论文检索

Jiaxin Huang

Frontier large language models are increasingly powerful though many of them are trained from vast proprietary data and intensive computes, raising barriers for academic labs and smaller institutions for exploration and improvement. In this talk, I will present a unified research agenda for breaking the resource monopoly in both post-training and serving. On the training side, I will describe label-free and even zero-data post-training pipelines that let models curate their own reasoning supervision. On the serving side, I will show how cost-aware inference can enable adaptive test-time scaling to be more efficient. Together, these components form a practical LLM system using modest data and compute resources.

Yi Ren (May) Fung

As foundation models grow in size and scope, crucial challenges remain in scaling their trustworthiness and adaptability to meet the diverse needs of individual users, as well as mitigating their risk of generating unhelpful, non-factual, or harmful content. To address this, we propose to reframe model reasoning through a unified paradigm of active knowledge grounding that coordinates different tools and modalities. First, to scale reasoning depth and creativity, we introduce the novel paradigm of Thinking with Images to encourage models to externalize intermediate structure and perform interleaved cross-modal advanced reasoning beyond text-centric cues. To further scale honesty and bridge knowledge gaps reliably, we develop one of the first vision-language deep research agents, WebWatcher, that actively gathers and verifies information from the web with enhanced fragmented reasoning capability. Ultimately, to scale effective and efficient human-AI collaboration, we propose AdaCtrl as a novel training mechanism for dynamically aligning model behavior with individual user preferences and difficulty awareness to adaptively allocate computational resources. Together, these three pillars of integrating advanced multimodal reasoning, autonomous discovery, and adaptive alignment form a foundational framework for advancing the frontier of next generation human-centric trustworthy AI systems.

Xuefeng Du

This talk surveys my research journey toward building reliable machine learning systems that behave safely and predictably in the open world. While modern machine learning models—including foundation models (FMs)—have demonstrated unprecedented capabilities, they often suffer from reliability failures under distribution shift, leading to overconfident mispredictions, hallucinated generations, or susceptibility to adversarial prompts. My research rethinks reliability not as an afterthought, but as a first-class algorithmic principle, to be optimized alongside accuracy with minimal human supervision. The talk is organized around three key threads. To respect the allotted 20-30 minutes, the first and second parts will be briefly discussed. 1. Unknown-Aware Learning via Outlier Synthesis. I introduce a class of learning algorithms that synthesize “virtual outliers” in representation or pixel space to explicitly teach models what they don’t know. This includes the VOS, NPOS, and Dream-OOD frameworks, which shape the energy landscape around in-distribution data to avoid overconfidence on OOD. 2. Learning in the Wild with Unlabeled Data. I present theoretical insights and practical algorithms for leveraging unlabeled in-the-wild data to improve reliability. This includes SAL framework, which uses a gradient-based spectral method to separate potential outliers, and SCONE, which handles semantic and covariate shifts via constrained optimization. These results turn unlabeled data contamination into a learning signal. 3. Reliable Foundation Models. I explore reliability failures in LLMs and multimodal systems. I introduce HaloScope for hallucination detection via subspace separation on LLM representations, and TSV that performs LLM latent steering for improved hallucination detection. I will also briefly cover the LLM security and alignment, which includes VLMGuard for detecting malicious prompts in vision-language models and a data-centric paradigm for AI alignment through source-aware feedback cleaning. Throughout the talk, I highlight how representation learning, data generation, and theoretical guarantees intersect to produce scalable, label-efficient reliability methods. I will also reflect on my broader vision: designing proactive and collaborative AI systems that anticipate uncertainty and support rich human-AI interaction—especially for underrepresented communities and emerging scientific domains. This talk will be accessible to a broad AAAI audience, combining foundational algorithmic insights with real-world applications and forward-looking perspectives on the future of responsible AI.

