We propose—somewhat tongue-in-cheek, yet with serious implications—a new test for artificial intelligence: the ability to watch a 90-minute episode of the long-running German crime drama Tatort, and to explain every relevant detail. This involves reconstructing the evolving social network of characters, identifying their beliefs, desires, and intentions, and, crucially, determining who committed the crime. We argue that this task integrates narrative understanding, common-sense reasoning, social cognition, and theory of mind—and thus provides a uniquely challenging benchmark for AI.
论文检索
输入标题、作者或关键词，从 100,903 篇学术成果中精准定位
100,903篇论文
第 152 / 5046 页
Recent research highlights the lack of reliability of large language models (LLMs) in tasks requiring complex reasoning. While they can produce impressively fluent text in response to prompts, they can fail on basic reasoning skills, such as recognizing that left is the opposite of right. They struggle even more with grounding such concepts in real-world contexts involving perception and action. Addressing real-world problems, however, typically requires models composed of multiple interdependent learners, with strong capabilities for composition and reasoning. In this talk, I will discuss the reasoning challenges of LLMs and discuss how symbolic representations can enhance neural models by enabling Spatial and Compositional Reasoning over complex linguistic structures, grounding language in visual perception, integrating multiple modalities, and dealing with uncertainty. I will overview recent research in Neurosymbolic (NeSy) modeling and emphasize the need for community-driven libraries to advance this direction. As part of this effort, I will introduce the DomiKnowS framework developed by my team, which combines symbolic and sub-symbolic representations to tackle complex, AI-complete problems, integrating symbolic and logical knowledge seamlessly into deep models and LLMs through a range of underlying algorithms.
As advances in artificial intelligence (AI) and machine learning (ML) continue to transform commercial applications, the scientific community is increasingly eager to harness AI/ML’s power to accelerate modeling and discovery. However, purely data-driven AI methods often lack interpretability, generalizability, and consistency with established scientific principles. Conversely, traditional process-based models embody deep scientific knowledge but suffer from limited scalability or incomplete representation of complex systems. Knowledge-guided machine learning (KGML) offers a promising path forward by integrating scientific knowledge with data-driven approaches to produce AI models that are robust, trustworthy, and capable of advancing both AI and science. This talk summarizes the foundations of KGML, outlines a taxonomy for organizing research efforts, and highlights emerging opportunities for broad scientific impact.
Large language models (LLMs) have become indispensable, but the most celebrated efficiency methods---mixture-of-experts (MoE), speculative decoding, and complex retrieval-augmented generation (RAG)---were built for hyperscale providers with vast infrastructure and elite teams. Outside that context, their benefits collapse into overhead, fragility, and wasted carbon. The result is that a handful of Big Tech companies benefit, while thousands of hospitals, schools, governments, and enterprises are left without viable options. We argue that the next frontier is not greater sophistication at scale, but robust simplicity: efficiency that thrives under modest resources and minimal expertise. We propose a new research agenda: retrofitting pretrained models with more efficient architectures without retraining, inventing lightweight fine-tuning that preserves alignment, making reasoning economical despite long chains of thought, enabling dynamic knowledge management without heavy RAG pipelines, and adopting Overhead-Aware Efficiency (OAE) as a standard benchmark. By redefining efficiency to include adoption cost, sustainability, and fairness, we can democratize LLM deployment---ensuring that optimization reduces inequality and carbon waste rather than amplifying them.
Recent advances in generative AI for music have achieved remarkable fidelity and stylistic diversity, yet these systems often fail to align with nuanced human preferences due to the specific loss functions they use. This paper advocates for the systematic application of preference alignment techniques to music generation, addressing the fundamental gap between computational optimization and human musical appreciation. Drawing on recent breakthroughs including MusicRL's large-scale preference learning, multi-preference alignment frameworks like diffusion-based preference optimization in DiffRhythm+, and inference-time optimization techniques like Text2midi-InferAlign, we discuss how these techniques can address music's unique challenges: temporal coherence, harmonic consistency, and subjective quality assessment. We identify key research challenges including scalability to long-form compositions, reliability amongst others in preference modelling. Looking forward, we envision preference-aligned music generation enabling transformative applications in interactive composition tools and personalized music services. This work calls for sustained interdisciplinary research combining advances in machine learning, music-theory to create music AI systems that truly serve human creative and experiential needs.
