论文检索

Babette Bühler, Ivo Bueno, Enkelejda Kasneci

Writing is a foundational skill for educational, professional, and civic participation, yet access to frequent and timely writing feedback remains deeply unequal. Teachers face significant workload constraints, particularly in large classes, and many learners lack alternative sources of individualized feedback. While large language models (LLMs) offer the opportunity for scalable, adaptive support, little is known about how students engage with such feedback tools in real-world, self-directed settings. We present a large-scale, year-long analysis of 23,650 voluntary interactions with an open-access AI writing feedback system used by students across diverse educational contexts and age groups, conducted in accordance with strict data protection standards. Using a clustering approach, we identify 2,800 iterative revision chains and apply a validated LLM-based multidimensional scoring framework to assess text quality over time. Our findings reveal that students who revised their texts after receiving AI feedback demonstrated statistically significant, albeit modest, improvements across both content and language-related dimensions (overall writing quality: ∆ = 0.067, p < .001, r = .17), with the greatest gains observed among initially low-performing writers. Revision frequency was positively associated with improvement, particularly in higher-order writing skills. However, engagement was uneven, with higher usage among students in academically oriented schools. These results demonstrate both the technical feasibility and social potential of deploying generative AI for educational support at scale, while highlighting the need for inclusive infrastructure, accessible design, and targeted outreach to truly democratize educational benefits.

Avrim Blum, Emily Diana, Kavya Ravichandran, Alexander Williams Tolbert

Ambition and risk-taking have been heralded as important ways for marginalized communities to get out of cycles of poverty. As a result, educational messaging often encourages individuals to strengthen their personal resolve and develop characteristics such as discipline and grit to succeed in ambitious ends. However, recent work in philosophy and sociology highlights that this messaging often does more harm than good for students in these situations. We study similar questions using a different epistemic approach and in simple theoretical models -- we provide a quantitative model of decision-making between stable and risky choices in the improving multi-armed bandits framework. We use this model to first study how individuals' "strategies" are affected by their level of grittiness and how this affects their accrued rewards. Then, we study the impact of various interventions, such as increasing grit or providing a financial safety net. Our investigation of rational decision making studies the competitive ratio between the accrued reward and the optimal reward.

Andrew Blair-Stanek, Nils Holzenberger, Benjamin Van Durme

We investigate whether large language models can discover and analyze U.S. tax-minimization strategies. This real-world domain challenges even seasoned human experts, and progress can reduce tax revenue lost from well-advised, wealthy taxpayers. We evaluate the most advanced LLMs on their ability to (1) interpret and verify tax strategies, (2) fill in gaps in partially specified strategies, and (3) generate complete, end-to-end strategies from scratch. This domain should be of particular interest to the LLM reasoning community: unlike synthetic challenge problems or scientific reasoning tasks, U.S. tax law involves navigating hundreds of thousands of pages of statutes, case law, and administrative guidance, all updated regularly. Notably, an LLM identified an apparently novel tax strategy, highlighting these models' potential to revolutionize tax agencies' fight against tax abuse.

Ting Bi, Chenghang Ye, Zheyu Yang, Ziyi Zhou, Cui Tang, Zui Tao, Jun Zhang, Kailong Wang, Liting Zhou, Yang Yang 等

Augmented Reality (AR) and Multimodal Large Language Models (LLMs) are rapidly evolving, providing unprecedented capabilities for human-computer interaction. However, their integration introduces a new attack surface for Social Engineering (SE). In this paper, we systematically investigate the feasibility of orchestrating AR-driven Social Engineering attacks using Multimodal LLM for the first time, via our proposed SEAR framework, which operates through three key phases: (1) AR-based social context synthesis, which fuses Multimodal inputs (visual, auditory and environmental cues); (2) role-based Multimodal RAG (Retrieval-Augmented Generation), which dynamically retrieves and integrates social context; and (3) ReInteract social engineering agents, which execute adaptive multiphase attack strategies through inference interaction loops. To verify SEAR, we conducted an IRB-approved study with 60 participants and build a novel dataset of 180 annotated conversations in different social scenarios (e.g., coffee shops, networking events). Our results show that SEAR is highly effective at eliciting high-risk behaviors (e.g., 93.3% of participants susceptible to email phishing). The framework was particularly effective in building trust, with 85% of targets willing to accept an attacker's call after an interaction. Also, we identified notable limitations such as authenticity gaps. This work provides proof-of-concept for AR-LLM driven social engineering attacks and insights for developing defenses against next-generation AR/LLM-based SE threats.