Hao-Wen Dong

In this talk, I will survey my work in three main research directions: 1) generative models for music creation, 2) AI-assisted music creation tools, and 3) multimodal generative models for content creation. In particular, I will discuss our recent work on AI-assisted video editing that explores novel machine learning models that can cut, select, and rearrange a long video into a short video. In the first TeaserGen project, we proposed a narration-centered teaser generation system that can effectively compress >30-min documentaries into

Mingyu Ding

Large multimodal models (LMMs) have revolutionized AI by demonstrating remarkable capabilities in vision, language, audio, and other domains, particularly in understanding and generalization tasks. Yet, moving beyond passive understanding to active interaction requires embodied agents, such as robots, that can harness the capabilities of AI models to act within the physical world. My core research aims to build embodied agents that reason about and interact with the physical world with human-like commonsense. Specifically, I design algorithms and representations that enable robots to perceive their environment, reason about physical properties, and plan long-horizon actions for both manipulation and locomotion. These advances are grounded in the integration of large-scale AI models with embodied control. I organize this agenda into three stages: (1) injecting actions into LMMs to form vision–language–action (VLA) models; (2) learning from human motion and contact to enrich physical reasoning; and (3) advancing whole-body robot loco-manipulation guided by LMMs toward embodied artificial general intelligence (AGI). The talk details recent advances in leveraging LMMs for robot learning, emphasizing the promise of robust generalization across diverse environments, tasks, and modalities. I will highlight contributions at the intersection of perception, reasoning, and control, and outline open challenges and future opportunities toward enabling humanoid robots that can robustly understand, interact, and collaborate with humans in complex real-world settings.

Yang Deng

Conversational AI agents are envisioned to provide social support or functional service to human users via natural language interactions. The popularity of conversational AI has grown unprecedentedly with the advent of ChatGPT, which showcases exceptional proficiency in the capabilities of context understanding and response generation with large language models (LLMs). However, typical conversational systems are built to follow instructions, which means that the conversation is led by the user, and the system simply follows the user’s instructions or intents. My research endows the conversational AI with the capabilities of creating or controlling the conversation to achieve the conversational goals by taking initiative and anticipating impacts on themselves or human users, namely Proactive Conversational AI. I will also highlight the importance of moving towards building human-centered proactive conversational AI that emphasize human needs and expectations, and that considers ethical and social implications of these agents, rather than solely focusing on technological capabilities.

Jiaming Cui

Public health and clinical decisions are intertwined. Public health crises place a high burden on healthcare facilities, forcing them to make decisions such as maintaining quality verses treating more people. Meanwhile, sub-optimal clinical decisions also cause downstream effects on communities. For ex- ample, discharging patients too early may increase disease spread. Motivated by this, we bring a data-centric perspective to bridge clinical decisions within the context of infectious diseases for public health. This work addresses multiple challenges arising from effectively utilizing rich clinical datasets and issues stemming from the complexity of disease spread dynamics in healthcare facilities. We will cover methods developed to address these challenges with better designed models to optimize disease surveillance and control policies and new techniques for end-to-end learning with mechanistic models. We will conclude by discussing emerging challenges and opportunities at the intersection of machine learning, scientific modeling, and clinical decision-making for computer scientists, epidemiologists, and computational biologists.

Grigorios Chrysos

As large language models (LLMs) scale, ensuring interpretability and privacy becomes critical. This talk addresses these interconnected challenges with novel approaches to model specialization and safety. First, we tackle the dense, distributed nature of LLM representations by casting Mixture-of-Experts (MoE) as a tensor decomposition, enabling specialized experts in a factorized space. Second, we argue that current neuron-level sparsity methods create a severe accuracy-sparsity trade-off, and we propose a paradigm shift to layer-level sparsity with the Mixture of Decoders (MxD). We explain how MxD uses tensor factorization to expand dense layers into thousands of specialized, full-rank sublayers, demonstrating how it significantly outperforms alternatives in preserving model faithfulness and performance across LLMs up to 3B parameters. Finally, we address privacy in open-weight models by proposing a scalable and certifiable algorithm that induces maximal uncertainty on protected instances, proving tight bounds that characterize the resulting privacy-utility tradeoff.

