Generations in Dialogue: Bridging Perspectives in AI is a podcast from AAAI featuring thought-provoking discussions between AI experts, practitioners, and enthusiasts from different age groups and backgrounds. Each episode delves into how generational experiences shape views on AI, exploring the challenges, opportunities, and ethical considerations that come with the advancement of this transformative technology. In the second episode of this new series from AAAI, host Ella Lan chats to Professor Andreea Bobu about choosing her research direction, working with humans in-the-loop, things to consider when working with data, system design challenges, the gap between what we think we're programming and what we've actually programmed, privacy and personalisation, and advice for early-career researchers interested in human-centric AI. Andreea Bobu is an Assistant Professor at MIT and leads the Collaborative Learning and Autonomy Research (CLEAR) Lab, where she develops autonomous agents that learn to act for, with, and around people. Her research focuses on aligning robot behavior with human expectations by studying how agents can acquire the right supervision--whether from direct human input or priors--build shared task representations with users, and address misalignment from differing human models.

Sam Altman says the one-person billion-dollar company is coming. Maybe I could be that person--if only I could get my colleagues to shut up and stop lying. One day a couple months ago, in the middle of lunch, I glanced at my phone and was puzzled to see my colleague Ash Roy calling. In and of itself it might not have seemed strange to get a call from Ash: He's the CTO and chief product officer of HurumoAI, a startup I cofounded last summer. We were in the middle of a big push to get our software product, an AI agent application, into beta. There was plenty to discuss. "Hey there," he said, when I picked up. He was calling, he said, because I'd requested a progress report on the app from Megan. "I've been good," I said, chewing my grilled cheese.

Counterfactual regret minimization (CFR) is a family of algorithms for effectively solving imperfect-information games. To enhance CFR's applicability in large games, researchers use neural networks to approximate its behavior. However, existing methods are mainly based on vanilla CFR and struggle to effectively integrate more advanced CFR variants. In this work, we propose an efficient model-free neural CFR algorithm, overcoming the limitations of existing methods in approximating advanced CFR variants. At each iteration, it collects variance-reduced sampled advantages based on a value network, fits cumulative advantages by bootstrapping, and applies discounting and clipping operations to simulate the update mechanisms of advanced CFR variants. Experimental results show that, compared with model-free neural algorithms, it exhibits faster convergence in typical imperfect-information games and demonstrates stronger adversarial performance in a large poker game.

The sim-to-real gap, where agents trained in a simulator face significant performance degradation during testing, is a fundamental challenge in reinforcement learning. Extansive works adopt the framework of distributionally robust RL, to learn a policy that acts robustly under worst case environment shift. Within this framework, our objective is to devise algorithms that are sample efficient with interactive data collection and large state spaces. By assuming d-rectangularity of environment dynamic shift, we identify a fundamental hardness result for learning in online Markov game, and address it by adopting minimum value assumption. Then, a novel least square value iteration type algorithm, DR-CCE-LSI, with exploration bonus devised specifically for multiple agents, is proposed to find an \episilon-approximate robust Coarse Correlated Equilibrium(CCE). To obtain sample efficient learning, we find that: when the feature mapping function satisfies certain properties, our algorithm, DR-CCE-LSI, is able to achieve ε-approximate CCE with a regret bound of O{dHmin{H,1/min{σ_i}}\sqrt{K}}, where K is the number of interacting episodes, H is the horizon length, d is the feature dimension, and \simga_i represents the uncertainty level of player i. Our work introduces the first sample-efficient algorithm for this setting, matches the best result so far in single agent setting, and achieves minimax optimalsample complexity in terms of the feature dimension d. Meanwhile, we also conduct simulation study to validate the efficacy of our algorithm in learning a robust equilibrium.

Weakly electric fish, like Gnathonemus petersii, use a remarkable electrical modality for active sensing and communication, but studying their rich electrosensing and electrocommunication behavior and associated neural activity in naturalistic settings remains experimentally challenging. Here, we present a novel biologically-inspired computational framework to study these behaviors, where recurrent neural network (RNN) based artificial agents trained via multi-agent reinforcement learning (MARL) learn to modulate their electric organ discharges (EODs) and movement patterns to collectively forage in virtual environments. Trained agents demonstrate several emergent features consistent with real fish collectives, including heavy tailed EOD interval distributions, environmental context dependent shifts in EOD interval distributions, and social interaction patterns like freeloading, where agents reduce their EOD rates while benefiting from neighboring agents' active sensing. A minimal two-fish assay further isolates the role of electro-communication, showing that access to conspecific EODs and relative dominance jointly shape foraging success. Notably, these behaviors emerge through evolution-inspired rewards for individual fitness and emergent inter-agent interactions, rather than through rewarding agents explicitly for social interactions. Our work has broad implications for the neuroethology of weakly electric fish, as well as other social, communicating animals in which extensive recordings from multiple individuals, and thus traditional data-driven modeling, are infeasible.