Ankit Bhardwaj, Rohail Asim, Sachin Kumar Chauhan, Yasir Zaki, Lakshmi Subramanian

Free-flow road networks, such as suburban highways, are increasingly experiencing traffic congestion due to growing commuter inflow and limited infrastructure. Traditional control mechanisms—traffic signals or local heuristics—are ineffective or infeasible in these high-speed, signal-free environments. We introduce self-regulating cars, a reinforcement learning-based traffic control protocol that dynamically modulates vehicle speeds to optimize throughput and prevent congestion, without requiring new physical infrastructure. Our approach integrates classical traffic flow theory, gap acceptance models, and microscopic simulation into a physics-informed RL framework. By abstracting roads into super-segments, the agent captures emergent flow dynamics and learns robust speed modulation policies from instantaneous traffic observations. Evaluated in the high-fidelity PTV Vissim simulator on a real-world highway network, our method improves total throughput by 5%, reduces average delay by 13%, and decreases total stops by 3% compared to the no-control setting. It also achieves smoother, congestion-resistant flow while generalizing across varied traffic patterns—demonstrating its potential for scalable, ML-driven traffic management.

Livia Betti, Farooq Sanni, Gnouyaro Z. Sogoyou, Togbe Agbagla, Cullen Molitor, Tamma Carleton, Esther Rolf

In applications across agriculture, ecology, and human development, machine learning with satellite imagery (SatML) is limited by the sparsity of labeled training data. While satellite data cover the globe, labeled training datasets for SatML are often small, spatially clustered, and collected for other purposes (e.g., administrative surveys or field measurements). Despite the pervasiveness of this issue in practice, past SatML research has largely focused on new model architectures and training algorithms to handle scarce training data, rather than modeling data conditions directly. This leaves scientists and policymakers who wish to use SatML for large-scale monitoring uncertain about whether and how to collect additional data to maximize performance. Here, we present the first problem formulation for the optimization of spatial training data in the presence of heterogeneous data collection costs and realistic budget constraints, as well as novel methods for addressing this problem. In experiments simulating different problem settings across three continents and four tasks, our strategies reveal substantial gains from sample optimization. Further experiments delineate settings for which optimized sampling is particularly effective. The problem formulation and methods we introduce are designed to generalize across application domains for SatML; we put special emphasis on a specific problem setting where our coauthors can immediately use our findings to augment clustered agricultural surveys for SatML monitoring in Togo.

Tina Behzad, Siddartha Devic, Vatsal Sharan, Aleksandra Korolova, David Kempe

We conduct an independent, third-party audit for bias of LinkedIn's Talent Search ranking system, focusing on potential ranking bias across two attributes: gender and race. To do so, we first construct a dataset of rankings produced by the system, collecting extensive Talent Search results across a diverse set of occupational queries. We then develop a robust labeling pipeline that infers the two demographic attributes of interest for the returned users. To evaluate potential biases in the collected dataset of real-world rankings, we utilize two exposure disparity metrics: deviation from group proportions and MinSkew@k. Our analysis reveals an under-representation of minority groups in early ranks across many queries. We further examine potential causes of this disparity, and discuss why they may be difficult or, in some cases, impossible to fully eliminate among the early ranks of queries. Beyond static metrics, we also investigate the concept of subgroup fairness over time, highlighting \emph{temporal disparities} in exposure and retention, which are often more difficult to audit for in practice. In employer recruiting platforms such as LinkedIn Talent Search, the persistence of a particular candidate over multiple days in the ranking can directly impact the probability that the given candidate is selected for opportunities. Our analysis reveals demographic disparities in this temporal stability, with some groups experiencing greater volatility in their ranked positions than others. We contextualize all our findings alongside LinkedIn’s published self-audits of its Talent Search system and reflect on the methodological constraints of a black-box external evaluation, including limited observability and noisy demographic inference. Our work contributes empirical insights and practical guidance for conducting third-party audits of modern socio-technical systems which go beyond the well-studied and standard algorithmic fairness guarantees of predictors.

