Search

Ensuring transparency in AI decision-making requires interpretable explanations, particularly at the instance level. Counterfactual explanations are a powerful tool for this purpose, but existing techniques frequently depend on synthetic examples, introducing biases from unrealistic assumptions, flawed models, or skewed data. Many methods also assume full dataset availability, an impractical constraint in real-time environments where data flows continuously. In contrast, streaming explanations offer adaptive, real-time insights without requiring persistent storage of the entire dataset. This work introduces a scalable, model-free approach to selecting diverse and relevant counterfactual examples directly from observed data. Our algorithm operates efficiently in streaming settings, maintaining $O(\log k)$ update complexity per item while ensuring high-quality counterfactual selection. Empirical evaluations on both real-world and synthetic datasets demonstrate superior performance over baseline methods, with robust behavior even under adversarial conditions.

The Hierarchical Navigable Small World (HNSW) algorithm is widely used for approximate nearest neighbor (ANN) search, leveraging the principles of navigable small-world graphs. However, it faces some limitations. The first is the local optima problem, which arises from the algorithm's greedy search strategy, selecting neighbors based solely on proximity at each step. This often leads to cluster disconnections. The second limitation is that HNSW frequently fails to achieve logarithmic complexity, particularly in high-dimensional datasets, due to the exhaustive traversal through each layer. To address these limitations, we propose a novel algorithm that mitigates local optima and cluster disconnections while enhancing the construction speed, maintaining inference speed. The first component is a dual-branch HNSW structure with LID-based insertion mechanisms, enabling traversal from multiple directions. This improves outlier node capture, enhances cluster connectivity, accelerates construction speed and reduces the risk of local minima. The second component incorporates a bridge-building technique that bypasses redundant intermediate layers, maintaining inference and making up the additional computational overhead introduced by the dual-branch structure. Experiments on various benchmarks and datasets showed that our algorithm outperforms the original HNSW in both accuracy and speed. We evaluated six datasets across Computer Vision (CV), and Natural Language Processing (NLP), showing recall improvements of 18\% in NLP, and up to 30\% in CV tasks while reducing the construction time by up to 20\% and maintaining the inference speed. We did not observe any trade-offs in our algorithm. Ablation studies revealed that LID-based insertion had the greatest impact on performance, followed by the dual-branch structure and bridge-building components.

In the task of reference-based image inpainting, an additional reference image is provided to restore a damaged target image to its original state. The advancement of diffusion models, particularly Stable Diffusion, allows for simple formulations in this task. However, existing diffusion-based methods often lack explicit constraints on the correlation between the reference and damaged images, resulting in lower faithfulness to the reference images in the inpainting results. In this work, we propose CorrFill, a training-free module designed to enhance the awareness of geometric correlations between the reference and target images. This enhancement is achieved by guiding the inpainting process with correspondence constraints estimated during inpainting, utilizing attention masking in self-attention layers and an objective function to update the input tensor according to the constraints. Experimental results demonstrate that CorrFill significantly enhances the performance of multiple baseline diffusion-based methods, including state-of-the-art approaches, by emphasizing faithfulness to the reference images., Comment: WACV 2025. Project page: https://corrfill.github.io/

We explore the potential of large-scale generative video models for autonomous driving, introducing an open-source auto-regressive video model (VaViM) and its companion video-action model (VaVAM) to investigate how video pre-training transfers to real-world driving. VaViM is a simple auto-regressive video model that predicts frames using spatio-temporal token sequences. We show that it captures the semantics and dynamics of driving scenes. VaVAM, the video-action model, leverages the learned representations of VaViM to generate driving trajectories through imitation learning. Together, the models form a complete perception-to-action pipeline. We evaluate our models in open- and closed-loop driving scenarios, revealing that video-based pre-training holds promise for autonomous driving. Key insights include the semantic richness of the learned representations, the benefits of scaling for video synthesis, and the complex relationship between model size, data, and safety metrics in closed-loop evaluations. We release code and model weights at https://github.com/valeoai/VideoActionModel, Comment: Code and model: https://github.com/valeoai/VideoActionModel, project page: https://valeoai.github.io/vavim-vavam/

We study the replacement paths problem in the $\mathsf{CONGEST}$ model of distributed computing. Given an $s$-$t$ shortest path $P$, the goal is to compute, for every edge $e$ in $P$, the shortest-path distance from $s$ to $t$ avoiding $e$. For unweighted directed graphs, we establish the tight randomized round complexity bound for this problem as $\widetilde{\Theta}(n^{2/3} + D)$ by showing matching upper and lower bounds. Our upper bound extends to $(1+\epsilon)$-approximation for weighted directed graphs. Our lower bound applies even to the second simple shortest path problem, which asks only for the smallest replacement path length. These results improve upon the very recent work of Manoharan and Ramachandran (SIROCCO 2024), who showed a lower bound of $\widetilde{\Omega}(n^{1/2} + D)$ and an upper bound of $\widetilde{O}(n^{2/3} + \sqrt{n h_{st}} + D)$, where $h_{st}$ is the number of hops in the given $s$-$t$ shortest path $P$.

We develop a novel ultrasound nasogastric tube (UNGT) dataset to address the lack of public nasogastric tube datasets. The UNGT dataset includes 493 images gathered from 110 patients with an average image resolution of approximately 879 $\times$ 583. Four structures, encompassing the liver, stomach, tube, and pancreas are precisely annotated. Besides, we propose a semi-supervised adaptive-weighting aggregation medical segmenter to address data limitation and imbalance concurrently. The introduced adaptive weighting approach tackles the severe unbalanced challenge by regulating the loss across varying categories as training proceeds. The presented multiscale attention aggregation block bolsters the feature representation by integrating local and global contextual information. With these, the proposed AAMS can emphasize sparse or small structures and feature enhanced representation ability. We perform extensive segmentation experiments on our UNGT dataset, and the results show that AAMS outperforms existing state-of-the-art approaches to varying extents. In addition, we conduct comprehensive classification experiments across varying state-of-the-art methods and compare their performance. The dataset and code will be available upon publication at https://github.com/NUS-Tim/UNGT., Comment: 31 pages, 6 figures

