id
stringlengths
9
16
title
stringlengths
4
278
categories
stringlengths
5
104
abstract
stringlengths
6
4.09k
2501.10642
Iterative Tree Analysis for Medical Critics
cs.CL
Large Language Models (LLMs) have been widely adopted across various domains, yet their application in the medical field poses unique challenges, particularly concerning the generation of hallucinations. Hallucinations in open-ended long medical text manifest as misleading critical claims, which are difficult to verify due to two reasons. First, critical claims are often deeply entangled within the text and cannot be extracted based solely on surface-level presentation. Second, verifying these claims is challenging because surface-level token-based retrieval often lacks precise or specific evidence, leaving the claims unverifiable without deeper mechanism-based analysis. In this paper, we introduce a novel method termed Iterative Tree Analysis (ITA) for medical critics. ITA is designed to extract implicit claims from long medical texts and verify each claim through an iterative and adaptive tree-like reasoning process. This process involves a combination of top-down task decomposition and bottom-up evidence consolidation, enabling precise verification of complex medical claims through detailed mechanism-level reasoning. Our extensive experiments demonstrate that ITA significantly outperforms previous methods in detecting factual inaccuracies in complex medical text verification tasks by 10%. Additionally, we will release a comprehensive test set to the public, aiming to foster further advancements in research within this domain.
2501.10644
UAV-Assisted Multi-Task Federated Learning with Task Knowledge Sharing
cs.LG cs.MA
The rapid development of Unmanned aerial vehicles (UAVs) technology has spawned a wide variety of applications, such as emergency communications, regional surveillance, and disaster relief. Due to their limited battery capacity and processing power, multiple UAVs are often required for complex tasks. In such cases, a control center is crucial for coordinating their activities, which fits well with the federated learning (FL) framework. However, conventional FL approaches often focus on a single task, ignoring the potential of training multiple related tasks simultaneously. In this paper, we propose a UAV-assisted multi-task federated learning scheme, in which data collected by multiple UAVs can be used to train multiple related tasks concurrently. The scheme facilitates the training process by sharing feature extractors across related tasks and introduces a task attention mechanism to balance task performance and encourage knowledge sharing. To provide an analytical description of training performance, the convergence analysis of the proposed scheme is performed. Additionally, the optimal bandwidth allocation for UAVs under limited bandwidth conditions is derived to minimize communication time. Meanwhile, a UAV-EV association strategy based on coalition formation game is proposed. Simulation results validate the effectiveness of the proposed scheme in enhancing multi-task performance and training speed.
2501.10645
Constrained Coding for Composite DNA: Channel Capacity and Efficient Constructions
cs.IT math.IT
Composite DNA is a recent novel method to increase the information capacity of DNA-based data storage above the theoretical limit of 2 bits/symbol. In this method, every composite symbol does not store a single DNA nucleotide but a mixture of the four nucleotides in a predetermined ratio. By using different mixtures and ratios, the alphabet can be extended to have much more than four symbols in the naive approach. While this method enables higher data content per synthesis cycle, potentially reducing the DNA synthesis cost, it also imposes significant challenges for accurate DNA sequencing since the base-level errors can easily change the mixture of bases and their ratio, resulting in changes to the composite symbols. With this motivation, we propose efficient constrained coding techniques to enforce the biological constraints, including the runlength-limited constraint and the GC-content constraint, into every DNA synthesized oligo, regardless of the mixture of bases in each composite letter and their corresponding ratio. Our goals include computing the capacity of the constrained channel, constructing efficient encoders/decoders, and providing the best options for the composite letters to obtain capacity-approaching codes. For certain codes' parameters, our methods incur only one redundant symbol.