Rohan Chandra

Deploying intelligent, autonomous agents e.g. autonomous vehicles and robots, in the real world has been a longstanding goal in robotics and artificial intelligence (AI). We have already begun to witness the emergence of vacuum robots in our homes, service robots in warehouses, and even self-driving cars on our way to work. These environments are often dense, constrained, and unstructured, with heterogeneous agents, each with their own unique behaviors and objectives. While agents today are designed to navigate these environments safely, their overly conservative nature often leads to slow and jerky motion (frequent stopping and freezing), lack of social compliance (not giving way to other people, blocking doorways and intersection), and poor adaptability across diverse complex environments (failure due to sudden accidents e.g. liquid spills). In other words, these robots often fail to capture the essence of human-like autonomy, which involves the ability to take calculated risks, even in complex environments. In this talk, I will describe my vision for a paradigm shift in the way intelligent physical agents navigate highly dense, heterogeneous, constrained, and unstructured environments using human-like autonomy.

Chirag Agarwal

Machine learning models have become ubiquitous in the last decade, and with their increasing use in critical applications (e.g., healthcare, financial systems, and crime forecasting), it is vital to ensure that ML developers and practitioners understand and trust their decisions. This problem has become paramount in the era of frontier models, which are developed by training billion-parameter models on broad, uncurated datasets and extensive computing. In this talk, we will first explore the (un)reliability of existing multimodal explainability techniques in large language and multimodal models and understand the robustness and safety implications of Mechanistic Interpretability tools. Next, we will delve into two complementary threads: i) domain-specific safety and related trustworthy evaluation that surfaces risks missed by generic red-teaming, focusing on multilingual and distribution-shifted settings; and ii) methods that explicitly train and assess reasoning in medical LLMs.

Mayank Vatsa, Aparna Bharati, Richa Singh

The architectural blueprint of today’s leading text-to-image models contains a fundamental flaw: an inability to handle logical composition. This survey investigates this breakdown across three core primitives—negation, counting, and spatial relations. Our analysis reveals a dramatic performance collapse: models that are accurate on single primitives fail precipitously when these are combined, exposing severe interference. We trace this failure to three key factors. First, training data show a near-total absence of explicit negations. Second, continuous attention architectures are fundamentally unsuitable for discrete logic. Third, evaluation metrics reward visual plausibility over constraint satisfaction. By analyzing recent benchmarks and methods, we show that current solutions and simple scaling cannot bridge this gap. Achieving genuine compositionality, we conclude, will require fundamental advances in representation and reasoning rather than incremental adjustments to existing architectures.

Warut Suksompong, Nicholas Teh

Voting is one of the most prominent applications of preference aggregation and computational social choice. While much of the literature focuses on models involving discrete candidates, there has been a growing interest in voting over divisible resources, such as budget, space, and time. In this survey, we review existing work on voting in divisible settings, including fundamental models of budget aggregation, fair mixing, and cake sharing. We also establish connections among these models, highlight unifying themes across different frameworks, and suggest directions for future research.

Aarti Singh

From expert AI systems of the 1970s to self-supervised systems of the 2020s, the pendulum of AI development has swung from heavy reliance on human feedback to no or minimal reliance in the last 50 years. Self-supervised approaches have contributed significantly to the success and scalable development of AI. However, today we are at a tipping point where the future of AI, and whether so-ciety ends up benefiting from this technology in the long run, depends critically on the subsequent AI develop-ment aligning with human goals and values. Realizing this, there has been ramping up of efforts to align AI models with human expectations and values. Human feedback, however, remains limited and difficult to elicit. Thus, a key question lingers – how can we scale up alignment of AI systems with individual expectations and societal norms? This talk and paper provides an overview and perspective on efforts at answering this question.

Ted Senator

AI is currently in the midst of a boom, mostly due to the success and predominance of large language models and associated models for other perceptual tasks such as computer vision. Yet AI has experienced several booms and busts over the past 75 years. While the booms are driven by commercial potential, the following busts affect not only commercial investment but also research funding and trends. This paper examines the expert systems boom of the 1980s and the following AI winter, identifies similarities, analogs, and differences with the current boom, and projects potential outcomes and directions for AI research that may follow when the current enthusiasm wanes based on these similarities, analogs, and differences. The presentation is distinct from currently active discussions and debates about the potential and limitations of large models such as whether problems such as hallucination will be solved, whether they can reason, or whether they will achieve AGI; rather, it examines previous AI techniques and how they evolved once their capabilities and limitations became well understood.