In the Multi-Agent Path Finding (MAPF) problem, the aim is to find collision free paths for multiple agents. MAPF has many practical applications and has spawned massive research interest in the past two decades. Most MAPF research assumed that every agent is assigned a target it must reach. This assumption often does not hold in several key applications such as automated warehouses and parking lots, where some agents are assigned targets to reach, and others, denoted as unassigned agents, can either stay idle or move to clear the way for the assigned agents. In this paper we introduce this important problem, explain its uniqueness and encourage the entire community to work on it.
Bandit algorithms and Large Language Models (LLMs) have emerged as powerful tools in artificial intelligence, each addressing distinct yet complementary challenges in decision-making and natural language processing. This survey explores the synergistic potential between these two fields, highlighting how bandit algorithms can enhance the performance of LLMs and how LLMs, in turn, can provide novel insights for improving bandit-based decision-making. We first examine the role of bandit algorithms in optimizing LLM fine-tuning, prompt engineering, and adaptive response generation, focusing on their ability to balance exploration and exploitation in large-scale learning tasks. Subsequently, we explore how LLMs can augment bandit algorithms through advanced contextual understanding, dynamic adaptation, and improved policy selection using natural language reasoning. By providing a comprehensive review of existing research and identifying key challenges and opportunities, this survey aims to bridge the gap between bandit algorithms and LLMs, paving the way for innovative applications and interdisciplinary research in AI.
The Shepherd Test: How Will Super Intelligent Agents Balance Care and Control in Asymmetric Relationships?
PDF ↗This paper introduces the Shepherd Test, a new conceptual test for assessing the moral and relational dimensions of superintelligent artificial agents. The test is inspired by human interactions with animals, where ethical considerations about care, manipulation, and consumption arise in contexts of asymmetric power and self-preservation. We argue that AI crosses an important, and potentially dangerous, threshold of intelligence when it exhibits the ability to manipulate, nurture, and instrumentally use less intelligent agents, while also managing its own survival and expansion goals. This includes the ability to weigh moral trade-offs between self-interest and the well-being of subordinate agents. The Shepherd Test thus challenges traditional AI evaluation paradigms by emphasizing moral agency, hierarchical behavior, and complex decision-making under existential stakes. We argue that this shift is critical for advancing AI governance, particularly as AI systems become increasingly integrated into multi-agent environments. We conclude by identifying key research directions, including the development of simulation environments for testing moral behavior in AI, and the formalization of ethical manipulation within multi-agent systems.
Current mainstream AI, at least as presented in media and measured by the number of people involved and papers published, is mainly about big data, deep learning, and recently trendy large language models. All these are techniques that are data-driven, model-free, and number-crunching. Their immense success in some areas, such as computer vision and natural language processing, started the next hype in the era of AI, which brings a question whether neural approaches, after being dismissed at the beginning of the AI era, finally conquered the world of AI and proved applicable to every problem. A deeper look at these new techniques shows they have similar issues as the old-fashioned AI techniques in the past: brittleness, making strange mistakes, and being highly dependent on data used for training. Moreover, there are problems with the explainability of results and no guarantees provided, which is a crucial issue in some application areas. In this paper, we look at core principles of the neural ML techniques, that is, being data-driven rather than knowledge-based and being model-free rather than model-based, and we argue that symbolic knowledge models can still contribute to the design of trustworthy and explainable AI systems. Specifically, we focus on hierarchical reasoning, namely hierarchical planning, which is useful for highly complex problems but is not addressed by current neural models. We propose a research plan consisting of solving specific problems in hierarchical planning as an example of a knowledge-intensive approach to problem-solving. We show close connections between these problems that allow a smooth transition between solving techniques used to solve these problems. We also propose an ultimate goal of this endeavor, that is, autonomous construction of hierarchical planning models, that addresses the crucial problem of knowledge-based approaches -- how to obtain a formal model (extract knowledge from data).