Large vision-language models (LVLMs) perform outstandingly across various multimodal tasks. However, their ability to evaluate generated content remains limited, and training vision-language reward models (VLRMs) with preference data is computationally expensive. This paper explores a training-free alternative by merging text-based reward models (RMs) with LVLMs to create VLRMs. Our approach shows that integrating these models leads to improved performance over LVLMs' scoring and text-based RMs, offering an efficient method for incorporating textual preferences into LVLMs., Comment: Preprint. Under Review

Speech representation models are highly effective at extracting general features for various tasks. While fine-tuning can enhance these representations for specific applications, it often compromises their generalization ability. To address this challenge, we propose Speech-FT, a fine-tuning strategy for speech representation models that leverages model merging to preserve generalization ability while still benefiting from fine-tuning. Speech-FT is effective across different fine-tuning scenarios and is compatible with various types of speech representation models, providing a versatile solution. Speech-FT offers an efficient and practical approach to further improving general speech representations after pre-training.

Quantum impurities give rise to rich physical phenomena, with some exhibiting critical behavior described by conformal field theories (CFTs) in the low-energy limit. In parallel, party-time ($\mathcal{PT}$) symmetric non-Hermitian systems host exceptional points (EPs) at criticality, leading to exotic features governed by non-unitary CFTs. Here, we establish a connection between non-Hermitian impurities and CFTs by demonstrating that the critical properties of a (1+1)-dimensional free-fermion chain with central charge $c=1$ can be drastically altered by the presence of a local non-Hermitian impurity. Through a systematic analysis of entanglement/R\'enyi entropy, the finite-size scaling of the many-body spectrum, and fidelity susceptibility, we identify that this impurity-induced non-Hermitian criticality is characterized by a non-unitary CFT with central charge $c=-2$. Furthermore, we find that these non-unitary critical properties exhibit strong sensitivity to boundary conditions., Comment: 9 pages, 6 figures

Explainable AI (XAI) is critical for ensuring transparency, accountability, and trust in machine learning systems as black-box models are increasingly deployed within high-stakes domains. Among XAI methods, Shapley values are widely used for their fairness and consistency axioms. However, prevalent Shapley value approximation methods commonly rely on abstract baselines or computationally intensive calculations, which can limit their interpretability and scalability. To address such challenges, we propose Pairwise Shapley Values, a novel framework that grounds feature attributions in explicit, human-relatable comparisons between pairs of data instances proximal in feature space. Our method introduces pairwise reference selection combined with single-value imputation to deliver intuitive, model-agnostic explanations while significantly reducing computational overhead. Here, we demonstrate that Pairwise Shapley Values enhance interpretability across diverse regression and classification scenarios--including real estate pricing, polymer property prediction, and drug discovery datasets. We conclude that the proposed methods enable more transparent AI systems and advance the real-world applicability of XAI.

Offline model-based reinforcement learning (MBRL) serves as a competitive framework that can learn well-performing policies solely from pre-collected data with the help of learned dynamics models. To fully unleash the power of offline MBRL, model selection plays a pivotal role in determining the dynamics model utilized for downstream policy learning. However, offline MBRL conventionally relies on validation or off-policy evaluation, which are rather inaccurate due to the inherent distribution shift in offline RL. To tackle this, we propose BOMS, an active model selection framework that enhances model selection in offline MBRL with only a small online interaction budget, through the lens of Bayesian optimization (BO). Specifically, we recast model selection as BO and enable probabilistic inference in BOMS by proposing a novel model-induced kernel, which is theoretically grounded and computationally efficient. Through extensive experiments, we show that BOMS improves over the baseline methods with a small amount of online interaction comparable to only $1\%$-$2.5\%$ of offline training data on various RL tasks., Comment: 18 pages, 6 figures

Dental panoramic radiographs (DPRs) are widely used in clinical practice for comprehensive oral assessment but present challenges due to overlapping structures and time constraints in interpretation. This study aimed to establish a solid baseline for the AI-automated assessment of findings in DPRs by developing, evaluating an AI system, and comparing its performance with that of human readers across multinational data sets. We analyzed 6,669 DPRs from three data sets (the Netherlands, Brazil, and Taiwan), focusing on 8 types of dental findings. The AI system combined object detection and semantic segmentation techniques for per-tooth finding identification. Performance metrics included sensitivity, specificity, and area under the receiver operating characteristic curve (AUC-ROC). AI generalizability was tested across data sets, and performance was compared with human dental practitioners. The AI system demonstrated comparable or superior performance to human readers, particularly +67.9% (95% CI: 54.0%-81.9%; p < .001) sensitivity for identifying periapical radiolucencies and +4.7% (95% CI: 1.4%-8.0%; p = .008) sensitivity for identifying missing teeth. The AI achieved a macro-averaged AUC-ROC of 96.2% (95% CI: 94.6%-97.8%) across 8 findings. AI agreements with the reference were comparable to inter-human agreements in 7 of 8 findings except for caries (p = .024). The AI system demonstrated robust generalization across diverse imaging and demographic settings and processed images 79 times faster (95% CI: 75-82) than human readers. The AI system effectively assessed findings in DPRs, achieving performance on par with or better than human experts while significantly reducing interpretation time. These results highlight the potential for integrating AI into clinical workflows to improve diagnostic efficiency and accuracy, and patient management.