Multi-agent systems with smaller language models (SLMs) present a viable alternative to single agent systems powered by large language models (LLMs) for addressing complex problems. In this work, we study how these alternatives compare in terms of both effectiveness and efficiency. To study this trade-off, we instantiate single and multi-agent systems for the complex problems in the AppWorld environment using different sized language models. We find that difficulties with long-trajectory learning in smaller language models (SLMs) limit their performance. Even when trained for specialized roles, SLMs fail to learn all subtasks effectively. To address this issue, we introduce a simple progressive sub-task training strategy, which introduces new sub-tasks progressively in each training epoch. We find that this novel strategy, analogous to instance level curriculum learning, consistently improves the effectiveness of multi-agents at all configurations. Our Pareto analysis shows that fine-tuned multi-agent systems yield better effectiveness-efficiency trade-offs. Additional ablations and analyses shows the importance of our progressive training strategy and its ability to reduce subtask error rates.

Identifying whether two product listings refer to the same Stock Keeping Unit (SKU) is a persistent challenge in ecommerce, especially when explicit identifiers are missing and product names vary widely across platforms. Rule based heuristics and keyword similarity often misclassify products by overlooking subtle distinctions in brand, specification, or bundle configuration. To overcome these limitations, we propose Question to Knowledge (Q2K), a multi agent framework that leverages Large Language Models (LLMs) for reliable SKU mapping. Q2K integrates: (1) a Reasoning Agent that generates targeted disambiguation questions, (2) a Knowledge Agent that resolves them via focused web searches, and (3) a Deduplication Agent that reuses validated reasoning traces to reduce redundancy and ensure consistency. A human in the loop mechanism further refines uncertain cases. Experiments on real world consumer goods datasets show that Q2K surpasses strong baselines, achieving higher accuracy and robustness in difficult scenarios such as bundle identification and brand origin disambiguation. By reusing retrieved reasoning instead of issuing repeated searches, Q2K balances accuracy with efficiency, offering a scalable and interpretable solution for product integration.

In this paper, we present SOCIA-$\nabla$, an end-to-end, agentic framework that treats simulator construction asinstance optimization over code within a textual computation graph. Specialized LLM-driven agents are embedded as graph nodes, and a workflow manager executes a loss-driven loop: code synthesis -> execution -> evaluation -> code repair. The optimizer performs Textual-Gradient Descent (TGD), while human-in-the-loop interaction is reserved for task-spec confirmation, minimizing expert effort and keeping the code itself as the trainable object. Across three CPS tasks, i.e., User Modeling, Mask Adoption, and Personal Mobility, SOCIA-$\nabla$ attains state-of-the-art overall accuracy. By unifying multi-agent orchestration with a loss-aligned optimization view, SOCIA-$\nabla$ converts brittle prompt pipelines into reproducible, constraint-aware simulator code generation that scales across domains and simulation granularities. We will release the code soon.

Though Large Vision-Language Models (LVLMs) are being actively explored in medicine, their ability to conduct complex real-world telemedicine consultations combining accurate diagnosis with professional dialogue remains underexplored. This paper presents 3MDBench (Medical Multimodal Multi-agent Dialogue Benchmark), an open-source framework for simulating and evaluating LVLM-driven telemedical consultations. 3MDBench simulates patient variability through temperament-based Patient Agent and evaluates diagnostic accuracy and dialogue quality via Assessor Agent. It includes 2996 cases across 34 diagnoses from real-world telemedicine interactions, combining textual and image-based data. The experimental study compares diagnostic strategies for widely used open and closed-source LVLMs. We demonstrate that multimodal dialogue with internal reasoning improves F1 score by 6.5% over non-dialogue settings, highlighting the importance of context-aware, information-seeking questioning. Moreover, injecting predictions from a diagnostic convolutional neural network into the LVLM's context boosts F1 by up to 20%. Source code is available at https://github.com/univanxx/3mdbench.

Common knowledge is crucial for safe group coordination. In its absence, humans must rely on shared knowledge, which is inherently limited in depth and therefore prone to coordination failures, because any finite-order knowledge attribution allows for an even higher order attribution that may change what is known by whom. In three separate experiments involving 802 participants, we investigate the extent to which humans can differentiate between common knowledge and nth-order shared knowledge. We designed a two-person coordination game with imperfect information to simplify the recursive game structure and higher-order uncertainties into a relatable everyday scenario. In this game, coordination for the highest payoff requires a specific fact to be common knowledge between players. However, this fact cannot become common knowledge in the game. The fact can at most be nth-order shared knowledge for some n. Our findings reveal that even at quite shallow depths of shared knowledge (low values of n), players behave as though they possess common knowledge, and claim similar levels of certainty in their actions, despite incurring significant penalties when falsely assuming guaranteed coordination. We term this phenomenon 'the curse of shared knowledge'. It arises either from the players' inability to distinguish between higher-order shared knowledge and common knowledge, or from their implicit assumption that their co-player cannot make this distinction.