Sumana Basu, Flemming Kondrup, Adriana Romero-Soriano, Doina Precup

Personalized insulin therapy for individuals with Type 1 Diabetes via closed‑loop artificial pancreas systems requires rapid adaptation of dosing strategies to each patient's unique insulin response. However, learning patient‑specific policies from scratch demands extensive exploration, which is often impractical. In this work, we study a framework that integrates insulin-response-informed transfer learning with model-based reinforcement learning for insulin dosing. We first train an LSTM‑based insulin responsiveness predictor on virtual patients, using their glucose, insulin, and meal history to forecast future glucose levels. Analysis of insulin responsiveness of in-silico patients uncovers natural insulin‑response groups characterized by similar sensitivity and dynamics profiles. For a new patient, we identify a representative model from their response group and use it to generate synthetic trajectories. These trajectories are integrated into an enhanced H-step Deep Dyna-Q algorithm, enabling accelerated policy optimization through model-based planning. The dynamics model trained entirely in simulation achieves 91.31% accuracy in predicting blood glucose ranges on the Ohio Type 1 Diabetes dataset, indicating strong zero-shot generalization. Additionally, we find that bootstrapping a new patient with a physiologically-matched reference model accelerates convergence of effective dosing policies across in-silico cohorts of children, adolescents, and adults. These findings suggest that leveraging response-group-specific synthetic experience can expedite personalized insulin therapy, offering a promising pathway towards clinical validation.

Abhipsa Basu, Aviral Gupta, Abhijnya Bhat, Venkatesh Babu Radhakrishnan

Image classification systems often inherit biases from uneven group representation in training data. For example, in face datasets for hair color classification, blond hair may be disproportionately associated with females, reinforcing stereotypes. A recent approach leverages the Stable Diffusion model to generate balanced training data, but these models often struggle to preserve the original data distribution. In this work, we explore multiple diffusion-finetuning techniques, e.g., LoRA and DreamBooth, to generate images that more accurately represent each training group by learning directly from their samples. Additionally, in order to prevent a single DreamBooth model from being overwhelmed by excessive intra-group variations, we explore a technique of clustering images within each group and train a DreamBooth model per cluster. These models are then used to generate group-balanced data for pretraining, followed by fine-tuning on real data. Experiments on multiple benchmarks demonstrate that the studied finetuning approaches outperform vanilla Stable Diffusion on average and achieve results comparable to SOTA debiasing techniques like Group-DRO, while surpassing them as the dataset bias severity increases.

Sidhika Balachandar, Shuvom Sadhuka, Bonnie Berger, Emma Pierson, Nikhil Garg

Graph neural networks (GNNs) are widely used in urban spatiotemporal forecasting, e.g., predicting infrastructure problems. In this setting, government officials aim to identify in which neighborhoods incidents like potholes or rodents occur. The true state of incidents is observed via government inspection ratings. However, these ratings are only conducted for a sparse set of neighborhoods and incident types. We also observe the state of incidents via crowdsourced reports, which are more densely observed but may be biased due to heterogeneous reporting. First, we propose a multiview, multioutput GNN-based model that uses both unbiased rating data and biased reporting data to predict the true latent state of incidents. Second, we investigate a case study of New York City urban incidents and collect a dataset of 9,615,863 crowdsourced reports and 1,041,415 government inspection ratings over 3 years and across 139 types of incidents. We show on both real and semi-synthetic data that our model can better predict the latent state compared to models that use only reporting data or only rating data. Finally, we quantify demographic biases in crowdsourced reporting, e.g., higher-income neighborhoods report problems at higher rates. Our analysis showcases a widely applicable approach for latent state prediction using heterogeneous, sparse, and biased data.

