational instabilities and sources of noise using some simple empirically-determined thresholds. It can be untenable for a human operator to monitor multiple data streams in this manual fashion and thus a distillation of these data-streams into a more human-friendly format is sought. In this paper, we present an end-to-end machine learning pipeline for features-based multivariate time series clustering to achieve this goal and to provide actionable insights to the detector operators by correlating found clusters with events of interest in the detector.

Retrieval-Augmented Generation (RAG) systems have demonstrated remarkable potential as question answering systems in the K-12 Education domain, where knowledge is typically queried within the restricted scope of authoritative textbooks. However, the discrepancy between textbooks and the parametric knowledge in Large Language Models (LLMs) could undermine the effectiveness of RAG systems. To systematically investigate the robustness of RAG systems under such knowledge discrepancies, we present EduKDQA, a question answering dataset that simulates knowledge discrepancies in real applications by applying hypothetical knowledge updates in answers and source documents. EduKDQA includes 3,005 questions covering five subjects, under a comprehensive question typology from the perspective of context utilization and knowledge integration. We conducted extensive experiments on retrieval and question answering performance. We find that most RAG systems suffer from a substantial performance drop in question answering with knowledge discrepancies, while questions that require integration of contextual knowledge and parametric knowledge pose a challenge to LLMs.

Interpretability methods seek to understand language model representations, yet the outputs of most such methods -- circuits, vectors, scalars -- are not immediately human-interpretable. In response, we introduce LatentQA, the task of answering open-ended questions about model activations in natural language. Towards solving LatentQA, we propose Latent Interpretation Tuning (LIT), which finetunes a decoder LLM on a dataset of activations and associated question-answer pairs, similar to how visual instruction tuning trains on question-answer pairs associated with images. We use the decoder for diverse reading applications, such as extracting relational knowledge from representations or uncovering system prompts governing model behavior. Our decoder also specifies a differentiable loss that we use to control models, such as debiasing models on stereotyped sentences and controlling the sentiment of generations. Finally, we extend LatentQA to reveal harmful model capabilities, such as generating recipes for bioweapons and code for hacking.

We introduce Concept Bottleneck Protein Language Models (CB-pLM), a generative masked language model with a layer where each neuron corresponds to an interpretable concept. Our architecture offers three key benefits: i) Control: We can intervene on concept values to precisely control the properties of generated proteins, achieving a 3 times larger change in desired concept values compared to baselines. ii) Interpretability: A linear mapping between concept values and predicted tokens allows transparent analysis of the model's decision-making process. iii) Debugging: This transparency facilitates easy debugging of trained models. Our models achieve pre-training perplexity and downstream task performance comparable to traditional masked protein language models, demonstrating that interpretability does not compromise performance. While adaptable to any language model, we focus on masked protein language models due to their importance in drug discovery and the ability to validate our model's capabilities through real-world experiments and expert knowledge. We scale our CB-pLM from 24 million to 3 billion parameters, making them the largest Concept Bottleneck Models trained and the first capable of generative language modeling.

The increasing rates of global urbanization and vehicle usage are leading to a shift of mobility to the third dimension-through Advanced Air Mobility (AAM)-offering a promising solution for faster, safer, cleaner, and more efficient transportation. As air transportation continues to evolve with more automated and autonomous systems, advancements in AAM require a deep understanding of human-computer interaction and human-AI collaboration to ensure safe and effective operations in complex urban and regional environments. There has been a significant increase in publications regarding these emerging applications; thus, there is a need to review developments in this area. This paper comprehensively reviews the current state of research on human-computer interaction and human-AI collaboration in AAM. Specifically, we focus on AAM applications related to the design of human-machine interfaces for various uses, including pilot training, air traffic management, and the integration of AI-assisted decision-making systems with immersive technologies such as extended, virtual, mixed, and augmented reality devices. Additionally, we provide a comprehensive analysis of the challenges AAM encounters in integrating human-computer frameworks, including unique challenges associated with these interactions, such as trust in AI systems and safety concerns. Finally, we highlight emerging opportunities and propose future research directions to bridge the gap between human factors and technological advancements in AAM.

It took over 15 years and more than 1,000 remotely operated vehicle (ROV) expeditions, but researchers aboard the NOAA Ocean Exploration Trust's Nautilus finally spotted their research vessel's namesake in the wild. On December 3, operators of the ship's Hercules ROV located four specimens of Palau nautilus (Nautilus belauensis) during the Nautilus Exploration Program's ongoing, 17-day survey in the Palau National Marine Sanctuary. While the team recorded these particular examples swimming 220-to-375 meters (roughly 721-to-1,230 ft) below the Pacific Ocean's surface, the pelagic marine mollusk cephalopods can survive at depths approaching 2,500 feet. Their spiral-shelled bodies belong to one of Earth's oldest families of animals, with fossil records indicating the squid relatives have changed comparatively little even after nearly 500 million years. Although their sight is limited due to rudimentary eyes that lack solid lenses, nine known nautilus species instead rely heavily on their olfactory senses to find food and mates.

In recent years, large language models (LLMs) have been widely adopted in political science tasks such as election prediction, sentiment analysis, policy impact assessment, and misinformation detection. Meanwhile, the need to systematically understand how LLMs can further revolutionize the field also becomes urgent. In this work, we--a multidisciplinary team of researchers spanning computer science and political science--present the first principled framework termed Political-LLM to advance the comprehensive understanding of integrating LLMs into computational political science. Specifically, we first introduce a fundamental taxonomy classifying the existing explorations into two perspectives: political science and computational methodologies. In particular, from the political science perspective, we highlight the role of LLMs in automating predictive and generative tasks, simulating behavior dynamics, and improving causal inference through tools like counterfactual generation; from a computational perspective, we introduce advancements in data preparation, fine-tuning, and evaluation methods for LLMs that are tailored to political contexts. We identify key challenges and future directions, emphasizing the development of domain-specific datasets, addressing issues of bias and fairness, incorporating human expertise, and redefining evaluation criteria to align with the unique requirements of computational political science. Political-LLM seeks to serve as a guidebook for researchers to foster an informed, ethical, and impactful use of Artificial Intelligence in political science. Our online resource is available at: http://political-llm.org/. Corresponding authors: Yushun Dong (yd24f@fsu.edu) is with the Department of Computer Science, Florida State University; Yue Zhao (yzhao010@usc.edu) is with the Department of Computer Science, University of Southern California; Fred Gui (pgui@lsu.edu) is with the Department of Political Science, Louisiana State University; Catherine Chen (catherinechen@lsu.edu) is with the Manship School of Mass Communication and the Department of Political Science, Louisiana State University.

The remarkable performance of large language models (LLMs) in content generation, coding, and common-sense reasoning has spurred widespread integration into many facets of society. However, integration of LLMs raises valid questions on their reliability and trustworthiness, given their propensity to generate hallucinations: plausible, factually-incorrect responses, which are expressed with striking confidence. Previous work has shown that hallucinations and other non-factual responses generated by LLMs can be detected by examining the uncertainty of the LLM in its response to the pertinent prompt, driving significant research efforts devoted to quantifying the uncertainty of LLMs. This survey seeks to provide an extensive review of existing uncertainty quantification methods for LLMs, identifying their salient features, along with their strengths and weaknesses. We present existing methods within a relevant taxonomy, unifying ostensibly disparate methods to aid understanding of the state of the art. Furthermore, we highlight applications of uncertainty quantification methods for LLMs, spanning chatbot and textual applications to embodied artificial intelligence applications in robotics. We conclude with open research challenges in uncertainty quantification of LLMs, seeking to motivate future research.