A widespread social sentiment suggests our world operates like a "makeshift world" —a system rife with hidden incompetence. Is this perception an inevitable outcome of our information ecosystem? This paper presents a formal mathematical theory to answer this question affirmatively. We model belief dynamics as a system of interacting agents governed by two operators: (1) an Attentional Update Operator formalizing how negatively biased information is assimilated, and (2) a Social Aggregation Operator modeling belief fusion over a network. Our main contribution is a rigorous proof: under minimal systemic negative bias and standard network connectivity, the collective belief system is a contraction mapping, guaranteed to converge to a unique pessimistic fixed-point that perceives the world as incompetent, regardless of objective truth. This work establishes a mathematical foundation for understanding systemic perceptual biases with applications to platform design and policy.
The recent proliferation of large language models (LLMs) holds the potential to revolutionize healthcare, with strong capabilities in diverse medical tasks. Yet, deploying LLMs in high-stakes healthcare settings requires rigorous verification and validation to understand any potential harm. This paper investigates the reliability and viability of using medical knowledge graphs (KGs) for the automated factuality evaluation of LLM-generated responses. To ground this investigation, we introduce FAITH, a framework designed to systematically probe the strengths and limitations of this KG-based approach. FAITH operates without reference answers by decomposing responses into atomic claims, linking them to a medical KG, and scoring them based on evidence paths. Experiments on diverse medical tasks with human subjective evaluations demonstrate that KG-grounded evaluation achieves considerably higher correlations with clinician judgments and can effectively distinguish LLMs with varying capabilities. It is also robust to textual variances. The inherent explainability of its scoring can further help users understand and mitigate the limitations of current LLMs. We conclude that while limitations exist, leveraging KGs is a prominent direction for automated factuality assessment in healthcare.
Large language models (LLMs) have achieved remarkable success in many domains, but concerns about data quality and privacy are growing. Federated Learning (FL) offers a privacy-preserving solution by training a model on local clients without sharing data. However, the impact of biased private data on LLMs fine-tuned through FL remains understudied. This work investigates how client-side biased data affects the global model during federated fine-tuning of LLMs. We simulate realistic scenarios where some clients possess datasets containing social biases (stereotypes, discriminatory language) while others have clean data through extensive experiments with popular FL algorithms (FedAvg, FedAdam and FedProx) and popular LLMs (LLaMA, Mistral, Phi-3 and Gemma) across datasets with varying bias proportions (33%, 66%, 100%). Our findings reveal that 1) FedAdam consistently shows the lowest bias propagation, reducing CrowS-Pairs scores by up to 15% compared to FedAvg; 2) Even small amounts of biased data (33%) can significantly influence global model bias; 3) Mixed biased and neutral data distributions lead to 5%-7% higher bias scores than segregated distributions. Additionally, we propose Bias-Aware Model Aggregation (BAMA), a novel debiasing method for federated fine-tuning that consistently reduces bias across various models and algorithms.
Early warning systems for disease outbreaks play a crucial role in public health for management and contingency planning. However, most predictive modeling works focus on flat models that incorporate exogenous inputs (e.g. climate, demographics) to predict future outbreaks at different locations, but do not jointly model multiple spatial aggregation levels. In this paper, we introduce HierarNet, a unique independent-interactive hierarchical forecasting framework that aims to predict disease outbreaks at different levels of spatial resolution, such as provinces, regions, and nations. HierarNet consists of two main phases. In the local phase, we train independent forecasting models for all locations at all levels. In the global phase, all models iteratively interact with others across different levels via their hierarchical relationships under an ensemble fashion to maximize their agreements. This global local hierarchical interactive scheme makes HierarNet a highly effective and flexible method (i.e. it can work with an arbitrary base prediction model and available exogenous data for each location independently). Extensive experiments are conducted on various disease datasets (e.g., Dengue fever, flu, diarrhea, and Bluetongue) in different countries (e.g., France, Vietnam, and USA) to show the performance of HierarNet compared to 19 state-of-the-art (SOTA) methods such as MinT, DYCHEM, WITRAN, SegRNN, TSMixer, PatchTST, or iTransformer. We also illustrate the generability of HierarNet in other domains, e.g., web traffic forecasting.