Mohammadsina Almasi, Hadis Anahideh

Equitably allocating limited resources in high-stakes domains—such as education, employment, and healthcare—requires balancing short-term utility with long-term impact, while accounting for delayed outcomes, hidden heterogeneity, and ethical constraints. However, most learning-based allocation frameworks either assume immediate feedback or ignore the complex interplay between individual characteristics and intervention dynamics. We propose a novel bi-level contextual bandit framework for individualized resource allocation under delayed feedback, designed to operate in real-world settings with dynamic populations, capacity constraints, and time-sensitive impact. At the meta level, the model optimizes subgroup-level budget allocations to satisfy fairness and operational constraints. At the base level, it identifies the most responsive individuals within each group using a neural network trained on observational data, while respecting cooldown windows and delayed treatment effects modeled via resource-specific delay kernels. By explicitly modeling temporal dynamics and feedback delays, the algorithm continually refines its policy as new data arrive, enabling more responsive and adaptive decision-making. We validate our approach on two real-world datasets from education and workforce development, showing that it achieves higher cumulative outcomes, better adapts to delay structures, and ensures equitable distribution across subgroups. Our results highlight the potential of delay-aware, data-driven decision-making systems to improve institutional policy and social welfare.

Xiangyu Zhou, Yao Qiang, Saleh Zare Zade, Douglas Zytko, Prashant Khanduri, Dongxiao Zhu

Large Language Models (LLMs), pre-trained on massive text corpora, exhibit remarkable human-level language understanding, reasoning, and decision-making abilities. However, they tend to memorize unwanted information, such as private or copyrighted content, raising significant privacy and legal concerns. Unlearning has emerged as a promising solution, but existing methods face a significant challenge of over-forgetting. This issue arises because they indiscriminately suppress the generation of all the tokens in forget samples, leading to a substantial loss of model utility. To overcome this challenge, we introduce the Targeted Information Forgetting (TIF) framework, which consists of (1) a flexible targeted information identifier designed to differentiate between unwanted words (UW) and general words (GW) in the forget samples, and (2) a novel Targeted Preference Optimization approach that leverages Logit Preference Loss to unlearn unwanted information associated with UW and Preservation Loss to retain general information in GW, effectively improving the unlearning process while mitigating utility degradation. Extensive experiments on the TOFU and MUSE benchmarks demonstrate that the proposed TIF framework enhances unlearning effectiveness while preserving model utility and achieving state-of-the-art results.

Tianyi Zhou, Johanne Medina, Sanjay Chawla

Large Language Models (LLMs) are prone to generating fluent but incorrect content, known as confabulation, which poses increasing risks in multi-turn or agentic applications where outputs may be reused as context. In this work, we investigate how in-context information influences model behavior and whether LLMs can identify their unreliable responses. We propose a reliability estimation that leverages token-level uncertainty to guide the aggregation of internal model representations. Specifically, we compute aleatoric and epistemic uncertainty from output logits to identify salient tokens and aggregate their hidden states into compact representations for response-level reliability prediction. Through controlled experiments on open QA benchmarks, we find that correct in-context information improves both answer accuracy and model confidence, while misleading context often induces confidently incorrect responses, revealing a misalignment between uncertainty and correctness. Our probing-based method captures these shifts in model behavior and improves the detection of unreliable outputs across multiple open-source LLMs. These results underscore the limitations of direct uncertainty signals and highlight the potential of uncertainty-guided probing for reliability-aware generation.