2501.10648
DNA 1.0 Technical Report
cs.CL
In this report, we present DNA 1.0 8B Instruct, a state-of-the-art bilingual language model optimized for Korean and English language tasks. By applying continual pre-training (CPT) with high-quality Korean datasets to Llama 3.1 8B and subsequent supervised fine-tuning (SFT), we create an instruction-following model with enhanced Korean language capabilities. This model is then merged with Llama 3.1 8B Instruct via spherical linear interpolation (SLERP) and undergoes further optimization through direct preference optimization (DPO) and knowledge distillation (KD). DNA 1.0 8B Instruct achieves state-of-the-art results on Korean-specific tasks, including KMMLU (53.26%), KoBEST (83.40%), and BELEBELE (57.99%), while maintaining strong English capabilities on MMLU (66.64%), MMLU-Pro (43.05%) and GSM8K (80.52%). As an open model, DNA 1.0 8B Instruct represents a significant advancement in bilingual language modeling. As an open model, DNA 1.0 8B Instruct is freely available through https://huggingface.co/dnotitia/Llama-DNA-1.0-8B-Instruct . For commercial licensing inquiries or feedback, please contact us at https://www.dnotitia.com/contact/post-form
2501.10651
MOFA: Discovering Materials for Carbon Capture with a GenAI- and Simulation-Based Workflow
cs.DC cond-mat.mtrl-sci cs.LG
We present MOFA, an open-source generative AI (GenAI) plus simulation workflow for high-throughput generation of metal-organic frameworks (MOFs) on large-scale high-performance computing (HPC) systems. MOFA addresses key challenges in integrating GPU-accelerated computing for GPU-intensive GenAI tasks, including distributed training and inference, alongside CPU- and GPU-optimized tasks for screening and filtering AI-generated MOFs using molecular dynamics, density functional theory, and Monte Carlo simulations. These heterogeneous tasks are unified within an online learning framework that optimizes the utilization of available CPU and GPU resources across HPC systems. Performance metrics from a 450-node (14,400 AMD Zen 3 CPUs + 1800 NVIDIA A100 GPUs) supercomputer run demonstrate that MOFA achieves high-throughput generation of novel MOF structures, with CO$_2$ adsorption capacities ranking among the top 10 in the hypothetical MOF (hMOF) dataset. Furthermore, the production of high-quality MOFs exhibits a linear relationship with the number of nodes utilized. The modular architecture of MOFA will facilitate its integration into other scientific applications that dynamically combine GenAI with large-scale simulations.
2501.10658
LUT-DLA: Lookup Table as Efficient Extreme Low-Bit Deep Learning Accelerator
cs.AR cs.AI cs.LG
The emergence of neural network capabilities invariably leads to a significant surge in computational demands due to expanding model sizes and increased computational complexity. To reduce model size and lower inference costs, recent research has focused on simplifying models and designing hardware accelerators using low-bit quantization. However, due to numerical representation limits, scalar quantization cannot reduce bit width lower than 1-bit, diminishing its benefits. To break through these limitations, we introduce LUT-DLA, a Look-Up Table (LUT) Deep Learning Accelerator Framework that utilizes vector quantization to convert neural network models into LUTs, achieving extreme low-bit quantization. The LUT-DLA framework facilitates efficient and cost-effective hardware accelerator designs and supports the LUTBoost algorithm, which helps to transform various DNN models into LUT-based models via multistage training, drastically cutting both computational and hardware overhead. Additionally, through co-design space exploration, LUT-DLA assesses the impact of various model and hardware parameters to fine-tune hardware configurations for different application scenarios, optimizing performance and efficiency. Our comprehensive experiments show that LUT-DLA achieves improvements in power efficiency and area efficiency with gains of $1.4$~$7.0\times$ and $1.5$~$146.1\times$, respectively, while maintaining only a modest accuracy drop. For CNNs, accuracy decreases by $0.1\%$~$3.1\%$ using the $L_2$ distance similarity, $0.1\%$~$3.4\%$ with the $L_1$ distance similarity, and $0.1\%$~$3.8\%$ when employing the Chebyshev distance similarity. For transformer-based models, the accuracy drop ranges from $1.4\%$ to $3.0\%$.
2501.10661
Unveiling the Mystery of Weight in Large Foundation Models: Gaussian Distribution Never Fades
cs.LG cs.AI cs.CL
This paper presents a pioneering exploration of the mechanisms underlying large foundation models' (LFMs) weights, aiming to simplify AI research. Through extensive observation and analysis on prevailing LFMs, we find that regardless of initialization strategies, their weights predominantly follow a Gaussian distribution, with occasional sharp, inverted T-shaped, or linear patterns. We further discover that the weights share the i.i.d. properties of Gaussian noise, and explore their direct relationship. We find that transformation weights can be derived from Gaussian noise, and they primarily serve to increase the standard deviation of pre-trained weights, with their standard deviation growing with layer depth. In other words, transformation weights broaden the acceptable deviation from the optimal weights, facilitating adaptation to downstream tasks. Building upon the above conclusions, we thoroughly discussed the nature of optimal weights, ultimately concluding that they should exhibit zero-mean, symmetry, and sparsity, with the sparse values being a truncated Gaussian distribution and a few outliers. Our experiments in LFM adaptation and editing demonstrate the effectiveness of these insights. We hope these findings can provide a foundational understanding to pave the way for future advancements in the LFM community.