Current autonomous driving vehicles rely mainly on their individual sensors to understand surrounding scenes and plan for future trajectories, which can be unreliable when the sensors are malfunctioning or occluded. To address this problem, cooperative perception methods via vehicle-to-vehicle (V2V) communication have been proposed, but they have tended to focus on detection and tracking. How those approaches contribute to overall cooperative planning performance is still under-explored. Inspired by recent progress using Large Language Models (LLMs) to build autonomous driving systems, we propose a novel problem setting that integrates an LLM into cooperative autonomous driving, with the proposed Vehicle-to-Vehicle Question-Answering (V2V-QA) dataset and benchmark. We also propose our baseline method Vehicle-to-Vehicle Large Language Model (V2V-LLM), which uses an LLM to fuse perception information from multiple connected autonomous vehicles (CAVs) and answer driving-related questions: grounding, notable object identification, and planning. Experimental results show that our proposed V2V-LLM can be a promising unified model architecture for performing various tasks in cooperative autonomous driving, and outperforms other baseline methods that use different fusion approaches. Our work also creates a new research direction that can improve the safety of future autonomous driving systems. Our project website: https://eddyhkchiu.github.io/v2vllm.github.io/ ., Comment: Our project website: https://eddyhkchiu.github.io/v2vllm.github.io/

We prove that the minimal size $M(\pi_n)$ of a maximal matching in the permutahedron $\pi_n$ is asymptotically $n!/3$. On the one hand, we obtain a lower bound $M(\pi_n) \ge n! (n-1) / (3n-2)$ by considering $4$-cycles in the permutahedron. On the other hand, we obtain an asymptotical upper bound $M(\pi_n) \le n!(1/3+o(1))$ by multiple applications of Hall's theorem (similar to the approach of Forcade (1973) for the hypercube) and an exact upper bound $M(\pi_n) \le n!/3$ by an explicit construction. We also derive bounds on minimum maximal matchings in products of permutahedra., Comment: 11 pages, 2 figures

With the rise of multimodal large language models, GPT-4o stands out as a pioneering model, driving us to evaluate its capabilities. This report assesses GPT-4o across various tasks to analyze its audio processing and reasoning abilities. We find that GPT-4o exhibits strong knowledge in audio, speech, and music understanding, performing well in tasks like intent classification, spoken command classification, semantic and grammatical reasoning., multilingual speech recognition, and singing analysis. It also shows greater robustness against hallucinations than other large audio-language models (LALMs). However, it struggles with tasks such as audio duration prediction and instrument classification. Additionally, GPT-4o's safety mechanisms cause it to decline tasks like speaker identification, age classification, MOS prediction, and audio deepfake detection. Notably, the model exhibits a significantly different refusal rate when responding to speaker verification tasks on different datasets. This is likely due to variations in the accompanying instructions or the quality of the input audio, suggesting the sensitivity of its built-in safeguards. Finally, we acknowledge that model performance varies with evaluation protocols. This report only serves as a preliminary exploration of the current state of LALMs., Comment: Work in progress

IR drop estimation is now considered a first-order metric due to the concern about reliability and performance in modern electronic products. Since traditional solution involves lengthy iteration and simulation flow, how to achieve fast yet accurate estimation has become an essential demand. In this work, with the help of modern AI acceleration techniques, we propose a comprehensive solution to combine both the advantages of image-based and netlist-based features in neural network framework and obtain high-quality IR drop prediction very effectively in modern designs. A customized convolutional neural network (CNN) is developed to extract PDN features and make static IR drop estimations. Trained and evaluated with the open-source dataset, experiment results show that we have obtained the best quality in the benchmark on the problem of IR drop estimation in ICCAD CAD Contest 2023, proving the effectiveness of this important design topic., Comment: IEEE/ACM International Conference on Computer-Aided Design (ICCAD '24), October 27--31, 2024

In this work, we initiate the research on controlling nonlinear waves propagating on lattices from a completely new perspective. We consider nonlinear waves on a lattice as a system of interacting particles and study their collective flocking behavior. By designing suitable feedback controls, we show that any admissible flock can be reached within a finite amount of time. Finally, we highlight the connection between our flocking problem and a minimal-time problem in the framework of nonlinear Hamilton-Jacobi equations and optimal control theory.

Online reinforcement learning (RL) enhances policies through direct interactions with the environment, but faces challenges related to sample efficiency. In contrast, offline RL leverages extensive pre-collected data to learn policies, but often produces suboptimal results due to limited data coverage. Recent efforts have sought to integrate offline and online RL in order to harness the advantages of both approaches. However, effectively combining online and offline RL remains challenging due to issues that include catastrophic forgetting, lack of robustness and sample efficiency. In an effort to address these challenges, we introduce A3 RL , a novel method that actively selects data from combined online and offline sources to optimize policy improvement. We provide theoretical guarantee that validates the effectiveness our active sampling strategy and conduct thorough empirical experiments showing that our method outperforms existing state-of-the-art online RL techniques that utilize offline data. Our code will be publicly available at: https://github.com/xuefeng-cs/A3RL.

Exploring the equation of state of dense matter is an essential part of interpreting the observable properties of neutron stars. We present here the first results for dense matter in the zero-temperature limit generated by the MUSES Calculation Engine, a composable workflow management system that orchestrates calculation and data processing stages comprising a collection of software modules designed within the MUSES framework. The modules presented in this work calculate equations of state using algorithms spanning three different theories/models: (1) Crust Density Functional Theory, valid starting at low densities, (2) Chiral Effective Field Theory, valid around saturation density, and (3) the Chiral Mean Field model, valid beyond saturation density. Lepton contributions are added through the Lepton module to each equation of state, ensuring charge neutrality and the possibility of $\beta$-equilibrium. Using the Synthesis module, we match the three equations of state using different thermodynamic variables and different methods. We then couple the complete equation of state to a novel full-general-relativity solver (QLIMR) module that calculates neutron star properties. We find that the matching performed using different thermodynamic variables affects differently the range obtained for neutron star masses and radii (although never beyond a few percent difference). We also investigate the universality of equation of state-independent relations for our matched stars. Finally, for the first time, we use the Flavor Equilibration module to estimate bulk viscosity and flavor relaxation charge fraction and rates (at low temperature) for Chiral Effective Field Theory and the Chiral Mean Field model., Comment: 38 pages, 17 figures, 6 tables