Junwen Zheng, Xinran Xu, Li Rong Wang, Chang Cai, Lucinda Siyun Tan, Dingyuan Wang, Hong Liang Tey, Xiuyi Fan

Deep learning has demonstrated expert-level performance in melanoma classification, positioning it as a powerful tool in clinical dermatology. However, model opacity and the lack of interpretability remain critical barriers to clinical adoption, as clinicians often struggle to trust the decision-making processes of black-box models. To address this gap, we present a Cross-modal Explainable Framework for Melanoma (CEFM) that leverages contrastive learning as the core mechanism for achieving interpretability. Specifically, CEFM maps clinical criteria for melanoma diagnosis—namely Asymmetry, Border, and Color (ABC)—into the Vision Transformer embedding space using dual projection heads, thereby aligning clinical semantics with visual features. The aligned representations are subsequently translated into structured textual explanations via natural language generation, creating a transparent link between raw image data and clinical interpretation. Experiments on public datasets demonstrate 92.79% accuracy and an AUC of 0.961, along with significant improvements across multiple interpretability metrics. Qualitative analyses further show that the spatial arrangement of the learned embeddings aligns with clinicians’ application of the ABC rule, effectively bridging the gap between high-performance classification and clinical trust.

Yiyang Zhao, Huiyu Bai, Xuejiao Zhao

Alignment of large language models (LLMs) with human preferences typically relies on supervised reward models or external judges that demand abundant annotations. However, in fields that rely on professional knowledge, such as medicine and law, such large-scale preference labels are often unachievable. In this paper, we propose a generative entropy-guided preference modeling approach named GEM for LLMs aligment at low-resource and domain-specific scenarios. Instead of training a discriminative reward model on preference data, we directly train the LLM to internalize a closed-loop optimization architecture that can extract and exploit the multi-dimensional, fine-grained cognitive signals implicit in human preferences. Specifically, our \textit{Cognitive Filtering} module, based on entropy theory in decision making, first leverages Chain-of-Thought (CoT) prompting to generate diverse candidate reasoning chains (CoTs) from preference data. Subsequently, it introduces a token scoring mechanism to rank and weight the sampled CoTs, boosting the importance of high-confidence answers and strategically high-entropy tokens. Building on these filtered preferences, we fine-tune the LLM using a novel self-evaluated group advantage algorithm, \textit{SEGA}, which effectively aggregates group-level cognitive signals and transforms the entropy-based scores into implicit rewards for policy optimization. In these ways, GEM empowers the LLM to rely on its own judgments and establishes an entropy-guided closed-loop cognitive optimization framework, enabling highly efficient few-shot alignment of LLMs. Experiments on general benchmarks and domain-specific tasks (such as mathematical reasoning and medical dialogues) demonstrate that our GEM achieves significant improvements with few-shot preference data.

Xin Zhao, Xiaojun Chen, Bingshan Liu, Zeyao Liu, Zhendong Zhao, Xiaoyan Gu

Generative vision-language models like Stable Diffusion demonstrate remarkable capabilities in creative media synthesis, but they also pose substantial risks of producing unsafe, offensive, or culturally inappropriate content when prompted adversarially. Current defenses struggle to align outputs with human values without sacrificing generation quality or incurring high costs. To address these challenges, we introduce VALOR (Value-Aligned LLM-Overseen Rewriter), a modular, zero-shot agentic framework for safer and more helpful text-to-image generation. VALOR integrates layered prompt analysis with human-aligned value reasoning: a multi-level NSFW detector filters lexical and semantic risks; a cultural value alignment module identifies violations of social norms, legality, and representational ethics; and an intention disambiguator detects subtle or indirect unsafe implications. When unsafe content is detected, prompts are selectively rewritten by a large language model under dynamic, role-specific instructions designed to preserve user intent while enforcing alignment. If the generated image still fails a safety check, VALOR optionally performs a stylistic regeneration to steer the output toward a safer visual domain without altering core semantics. Experiments across adversarial, ambiguous, and value-sensitive prompts show that VALOR significantly reduces unsafe outputs by up to 100.00% while preserving prompt usefulness and creativity. These results highlight VALOR as a scalable and effective approach for deploying safe, aligned, and helpful image generation systems in open-world settings.

