Offline Reinforcement Learning (RL) is a promising approach for learning optimal policies in environments where direct exploration is expensive or unfeasible. However, the adoption of such policies in practice is often challenging, as they are hard to interpret within the application context, and lack measures of uncertainty for the learned policy value and its decisions. To overcome these issues, we propose an Expert-Supervised RL (ESRL) framework which uses uncertainty quantification for offline policy learning. In particular, we have three contributions: 1) the method can learn safe and optimal policies through hypothesis testing, 2) ESRL allows for different levels of risk averse implementations tailored to the application context, and finally, 3) we propose a way to interpret ESRL's policy at every state through posterior distributions, and use this framework to compute off-policy value function posteriors. We provide theoretical guarantees for our estimators and regret bounds consistent with Posterior Sampling for RL (PSRL). Sample efficiency of ESRL is independent of the chosen risk aversion threshold and quality of the behavior policy.

Recent advances in AI research make it increasingly plausible that artificial agents with consequential real-world impact will soon operate beyond tightly controlled environments. Ensuring that these agents are not only safe but that they adhere to broader normative expectations is thus an urgent interdisciplinary challenge. Multiple fields -- notably AI Safety, AI Alignment, and Machine Ethics -- claim to contribute to this task. However, the conceptual boundaries and interrelations among these domains remain vague, leaving researchers without clear guidance in positioning their work. To address this meta-challenge, we develop a structured conceptual framework for understanding AI alignment. Rather than focusing solely on alignment goals, we introduce a taxonomy distinguishing the alignment aim (safety, ethicality, legality, etc.), scope (outcome vs. execution), and constituency (individual vs. collective). This structural approach reveals multiple legitimate alignment configurations, providing a foundation for practical and philosophical integration across domains, and clarifying what it might mean for an agent to be aligned all-things-considered.

Large language models are increasingly being integrated into applications that shape the production and discovery of societal knowledge such as search, online education, and travel planning. As a result, language models will shape how people learn about, perceive and interact with global cultures making it important to consider whose knowledge systems and perspectives are represented in models. Recognizing this importance, increasingly work in Machine Learning and NLP has focused on evaluating gaps in global cultural representational distribution within outputs. However, more work is needed on developing benchmarks for cross-cultural impacts of language models that stem from a nuanced sociologically-aware conceptualization of cultural impact or harm. We join this line of work arguing for the need of metricizable evaluations of language technologies that interrogate and account for historical power inequities and differential impacts of representation on global cultures, particularly for cultures already under-represented in the digital corpora. We look at two concepts of erasure: omission: where cultures are not represented at all and simplification i.e. when cultural complexity is erased by presenting one-dimensional views of a rich culture. The former focuses on whether something is represented, and the latter on how it is represented. We focus our analysis on two task contexts with the potential to influence global cultural production. First, we probe representations that a language model produces about different places around the world when asked to describe these contexts. Second, we analyze the cultures represented in the travel recommendations produced by a set of language model applications. Our study shows ways in which the NLP community and application developers can begin to operationalize complex socio-cultural considerations into standard evaluations and benchmarks.

Offline Reinforcement Learning (RL) is a promising approach for learning optimal policies in environments where direct exploration is expensive or unfeasible. However, the adoption of such policies in practice is often challenging, as they are hard to interpret within the application context, and lack measures of uncertainty for the learned policy value and its decisions. To overcome these issues, we propose an Expert-Supervised RL (ESRL) framework which uses uncertainty quantification for offline policy learning. In particular, we have three contributions: 1) the method can learn safe and optimal policies through hypothesis testing, 2) ESRL allows for different levels of risk averse implementations tailored to the application context, and finally, 3) we propose a way to interpret ESRL's policy at every state through posterior distributions, and use this framework to compute off-policy value function posteriors. We provide theoretical guarantees for our estimators and regret bounds consistent with Posterior Sampling for RL (PSRL). Sample efficiency of ESRL is independent of the chosen risk aversion threshold and quality of the behavior policy.

Pros: - The paper combines submodular function maximization with contextual bandit, and propose the problem of interactive submodular bandit formulation and SM-UCB solution. It includes several existing studies as special cases. Cons: - The discussion on the key RKHS condition is not enough, and its applicability for actual applications is unclear. More importantly, there is no discussion about constant B for the RKHS norm for various applications, and also the maximum information gain \gamma_t, but both of them are needed in the algorithm. The additional technical contribution is small.

In modern society, service robots are increasingly recognized for their wide range of practical applications. In large and crowded social spaces, such as museums and hospitals, these robots are required to safely move in the environment while exhibiting user-friendly behavior. Ensuring the safe and socially acceptable operation of robots in such settings presents several challenges. To enhance the social acceptance in the design process of service robots, we present a systematic analysis of requirements, categorized into functional and non-functional. These requirements are further classified into different categories, with a single requirement potentially belonging to multiple categories. Finally, considering the specific case of a receptionist robotic agent, we discuss the requirements it should possess to ensure social acceptance.

Adversarial testing of large language models (LLMs) is crucial for their safe and responsible deployment. We introduce a novel approach for automated generation of adversarial evaluation datasets to test the safety of LLM generations on new downstream applications. We call it AI-assisted Red-Teaming (AART) - an automated alternative to current manual red-teaming efforts. AART offers a data generation and augmentation pipeline of reusable and customizable recipes that reduce human effort significantly and enable integration of adversarial testing earlier in new product development. AART generates evaluation datasets with high diversity of content characteristics critical for effective adversarial testing (e.g. sensitive and harmful concepts, specific to a wide range of cultural and geographic regions and application scenarios). The data generation is steered by AI-assisted recipes to define, scope and prioritize diversity within the application context. This feeds into a structured LLM-generation process that scales up evaluation priorities. Compared to some state-of-the-art tools, AART shows promising results in terms of concept coverage and data quality.

The number of Internet-of-Things (IoT) and edge devices has exploded in the last decade (IoT000; IoT00; AGG04), providing new opportunities to transform everyday people's lives. Coupled with advances in learning technologies (ML00; ML01), these can transform how people interact with their environment. A typical machine learning workflow in sensor-based applications starts with unlabeled data. That data is visualized, featurized, and clustered in search of patterns. Typically, labels are obtained, and subsequent sample-label pairs are used to train a classifier.

Meteorological forecasting provides reliable prediction about the future weather within a given interval of time. Meteorological forecasting can be viewed as a form of hybrid diagnostic reasoning and can be mapped onto an integrated conceptual framework. The automation of the forecasting process would be helpful in a number of contexts, in particular: when the amount of data is too wide to be dealt with manually; to support forecasters education; when forecasting about underpopulated geographic areas is not interesting for everyday life (and then is out from human forecasters' tasks) but is central for tourism sponsorship. We present logic MeteoLOG, a framework that models the main steps of the reasoner the forecaster adopts to provide a bulletin. MeteoLOG rests on several traditions, mainly on fuzzy, temporal and probabilistic logics. On this basis, we also introduce the algorithm Tournament, that transforms a set of MeteoLOG rules into a defeasible theory, that can be implemented into an automatic reasoner. We finally propose an example that models a real world forecasting scenario.

We demonstrate an embodied cognitive agent that helps scientists visualize and analyze exo-planets and their host stars. The prototype is situated in a room equipped with a large display, microphones, cameras, speakers, and pointing devices. Users communicate with the agent via speech, gestures, and combinations thereof, and it responds by displaying content and generating synthesized speech.