In decision-making systems, algorithmic recourse aims to identify minimal-cost actions to alter an individual features, thereby obtaining a desired outcome. This empowers individuals to understand, question, or alter decisions that negatively affect them. However, due to the variety and sensitivity of system environments and individual personalities, quantifying the cost of a single function is nearly impossible while considering multiple criteria situations. Most current recourse mechanisms use gradient-based methods that assume cost functions are differentiable, often not applicable in real-world scenarios, resulting in sub-optimal solutions that compromise various criteria. These solutions are typically intractable and lack rigorous theoretical foundations, raising concerns regarding interpretability, reliability, and transparency from the explainable AI (XAI) perspective. To address these issues, this work proposes an algorithmic recourse framework that handles non-differentiable and discrete multi-cost functions. By formulating recourse as a multi-objective optimization problem and assigning weights to different criteria based on their importance, our method identifies Pareto optimal recourse recommendations. To demonstrate scalability, we incorporate the concept of epsilon-net, proving the ability to find approximated Pareto optimal actions. Experiments show the trade-off between different criteria and the methods scalability in large graphs. Compared to current heuristic practices, our approach provides a stronger theoretical foundation and better aligns recourse suggestions with real-world requirements.

Let $A, B$ be subsets of $(\mathbb{Z}/p^r\mathbb{Z})^2$. In this note, we provide conditions on the densities of $A$ and $B$ such that $|gA-B|\gg p^{2r}$ for a positive proportion of $g\in SO_2(\mathbb{Z}/p^r\mathbb{Z})$. The conditions are sharp up to constant factors in the unbalanced case, and the proof makes use of tools from discrete Fourier analysis and results in restriction/extension theory., Comment: 14 pages

Let $(R, \frak m)$ be a local ring of prime characteristic $p$ and of dimension $d$ with the embedding dimension $v$, type $s$ and the Frobenius test exponent for parameter ideals $\mathrm{Fte}(R)$. We will give an upper bound for the multiplicity of Cohen-Macaulay rings in prime characteristic in terms of $\mathrm{Fte}(R),d,v$ and $s$. Our result extends the main results for Gorenstein rings due to Huneke and Watanabe., Comment: Thang Long Journal of Science: Mathematics and Mathematical Sciences, Vol. 3, 17--27. arXiv admin note: substantial text overlap with arXiv:1901.00849

Environmental experts have developed the DPSIR (Driver, Pressure, State, Impact, Response) framework to systematically study and communicate key relationships between society and the environment. Using this framework requires experts to construct a DPSIR taxonomy from a corpus, annotate the documents, and identify DPSIR variables and relationships, which is laborious and inflexible. Automating it with conventional text mining faces technical challenges, primarily because the taxonomy often begins with abstract definitions, which experts progressively refine and contextualize as they annotate the corpus. In response, we develop GreenMine, a system that supports interactive text mining with prompt engineering. The system implements a prompting pipeline consisting of three simple and evaluable subtasks. In each subtask, the DPSIR taxonomy can be defined in natural language and iteratively refined as experts analyze the corpus. To support users evaluate the taxonomy, we introduce an uncertainty score based on response consistency. Then, we design a radial uncertainty chart that visualizes uncertainties and corpus topics, which supports interleaved evaluation and exploration. Using the system, experts can progressively construct the DPSIR taxonomy and annotate the corpus with LLMs. Using real-world interview transcripts, we present a case study to demonstrate the capability of the system in supporting interactive mining of DPSIR relationships, and an expert review in the form of collaborative discussion to understand the potential and limitations of the system. We discuss the lessons learned from developing the system and future opportunities for supporting interactive text mining in knowledge-intensive tasks for other application scenarios.

Recent advancements in controllable expressive speech synthesis, especially in text-to-speech (TTS) models, have allowed for the generation of speech with specific styles guided by textual descriptions, known as style prompts. While this development enhances the flexibility and naturalness of synthesized speech, there remains a significant gap in understanding how these models handle vague or abstract style prompts. This study investigates the potential gender bias in how models interpret occupation-related prompts, specifically examining their responses to instructions like "Act like a nurse". We explore whether these models exhibit tendencies to amplify gender stereotypes when interpreting such prompts. Our experimental results reveal the model's tendency to exhibit gender bias for certain occupations. Moreover, models of different sizes show varying degrees of this bias across these occupations., Comment: NAACL 2025 Findings

MIMOSIS is a CMOS Monolithic Active Pixel Sensor developed to equip the Micro Vertex Detector of the Compressed Baryonic Matter (CBM) experiment at FAIR/GSI. The sensor will combine an excellent spatial precision of $5~\mu m$ with a time resolution of $5~\mu s$ and provide a peak hit rate capability of $\mathrm{\sim 80~ MHz/cm^2}$. To fulfill its task, MIMOSIS will have to withstand ionising radiation doses of $\sim 5~ \mathrm{MRad}$ and fluences of $\sim 7 \times 10^{13}~\mathrm{n_{eq}/cm^2}$ per year of operation. This paper introduces the reticle size full feature sensor prototype MIMOSIS-2.1, which was improved with respect to earlier prototypes by adding on-chip grouping circuts and by improving the analog power grid. Moreover, it features for a first time a $50~\mu m$ epitaxial layer, which is found to improve the performances of the non-irradiated device significantly. We discuss the in beam sensor performances as measured during beam tests at the CERN-SPS., Comment: 6 pages, 7 figures

Three-dimensional scene inpainting is crucial for applications from virtual reality to architectural visualization, yet existing methods struggle with view consistency and geometric accuracy in 360{\deg} unbounded scenes. We present AuraFusion360, a novel reference-based method that enables high-quality object removal and hole filling in 3D scenes represented by Gaussian Splatting. Our approach introduces (1) depth-aware unseen mask generation for accurate occlusion identification, (2) Adaptive Guided Depth Diffusion, a zero-shot method for accurate initial point placement without requiring additional training, and (3) SDEdit-based detail enhancement for multi-view coherence. We also introduce 360-USID, the first comprehensive dataset for 360{\deg} unbounded scene inpainting with ground truth. Extensive experiments demonstrate that AuraFusion360 significantly outperforms existing methods, achieving superior perceptual quality while maintaining geometric accuracy across dramatic viewpoint changes. See our project page for video results and the dataset at https://kkennethwu.github.io/aurafusion360/., Comment: Project page: https://kkennethwu.github.io/aurafusion360/