Soujanya Poria, Navonil Majumder, Chia-Yu Hung, Amir Ali Bagherzadeh, Chuan Li, Kenneth Kwok, Ziwei Wang, Cheston Tan, Jiajun Wu, David Hsu

Due to their ability of follow natural language instructions, vision-language-action (VLA) models are increasingly preva- lent in the embodied AI arena, following the widespread suc- cess of their precursors—LLMs and VLMs. In this paper, we discuss 10 principal milestones in the ongoing develop- ment of VLA models—multimodality, reasoning, data, eval- uation, cross-robkot action generalization, efficiency, whole- body coordination, safety, agents, and coordination with hu- mans. Furthermore, we discuss the emerging trends of us- ing spatial understanding, modeling world dynamics, post training, and data synthesis—all aiming to reach these mile- stones. Through these discussions, we hope to bring attention to the research avenues that may accelerate the development of VLA models into wider acceptability.

David Poole

Artificial intelligence seems to be taking over the world with systems that model pixels, words, and phonemes. The world is arguably made up, not of pixels, words, and phonemes but of entities (objects, things, including events) with properties and relations among them. Surely we should model these, not the perception or description of them. You might suspect that concentrating on modeling words and pixels is because all of the (valuable) data in the world is in terms of text and images. If you look into almost any company you will find their most valuable data is in spreadsheets, databases and other relational formats. These are not the form that are studied in introductory machine learning, but are full of product numbers, student numbers, transaction numbers and other identifiers that can't be interpreted naively as numbers. The field that studies this sort of data has various names including relational learning, statistical relational AI, and many others. This paper explains why relational learning is not taking over the world -- except in a few cases with restricted relations -- and what needs to be done to bring it to it's rightful prominence.

Roberto Navigli

Word Sense Disambiguation (WSD) has been a central challenge since the earliest proposals for Machine Translation (MT), most famously Weaver's 1949 memorandum. Classical systems treated WSD as an explicit task, grounded in lexical resources and annotated data. Recently, however, Large Language Models (LLMs) have blurred the boundary between disambiguation and general language understanding, leading some to suggest that WSD might be obsolete. This paper surveys the role of WSD in the LLM era, drawing on recent studies of encoder-based sense separation and disambiguation, and decoder-based definition selection and generation, as well as multilingual evaluation. Closed-source instruction-tuned LLMs now achieve performance comparable to specialized WSD systems, yet systematic weaknesses remain: non-predominant senses are often misclassified and disambiguation biases in MT persist. We argue that WSD is not "dead" but redefined as a diagnostic lens for assessing lexical-semantic competence, robustness, and interpretability in LLMs.

Pavel Naumov

Traditionally, the goal of mechanism design was to promote socially desirable behaviour of rational agents, to achieve fairness, or to promote efficiency. I would like to suggest a new subfield of mechanism design, Responsible Mechanism Design, focused on achieving individual accountability of agents for their contributions to the outcome of collective decisions.

Sridhar Mahadevan

We propose a new theoretical foundation for artificial intelligence (AI) and machine learning (ML), building on ideas in pure mathematics relating to categories and functors. This paper builds on our AAAI 2025 tutorial Thinking with Functors: Category Theory for A(G)I, which provides background material. In addition, our recent papers on intuitionistic j-do calculus in Topos Causal Models} and GAIA: Categorical Foundations of Generative AI, illustrate how to generalize well-known formalisms in AI, such as causal inference and deep learning, to a category-theoretic setting.

Chuanyi Li, Yi Feng, Vincent Ng

Dual-system theory distinguishes between fast, intuitive System 1 and slow, deliberative System 2. While this dichotomy describes many forms of reasoning, it oversimplifies the reality of expert legal reasoning. Legal reasoning is not merely a process of slow, logical deliberation. It is intrinsically normative, embedding precedent analysis, statutory interpretation, policy balancing, and social values. This paper envisions a reasoning architecture for legal reasoning, System L (Legal System 2), which extends traditional System 2 by integrating domain-specific normative frameworks in a structured manner. Using the IRAC (Issue–Rule–Application–Conclusion) structure as a backbone model, System L represents a blueprint for the next generation of cognitive and AI systems capable of human-like legal reasoning.