Adaptive Regulation via Dual-Layer Evolution (ARDE): A Multi-Agent Approach to Balancing Efficiency, Fairness, and Diversity in Crowdsourced Platforms
PDF ↗Crowdsourced delivery platforms (e.g., Meituan, Uber Eats, DoorDash) have become vital infrastructure in urban logistics, yet their competitive order-grabbing mechanisms often lead to strategy homogenization, inefficiency, and income inequality. This paper presents ARDE (Adaptive Regulation via Dual-layer Evolution), an evolutionary governance framework that integrates individual reinforcement learning with adaptive platform-level regulation. The outer agent dynamically generates governance signals based on system diagnostics (strategy entropy, Gini coefficient, completion rate), while inner agents employ Diffusion Q-Learning guided by a language-model-driven reward shaping module to promote fairness and strategy diversity. Experiments on real-world datasets show that ARDE achieves stable diversity (0.997 ± 0.184), reduces inequality (Gini change 1.3%), and maintains high efficiency. Further comparison (ARDE-PPO vs. MAPPO) confirms that its advantages stem from explicit hierarchical governance rather than algorithmic coincidence. Overall, ARDE offers a scalable and interpretable paradigm for reconciling individual rationality with collective welfare in gig economies and other multi-agent socio-technical systems.
Realistic Curriculum Reinforcement Learning for Autonomous and Sustainable Marine Vessel Navigation
PDF ↗Sustainability is becoming increasingly critical in the maritime transport, encompassing both environmental and social impacts, such as Greenhouse Gas (GHG) emissions and navigational safety. Traditional vessel navigation heavily relies on human experience, often lacking autonomy and emission awareness, and is prone to human errors that may compromise safety. In this paper, we propose a Curriculum Reinforcement Learning (CRL) framework integrated with a realistic, data-driven marine simulation environment and a machine learning-based fuel consumption prediction module. The simulation environment is constructed using real-world vessel movement data and enhanced with a Diffusion Model to simulate dynamic maritime conditions. Vessel fuel consumption is estimated using historical operational data and learning-based regression. The surrounding environment is represented as image-based inputs to capture spatial complexity. We design a lightweight, policy-based CRL agent with a comprehensive reward mechanism that considers safety, emissions, timeliness, and goal completion. This framework effectively handles complex tasks progressively while ensuring stable and efficient learning in continuous action spaces. We validate the proposed approach in a sea area of the Indian Ocean, demonstrating its efficacy in enabling sustainable and safe vessel navigation.
Understanding opinion evolution in complex social networks is crucial for modeling social influence and predicting collective behavior. Yet, most models overlook how community structures shape opinion updates, often assuming homogeneous influence. This abstraction neglects individuals’ stronger responsiveness to intra-community peers—an empirically observed driver of localized consensus and inter-group polarization. We propose GCAOFP, a co-evolutionary framework that jointly models opinion dynamics and community formation as an integrated process. In GCAOFP, agents strategically alternate between two coupled modules: (1) a Community Dynamics Module, where agents play a non-cooperative game to optimize community memberships based on opinion alignment and structural cohesion; and (2) an Opinion Adjustment Module, where agents revise opinions via a bounded-confidence mechanism modulated by community-induced influence weights. This dual-stage process captures the feedback loop between structure and opinion. We prove that GCAOFP converges to stable equilibria, ensuring intra-community consensus and inter-community diversity—dynamics that standard models fail to replicate. Experiments on real-world networks show that GCAOFP better reproduces localized opinion clusters, while offering strong scalability and interpretability, illuminating the strategic foundations of polarization.
Scaling Equitable Reflection Assessment in Education via Large Language Models and Role-Based Feedback Agents
PDF ↗Formative feedback is widely recognized as one of the most effective drivers of student learning, yet it remains difficult to implement equitably at scale. In large or low-resource courses, instructors often lack the time, staffing, and bandwidth required to review and respond to every student reflection, creating gaps in support precisely where learners would benefit most. This paper presents a theory-grounded system that uses five coordinated role-based LLM agents (Evaluator, Equity Monitor, Metacognitive Coach, Aggregator, and Reflexion Reviewer) to score learner reflections with a shared rubric and to generate short, bias-aware, learner-facing comments. The agents first produce structured rubric scores, then check for potentially biased or exclusionary language, add metacognitive prompts that invite students to think about their own thinking, and finally compose a concise feedback message of at most 120 words. The system includes simple fairness checks that compare scoring error across lower and higher scoring learners, enabling instructors to monitor and bound disparities in accuracy. We evaluate the pipeline in a 12-session AI literacy program with adult learners. In this setting, the system produces rubric scores that approach expert-level agreement, and trained graders rate the AI-generated comments as helpful, empathetic, and well aligned with instructional goals. Taken together, these results show that multi-agent LLM systems can deliver equitable, high-quality formative feedback at a scale and speed that would be impossible for human graders alone. The approach demonstrates how structured agent roles, fairness checks, and learning-science principles can work together to support instructors while preserving pedagogical intent. More broadly, the work points toward a future where feedback-rich learning becomes feasible for any course size or context, advancing long-standing goals of equity, access, and instructional capacity in education.