The Micro Vertex Detector (MVD) is the most upstream detector of the fixed-target Compressed Baryonic Matter Experiment (CBM) at the future Facility for Antiproton and Ion Research (FAIR). It enables high-precision low-momentum tracking in direct proximity of the target. Reaching the stringent requirements for the MVD, a material budget of~$0.3\,-\,0.5\%\,X_0$ per layer, operating the dedicated CMOS MAPS~(`MIMOSIS') in the target vacuum, the strong magnetic dipole field, and a harsh radiation environment~(5\,Mrad, $7\times10^{13}\,n_{\text{eq}}/\text{cm}^2$ per CBM year), poses an unprecedented integration challenge. In this paper, the integration concept of the detector is be outlined, elaborating on the selection and preparation of materials, assembly procedures, and quality assessment steps in the ongoing preparation of pre-series production and detector commissioning in 2028., Comment: Titlepage + 5 pages, 9 figures, proceedings for Pixel 2024

In this paper, we investigate error bounds for cone-convex inclusion problems in finite-dimensional settings of the form $f(x)\in K$, where $K$ is a smooth cone and $f$ is a continuously differentiable and $K$-concave function. We show that local error bounds for the inclusion can be characterized by the Abadie constraint qualification around the reference point. In the case where $f$ is an affine function, we precisely identify the conditions under which the inclusion admits global error bounds. Additionally, we derive some properties of smooth cones, as well as regular cones and strictly convex cones.

Medication adherence is critical for the recovery of adolescents and young adults (AYAs) who have undergone hematopoietic cell transplantation (HCT). However, maintaining adherence is challenging for AYAs after hospital discharge, who experience both individual (e.g. physical and emotional symptoms) and interpersonal barriers (e.g., relational difficulties with their care partner, who is often involved in medication management). To optimize the effectiveness of a three-component digital intervention targeting both members of the dyad as well as their relationship, we propose a novel Multi-Agent Reinforcement Learning (MARL) approach to personalize the delivery of interventions. By incorporating the domain knowledge, the MARL framework, where each agent is responsible for the delivery of one intervention component, allows for faster learning compared with a flattened agent. Evaluation using a dyadic simulator environment, based on real clinical data, shows a significant improvement in medication adherence (approximately 3%) compared to purely random intervention delivery. The effectiveness of this approach will be further evaluated in an upcoming trial.

Rust supports type conversions and safe Rust guarantees the security of these conversions through robust static type checking and strict ownership guidelines. However, there are instances where programmers need to use unsafe Rust for certain type conversions, especially those involving pointers. Consequently, these conversions may cause severe memory corruption problems. Despite extensive research on type confusion bugs in C/C++, studies on type confusion bugs in Rust are still lacking. Also, due to Rust's new features in the type system, existing solutions in C/C++ cannot be directly applied to Rust. In this paper, we develop a static analysis tool called TYPEPULSE to detect three main categories of type confusion bugs in Rust including misalignment, inconsistent layout, and mismatched scope. TYPEPULSE first performs a type conversion analysis to collect and determine trait bounds for type pairs. Moreover, it performs a pointer alias analysis to resolve the alias relationship of pointers. Following the integration of information into the property graph, it constructs type patterns and detects each type of bug in various conversion scenarios. We run TYPEPULSE on the top 3,000 Rust packages and uncover 71 new type confusion bugs, exceeding the total number of type confusion bugs reported in RUSTSEC over the past five years. We have received 32 confirmations from developers, along with one CVE ID and six RUSTSEC IDs., Comment: To Appear in the 34th USENIX Security Symposium, August 13-15, 2025

Materials design is an important component of modern science and technology, yet traditional approaches rely heavily on trial-and-error and can be inefficient. Computational techniques, enhanced by modern artificial intelligence (AI), have greatly accelerated the design of new materials. Among these approaches, inverse design has shown great promise in designing materials that meet specific property requirements. In this mini-review, we summarize key computational advancements for materials design over the past few decades. We follow the evolution of relevant materials design techniques, from high-throughput forward machine learning (ML) methods and evolutionary algorithms, to advanced AI strategies like reinforcement learning (RL) and deep generative models. We highlight the paradigm shift from conventional screening approaches to inverse generation driven by deep generative models. Finally, we discuss current challenges and future perspectives of materials inverse design. This review may serve as a brief guide to the approaches, progress, and outlook of designing future functional materials with technological relevance., Comment: 18 pages, 7 figures, 1 table; Review article

Software testing ensures the quality and reliability of software products, but manual test case creation is labor-intensive. With the rise of large language models (LLMs), there is growing interest in unit test creation with LLMs. However, effective assessment of LLM-generated test cases is limited by the lack of standardized benchmarks that comprehensively cover diverse programming scenarios. To address the assessment of LLM's test case generation ability and lacking dataset for evaluation, we propose the Generated Benchmark from Control-Flow Structure and Variable Usage Composition (GBCV) approach, which systematically generates programs used for evaluating LLMs' test generation capabilities. By leveraging basic control-flow structures and variable usage, GBCV provides a flexible framework to create a spectrum of programs ranging from simple to complex. Because GPT-4o and GPT-3-Turbo are publicly accessible models, to present real-world regular user's use case, we use GBCV to assess LLM performance on them. Our findings indicate that GPT-4o performs better on complex program structures, while all models effectively detect boundary values in simple conditions but face challenges with arithmetic computations. This study highlights the strengths and limitations of LLMs in test generation, provides a benchmark framework, and suggests directions for future improvement., Comment: 17 pages, 9 figures