Xiaofei Zhao

The increasing complexity of modern AI systems exposes a significant assurance gap: safety evidence from practices like red-teaming and robustness testing remains fragmented, lacking a formal mechanism for composition and propagation throughout the development lifecycle. This prevents the construction of rigorous, dynamic safety cases essential for trustworthy AI. We introduce the Composable Assurance Framework (CAF), a novel engineering methodology that integrates safety assurance directly into MLOps workflows. At its core is the Formal Safety Assertion (FSA), a standardized, machine-readable structure that verifiably links safety properties—such as robustness scores or the absence of deceptive circuits—to specific AI artifacts. We then define a Composition Calculus, a set of formal rules governing how FSAs are propagated and aggregated as components are combined into a system. This approach transforms the development pipeline into an automated evidence-gathering engine, whose output is a dynamic Directed Acyclic Graph (DAG) of assertions that constitutes a living safety case. Through a prototype and a Retrieval-Augmented Generation (RAG) case study, we demonstrate how CAF automatically enforces a predefined safety policy, blocking non-compliant deployments.

Weixiang Zhao, Haozhen Li, Yanyan Zhao, Haixiao Liu, Biye Li, Ting Liu, Bing Qin

As large language models (LLMs) are increasingly deployed across culturally diverse regions, ensuring that their responses align with users’ cultural norms has become a critical challenge. Existing approaches to cultural alignment primarily rely on prompting or data-augmentation-based supervised finetuning, which teach models to follow norms indirectly through example-based supervision. However, these methods are difficult to scale and often fail to generalize, particularly in low-resource cultural settings. In this work, we propose CultureRL, a culture-norm-driven reinforcement learning framework that directly encodes cultural principles into model behavior. Rather than relying on output imitation, CultureRL provides normative feedback during training, enabling the model to internalize high-level cultural rules. It consists of two key components: (1) Norm Pool Construction (NPC), which clusters data from the World Values Survey into abstract cultural concepts to form a structured and retrievable norm pool; and (2) Norm Cluster-based Reward Mechanism (NCRM), which retrieves the relevant norm for each input and uses an external reward model to assess conformity, guiding model updates through cultural alignment. We evaluate CultureRL in both one-for-one (per-culture) and one-for-all (multi-culture) settings across nine cultures and three benchmarks. Results show that CultureRL consistently outperforms strong baselines, especially in terms of cultural consistency and adaptability.

Yitong Zhang, Jia Li, Liyi Cai, Ge Li

Large Vision-Language Models (LVLMs) have achieved impressive progress across various applications but remain vulnerable to malicious queries. Existing safety alignment approaches typically fail to resist malicious queries while preserving utility on benign ones effectively. To address these challenges, we propose DAVSP, which is built upon two key innovations. First, we introduce Visual Safety Prompt, which appends a trainable padding region around the input image. It preserves visual features and expands the optimization space. Second, we propose Deep Alignment, a novel approach to train the visual safety prompt through supervision in the model's activation space. It enhances the inherent ability of LVLMs to perceive malicious queries, achieving deeper alignment than prior works. Extensive experiments demonstrate that DAVSP effectively resists malicious queries while preserving benign input utility. Furthermore, DAVSP exhibits great cross-model generation ability. Ablation studies further reveal that both the Visual Safety Prompt and Deep Alignment are essential to the overall effectiveness.

Peng Zhang, Peijie Sun

Safety alignment instills in Large Language Models (LLMs) a critical capacity to refuse malicious requests. Prior works have modeled this refusal mechanism as a single linear direction in the activation space. We posit that this is an oversimplification that conflates two functionally distinct neural processes: the detection of harm and the execution of a refusal. In this work, we deconstruct this single representation into a Harm Detection Direction and a Refusal Execution Direction. Leveraging this fine-grained model, we introduce Differentiated Bi-Directional Intervention (DBDI), a new white-box framework that precisely neutralizes the safety alignment at critical layer. DBDI applies adaptive projection nullification to the refusal execution direction while suppressing the harm detection direction via direct steering. Extensive experiments demonstrate that DBDI outperforms prominent jailbreaking methods, achieving up to a 97.88% attack success rate on models such as Llama-2. By providing a more granular and mechanistic framework, our work offers a new direction for the in-depth understanding of LLM safety alignment.