Improving Day-Ahead Grid Carbon Intensity Forecasting by Joint Modeling of Local-Temporal and Cross-Variable Dependencies Across Different Frequencies
PDF ↗Accurate forecasting of the grid carbon intensity factor (CIF) is critical for enabling demand-side management and reducing emissions in modern electricity systems. Leveraging multiple interrelated time series, CIF prediction is typically formulated as a multivariate time series forecasting problem. Despite advances in deep learning-based methods, it remains challenging to capture the fine-grained local-temporal dependencies, dynamic higher-order cross-variable dependencies, and complex multi-frequency patterns for CIF forecasting. To address these issues, we propose a novel model that integrates two parallel modules: 1) one enhances the extraction of local-temporal dependencies under multi-frequency by applying multiple wavelet-based convolutional kernels to overlapping patches of varying lengths; 2) the other captures dynamic cross-variable dependencies under multi-frequency to model how inter-variable relationships evolve across the time-frequency domain. Evaluations on four representative electricity markets from Australia, featuring varying levels of renewable penetration, demonstrate that the proposed method outperforms the state-of-the-art models. An ablation study further validates the complementary benefits of the two proposed modules. Designed with built-in interpretability, the proposed model also enables better understanding of its predictive behavior, as shown in a case study where it adaptively shifts attention to relevant variables and time intervals during a disruptive event.
Effective crime linkage analysis is crucial for identifying serial offenders and enhancing public safety. To address the limitations of traditional crime linkage methods when handling high-dimensional, sparse, and heterogeneous data, this paper proposes a Siamese Autoencoder framework to learn meaningful latent representations and uncover correlations in highly complex data. Using a dataset from the Violent Crime Linkage Analysis System—a database maintained by the Serious Crime Analysis Section of the UK’s National Crime Agency—our approach mitigates signal dilution in high-dimensional sparse data through decoder-stage integration of geographic-temporal features. This integration amplifies learned behavioral representations rather than allowing them to be overwhelmed at the input stage, leading to consistent improvements over baseline methods across multiple metrics. We further examine how different data reduction strategies based on domain-expert can impact model performance, offering practical insights into preprocessing for crime linkage. Our solution shows that advanced machine learning approaches can enhance linkage accuracy, improving AUC by up to 9% over traditional methods and providing insights to support human decision-making in crime investigation.
Uncovering Bias Paths with LLM-guided Causal Discovery: An Active Learning and Dynamic Scoring Approach
PDF ↗Ensuring fairness in machine learning requires understanding how sensitive attributes like race or gender causally influence outcomes. Existing causal discovery (CD) methods often struggle to recover fairness-relevant pathways in the presence of noise, confounding, or data corruption. Large language models (LLMs) offer a complementary signal by leveraging semantic priors from variable metadata. We propose a hybrid LLM-guided CD framework that extends a breadth-first search strategy with active learning and dynamic scoring. Variable pairs are prioritized for querying using a composite score combining mutual information, partial correlation, and LLM confidence, enabling more efficient and robust structure discovery. To evaluate fairness sensitivity, we introduce a semi-synthetic benchmark based on the UCI Adult dataset, embedding domain-informed bias pathways alongside noise and latent confounders. We assess how well CD methods recover both global graph structure and fairness-critical paths (e.g., sex→education→income). Our results demonstrate that LLM-guided methods, including our active, dynamically scored variant, outperform baselines in recovering fairness-relevant structure under noisy conditions. We analyze when LLM-driven insights complement statistical dependencies and discuss implications for fairness auditing in high-stakes domains.