In light of breakthroughs in superconductivity under high pressure, and considering that record critical temperatures (T$_c$s) across various systems have been achieved under high pressure, the primary challenge for higher Tc should no longer solely be to increase T$_c$ under extreme conditions but also to reduce, or ideally eliminate, the need for applied pressure in retaining pressure-induced or -enhanced superconductivity. The topological semiconductor Bi$_{0.5}$Sb$_{1.5}$Te$_3$ (BST) was chosen to demonstrate our approach to addressing this challenge and exploring its intriguing physics. Under pressures up to ~ 50 GPa, three superconducting phases (BST-I, -II, and -III) were observed. A superconducting phase in BST-I appears at ~ 4 GPa, without a structural transition, suggesting the possible topological nature of this phase. Using the pressure-quench protocol (PQP) recently developed by us, we successfully retained this pressure-induced phase at ambient pressure and revealed the bulk nature of the state. Significantly, this demonstrates recovery of a pressure-quenched sample from a diamond anvil cell at room temperature with the pressure-induced phase retained at ambient pressure. Other superconducting phases were retained in BST-II and -III at ambient pressure and subjected to thermal and temporal stability testing. Superconductivity was also found in BST with T$_c$ up to 10.2 K, the record for this compound series. While PQP maintains superconducting phases in BST at ambient pressure, both depressurization and PQP enhance its T$_c$, possibly due to microstructures formed during these processes, offering an added avenue to raise T$_c$. These findings are supported by our density-functional theory calculations., Comment: 26 pages, 14 figures

The proliferation of datasets across open data portals and enterprise data lakes presents an opportunity for deriving data-driven insights. However, widely-used dataset search systems rely on keyword searches over dataset metadata, including descriptions, to facilitate discovery. When these descriptions are incomplete, missing, or inconsistent with dataset contents, findability is severely hindered. In this paper, we address the problem of automatic dataset description generation: how to generate informative descriptions that enhance dataset discovery and support relevance assessment. We introduce AutoDDG, a framework for automated dataset description generation tailored for tabular data. To derive descriptions that are comprehensive, accurate, readable and concise, AutoDDG adopts a data-driven approach to summarize the contents of a dataset, and leverages LLMs to both enrich the summaries with semantic information and to derive human-readable descriptions. An important challenge for this problem is how to evaluate the effectiveness of methods for data description generation and the quality of the descriptions. We propose a multi-pronged evaluation strategy that: (1) measures the improvement in dataset retrieval within a dataset search engine, (2) compares generated descriptions to existing ones (when available), and (3) evaluates intrinsic quality metrics such as readability, faithfulness to the data, and conciseness. Additionally, we introduce two new benchmarks to support this evaluation. Our experimental results, using these benchmarks, demonstrate that AutoDDG generates high-quality, accurate descriptions and significantly improves dataset retrieval performance across diverse use cases.

Task vectors, which are derived from the difference between pre-trained and fine-tuned model weights, enable flexible task adaptation and model merging through arithmetic operations such as addition and negation. However, existing approaches often rely on heuristics with limited theoretical support, often leading to performance gaps comparing to direct task fine tuning. Meanwhile, although it is easy to manipulate saved task vectors with arithmetic for different purposes, such compositional flexibility demands high memory usage, especially when dealing with a huge number of tasks, limiting scalability. This work addresses these issues with a theoretically grounded framework that explains task vector arithmetic and introduces the task vector bases framework. Building upon existing task arithmetic literature, our method significantly reduces the memory cost for downstream arithmetic with little effort, while achieving competitive performance and maintaining compositional advantage, providing a practical solution for large-scale task arithmetic., Comment: 25 pages, 11 figures

Efficient real-time trajectory planning and control for fixed-wing unmanned aerial vehicles is challenging due to their non-holonomic nature, complex dynamics, and the additional uncertainties introduced by unknown aerodynamic effects. In this paper, we present a fast and efficient real-time trajectory planning and control approach for fixed-wing unmanned aerial vehicles, leveraging the differential flatness property of fixed-wing aircraft in coordinated flight conditions to generate dynamically feasible trajectories. The approach provides the ability to continuously replan trajectories, which we show is useful to dynamically account for the curvature constraint as the aircraft advances along its path. Extensive simulations and real-world experiments validate our approach, showcasing its effectiveness in generating trajectories even in challenging conditions for small FW such as wind disturbances., Comment: Approved at Icra 25

In Transformer-based sequence-to-sequence generation, beam search has proven effective in enhancing the quality of generated sequences compared to greedy decoding. Conventional beam search methods typically adopt either a sequential or batch-based approach. The sequential approach, while memory-efficient, requires multiple decoding passes to construct a complete search tree, leading to significantly slower inference. On the other hand, the batch-based approach enables parallel computation across beams, but at the expense of high memory consumption due to the need to maintain separate key-value (KV) caches for each beam. In this study, we introduce a novel trie (prefix-tree)-based parallel decoding method that addresses the memory inefficiency of batch-based beam search. By sharing a single KV cache among all beams that share the same prefix, the proposed method not only reduces memory consumption dramatically but also enables parallel decoding across all branches. This innovative use of a prefix tree offers an efficient alternative for beam search, achieving significant memory savings while preserving inference speed, making it particularly well-suited for memory-constrained environments or large-scale model deployments., Comment: 9 pages

Inverse design enables automating the discovery and optimization of devices achieving performance significantly exceeding that of traditional human-engineered designs. However, existing methodologies to inverse design electromagnetic devices require computationally expensive and time-consuming full-wave electromagnetic simulation at each inverse design iteration or generation of large datasets for training neural-network surrogate models. This work introduces the Precomputed Numerical Green Function method, an approach for ultrafast electromagnetic inverse design. The static components of the design are incorporated into a numerical Green function obtained from a single fully parallelized precomputation step, reducing the cost of evaluating candidate designs during optimization to only being proportional to the size of the region under modification. A low-rank matrix update technique is introduced that further decreases the cost of the method to less than a second per iteration without any approximations or compromises in accuracy. The complete method is shown to have linear time complexity, reducing the total runtime for an inverse design by several orders of magnitude compared to using conventional Maxwell solvers. The design examples considered demonstrate speedups of up to 700x, lowering the design process from multiple days to weeks down to less than an hour. The approach stands to transform inverse design in electromagnetics.

We present BreezyVoice, a Text-to-Speech (TTS) system specifically adapted for Taiwanese Mandarin, highlighting phonetic control abilities to address the unique challenges of polyphone disambiguation in the language. Building upon CosyVoice, we incorporate a $S^{3}$ tokenizer, a large language model (LLM), an optimal-transport conditional flow matching model (OT-CFM), and a grapheme to phoneme prediction model, to generate realistic speech that closely mimics human utterances. Our evaluation demonstrates BreezyVoice's superior performance in both general and code-switching contexts, highlighting its robustness and effectiveness in generating high-fidelity speech. Additionally, we address the challenges of generalizability in modeling long-tail speakers and polyphone disambiguation. Our approach significantly enhances performance and offers valuable insights into the workings of neural codec TTS systems.

Neural Architecture Search (NAS) aims to automate the design of deep neural networks. However, existing NAS techniques often focus on maximising accuracy, neglecting model efficiency. This limitation restricts their use in resource-constrained environments like mobile devices and edge computing systems. Moreover, current evaluation metrics prioritise performance over efficiency, lacking a balanced approach for assessing architectures suitable for constrained scenarios. To address these challenges, this paper introduces the M-factor, a novel metric combining model accuracy and size. Four diverse NAS techniques are compared: Policy-Based Reinforcement Learning, Regularised Evolution, Tree-structured Parzen Estimator (TPE), and Multi-trial Random Search. These techniques represent different NAS paradigms, providing a comprehensive evaluation of the M-factor. The study analyses ResNet configurations on the CIFAR-10 dataset, with a search space of 19,683 configurations. Experiments reveal that Policy-Based Reinforcement Learning and Regularised Evolution achieved M-factor values of 0.84 and 0.82, respectively, while Multi-trial Random Search attained 0.75, and TPE reached 0.67. Policy-Based Reinforcement Learning exhibited performance changes after 39 trials, while Regularised Evolution optimised within 20 trials. The research investigates the optimisation dynamics and trade-offs between accuracy and model size for each strategy. Findings indicate that, in some cases, random search performed comparably to more complex algorithms when assessed using the M-factor. These results highlight how the M-factor addresses the limitations of existing metrics by guiding NAS towards balanced architectures, offering valuable insights for selecting strategies in scenarios requiring both performance and efficiency.

Next-frame prediction in videos is crucial for applications such as autonomous driving, object tracking, and motion prediction. The primary challenge in next-frame prediction lies in effectively capturing and processing both spatial and temporal information from previous video sequences. The transformer architecture, known for its prowess in handling sequence data, has made remarkable progress in this domain. However, transformer-based next-frame prediction models face notable issues: (a) The multi-head self-attention (MHSA) mechanism requires the input embedding to be split into $N$ chunks, where $N$ is the number of heads. Each segment captures only a fraction of the original embeddings information, which distorts the representation of the embedding in the latent space, resulting in a semantic dilution problem; (b) These models predict the embeddings of the next frames rather than the frames themselves, but the loss function based on the errors of the reconstructed frames, not the predicted embeddings -- this creates a discrepancy between the training objective and the model output. We propose a Semantic Concentration Multi-Head Self-Attention (SCMHSA) architecture, which effectively mitigates semantic dilution in transformer-based next-frame prediction. Additionally, we introduce a loss function that optimizes SCMHSA in the latent space, aligning the training objective more closely with the model output. Our method demonstrates superior performance compared to the original transformer-based predictors.

Normal crystals (also known as Stembridge crystals) are commonly used to establish the Schur positivity of symmetric functions, as their characters are sums of Schur polynomials. In this paper, we develop a combinatorial framework for a novel family of objects called normal square root crystals, which are closely related to symmetric Grothendieck functions, the $K$-theoretic analogue of Schur functions. Among other applications, this tool leads to a new proof of Buch's combinatorial rule for the multiplication of symmetric Grothendieck functions. The definition of a normal square root crystal, originally formulated by the first two authors, largely mirrors that of normal crystals. Our main result is to show that the character of such a crystal is always a sum of symmetric Grothendieck polynomials. The proof relies on an unexpected connection between the raising operators for our crystals and the Hecke insertion algorithm developed by Buch, Kresch, Shimozono, Tamvakis, and Yong., Comment: 36 pages, 3 figures

Play is pivotal in fostering the emotional, social, and cultural dimensions of urban spaces. While generative AI (GAI) potentially supports playful urban interaction, a balanced and critical approach to the design opportunities and challenges is needed. This work develops iWonder, an image-to-image GAI tool engaging fourteen designers in urban explorations to identify GAI's playful features and create design ideas. Fourteen citizens then evaluated these ideas, providing expectations and critical concerns from a bottom-up perspective. Our findings reveal the dynamic interplay between users, GAI, and urban contexts, highlighting GAI's potential to facilitate playful urban experiences through generative agency, meaningful unpredictability, social performativity, and the associated offensive qualities. We propose design considerations to address citizen concerns and the `tourist metaphor' to deepen our understanding of GAI's impact, offering insights to enhance cities' socio-cultural fabric. Overall, this research contributes to the effort to harness GAI's capabilities for urban enrichment., Comment: Accepted for publication in the 2025 ACM CHI Conference on Human Factors in Computing Systems (CHI'25)

Matsuki and Oshima introduced the notion of clans, which are incomplete matchings between $[1,n]$ with positive or negative signs on isolated vertices. They discovered that clans can parametrise $K$-orbits in the flag varieties for classical linear groups, where $K$ is a fixed point subgroup of an involution in the same classical linear group. Yamamoto gave a full proof for the type AIII $\textsf{GL}_p(\mathbb{C}) \times \textsf{GL}_q(\mathbb{C})$-orbits in type A flag variety. In this work we investigate the affine version of type AIII orbits. For a field $\mathbb{K}$ with characteristic not equal to two, we construct explicit bijections between the $\textsf{GL}_p(\mathbb{K}(\hspace{-0.5mm}(t)\hspace{-0.5mm})) \times \textsf{GL}_q(\mathbb{K}(\hspace{-0.5mm}(t)\hspace{-0.5mm}))$-orbits in the affine flag variety and certain objects called affine $(p,q)$-clans. These affine $(p,q)$-clans can be concretely interpreted as involutions in the affine permutation group with positive or negative signs in fixed points., Comment: 43 pages, 2 tables

In this paper, we employ a Schauder-type estimate method, as developed in \cite{CHN}, to establish critical well-posedness result for the Fractional Fokker-Planck Equation. This equation serves as a fundamental model in kinetic theory and can be regarded as a semi-linear analogue of the non-cutoff Boltzmann equation. We demonstrate that the techniques introduced in this study are not only effective for the FFPE but also hold promise for broader applications, particularly in addressing the non-cutoff Boltzmann equation and the Landau equation. Our results contribute to a deeper understanding of the analytical framework required for these complex kinetic models.

Y$_2$Co$_3$ is a newly discovered antiferromagnetic (AFM) compound with distorted kagome layers. Previous investigations via bulk magnetization measurements suggested a complex noncollinear magnetic behavior, with magnetic moments primarily anti-aligned along the $b$ axis and some canting towards the $ac$ plane. In this study, we report the magnetic structure of Y$_2$Co$_3$ to be an A-type AFM structure with ferromagnetic (FM) interactions within the distorted kagome plane and an interplane antiferromagnetic interaction, as determined by single-crystal neutron diffraction. The magnetic moments align along the $b$ axis, with minimal canting towards the $c$ axis, at odds with the previous interpretation of bulk magnetization measurements. The magnetic moments on the two distinct Co sites are [0, -0.68(3), 0] $\mu_B$ and [0, 1.25(4), 0.07(1)] $\mu_B$. We attribute the previously reported "noncollinear" behavior to the considerable temperature dependence of itinerant AFM exchange interactions, induced by thermal contraction along the $b$ axis. Additionally, our examination of lattice constants through pressure studies reveals compensating effects on FM and AFM interactions, resulting in negligible pressure dependence of $T_\textrm{N}$.

Decentralized Multi-Agent Reinforcement Learning (Dec-MARL) has emerged as a pivotal approach for addressing complex tasks in dynamic environments. Existing Multi-Agent Reinforcement Learning (MARL) methodologies typically assume a shared objective among agents and rely on centralized control. However, many real-world scenarios feature agents with individual goals and limited observability of other agents, complicating coordination and hindering adaptability. Existing Dec-MARL strategies prioritize either communication or coordination, lacking an integrated approach that leverages both. This paper presents a novel Dec-MARL framework that integrates peer-to-peer communication and coordination, incorporating goal-awareness and time-awareness into the agents' knowledge-sharing processes. Our framework equips agents with the ability to (i) share contextually relevant knowledge to assist other agents, and (ii) reason based on information acquired from multiple agents, while considering their own goals and the temporal context of prior knowledge. We evaluate our approach through several complex multi-agent tasks in environments with dynamically appearing obstacles. Our work demonstrates that incorporating goal-aware and time-aware knowledge sharing significantly enhances overall performance.

In today's fast-evolving business landscape, having insight into the technology stacks that organizations use is crucial for forging partnerships, uncovering market openings, and informing strategic choices. However, conventional technology mapping, which typically hinges on keyword searches, struggles with the sheer scale and variety of data available, often failing to capture nascent technologies. To overcome these hurdles, we present STARS (Semantic Technology and Retrieval System), a novel framework that harnesses Large Language Models (LLMs) and Sentence-BERT to pinpoint relevant technologies within unstructured content, build comprehensive company profiles, and rank each firm's technologies according to their operational importance. By integrating entity extraction with Chain-of-Thought prompting and employing semantic ranking, STARS provides a precise method for mapping corporate technology portfolios. Experimental results show that STARS markedly boosts retrieval accuracy, offering a versatile and high-performance solution for cross-industry technology mapping., Comment: Technical Report

Maintaining consistent model performance across domains is a fundamental challenge in machine learning. While recent work has explored using LLM-generated data for fine-tuning, its impact on cross-domain generalization remains poorly understood. In this paper, we present a systematic analysis revealing that fine-tuning with LLM-generated data not only improves target task performance but also reduces out-of-domain (OOD) degradation compared to fine-tuning with ground truth data. Through analyzing the data sequence in tasks of various domains, we demonstrate that this enhanced OOD robustness stems from a reduced prevalence of high perplexity tokens in LLM-generated sequences. Following this hypothesis we showed that masking high perplexity tokens in ground truth training data also achieves similar OOD preservation comparable to using LLM-generated data. Extensive experiments across diverse model architectures and scales, including Gemma2-2B, Mistral-7B and Llama3-8B, corroborate the consistency of our findings. To the best of our knowledge, this work provides the first mechanistic explanation for the superior OOD robustness conferred by LLM-generated training data, offering valuable insights for developing more robust fine-tuning strategies.

The widespread use of Multi-layer perceptrons (MLPs) often relies on a fixed activation function (e.g., ReLU, Sigmoid, Tanh) for all nodes within the hidden layers. While effective in many scenarios, this uniformity may limit the networks ability to capture complex data patterns. We argue that employing the same activation function at every node is suboptimal and propose leveraging different activation functions at each node to increase flexibility and adaptability. To achieve this, we introduce Local Control Networks (LCNs), which leverage B-spline functions to enable distinct activation curves at each node. Our mathematical analysis demonstrates the properties and benefits of LCNs over conventional MLPs. In addition, we demonstrate that more complex architectures, such as Kolmogorov-Arnold Networks (KANs), are unnecessary in certain scenarios, and LCNs can be a more efficient alternative. Empirical experiments on various benchmarks and datasets validate our theoretical findings. In computer vision tasks, LCNs achieve marginal improvements over MLPs and outperform KANs by approximately 5\%, while also being more computationally efficient than KANs. In basic machine learning tasks, LCNs show a 1\% improvement over MLPs and a 0.6\% improvement over KANs. For symbolic formula representation tasks, LCNs perform on par with KANs, with both architectures outperforming MLPs. Our findings suggest that diverse activations at the node level can lead to improved performance and efficiency.