2501.10663
PB-NBV: Efficient Projection-Based Next-Best-View Planning Framework for Reconstruction of Unknown Objects
cs.RO
Completely capturing the three-dimensional (3D) data of an object is essential in industrial and robotic applications. The task of next-best-view (NBV) planning is to calculate the next optimal viewpoint based on the current data, gradually achieving a complete 3D reconstruction of the object. However, many existing NBV planning algorithms incur heavy computational costs due to the extensive use of ray-casting. Specifically, this framework refits different types of voxel clusters into ellipsoids based on the voxel structure. Then, the next optimal viewpoint is selected from the candidate views using a projection-based viewpoint quality evaluation function in conjunction with a global partitioning strategy. This process replaces extensive ray-casting, significantly improving the computational efficiency. Comparison experiments in the simulation environment show that our framework achieves the highest point cloud coverage with low computational time compared to other frameworks. The real-world experiments also confirm the efficiency and feasibility of the framework. Our method will be made open source to benefit the community.
2501.10666
Speech Emotion Detection Based on MFCC and CNN-LSTM Architecture
cs.SD cs.LG eess.AS
Emotion detection techniques have been applied to multiple cases mainly from facial image features and vocal audio features, of which the latter aspect is disputed yet not only due to the complexity of speech audio processing but also the difficulties of extracting appropriate features. Part of the SAVEE and RAVDESS datasets are selected and combined as the dataset, containing seven sorts of common emotions (i.e. happy, neutral, sad, anger, disgust, fear, and surprise) and thousands of samples. Based on the Librosa package, this paper processes the initial audio input into waveplot and spectrum for analysis and concentrates on multiple features including MFCC as targets for feature extraction. The hybrid CNN-LSTM architecture is adopted by virtue of its strong capability to deal with sequential data and time series, which mainly consists of four convolutional layers and three long short-term memory layers. As a result, the architecture achieved an accuracy of 61.07% comprehensively for the test set, among which the detection of anger and neutral reaches a performance of 75.31% and 71.70% respectively. It can also be concluded that the classification accuracy is dependent on the properties of emotion to some extent, with frequently-used and distinct-featured emotions having less probability to be misclassified into other categories. Emotions like surprise whose meaning depends on the specific context are more likely to confuse with positive or negative emotions, and negative emotions also have a possibility to get mixed with each other.
2501.10667
Precision Adaptive Imputation Network : An Unified Technique for Mixed Datasets
cs.LG stat.ML
The challenge of missing data remains a significant obstacle across various scientific domains, necessitating the development of advanced imputation techniques that can effectively address complex missingness patterns. This study introduces the Precision Adaptive Imputation Network (PAIN), a novel algorithm designed to enhance data reconstruction by dynamically adapting to diverse data types, distributions, and missingness mechanisms. PAIN employs a tri-step process that integrates statistical methods, random forests, and autoencoders, ensuring balanced accuracy and efficiency in imputation. Through rigorous evaluation across multiple datasets, including those characterized by high-dimensional and correlated features, PAIN consistently outperforms traditional imputation methods, such as mean and median imputation, as well as other advanced techniques like MissForest. The findings highlight PAIN's superior ability to preserve data distributions and maintain analytical integrity, particularly in complex scenarios where missingness is not completely at random. This research not only contributes to a deeper understanding of missing data reconstruction but also provides a critical framework for future methodological innovations in data science and machine learning, paving the way for more effective handling of mixed-type datasets in real-world applications.
2501.10668
MappedTrace: Tracing Pointer Remotely with Compiler-generated Maps
cs.PL cs.CL
Existing precise pointer tracing methods introduce substantial runtime overhead to the program being traced and are applicable only at specific program execution points. We propose MappedTrace that leverages compiler-generated read-only maps to accurately identify all pointers in any given snapshot of a program's execution state. The maps record the locations and types of pointers, allowing the tracer to precisely identify pointers without requiring the traced program to maintain bookkeeping data structures or poll at safe points, thereby reducing runtime overhead. By running the tracer from a different address space or machine, MappedTrace presents new opportunities to improve memory management techniques like memory leak detection and enables novel use cases such as infinite memory abstraction for resource-constrained environments.
2501.10670
Computing Capacity-Cost Functions for Continuous Channels in Wasserstein Space
cs.IT eess.SP math.IT math.OC
This paper investigates the problem of computing capacity-cost (C-C) functions for continuous channels. Motivated by the Kullback-Leibler divergence (KLD) proximal reformulation of the classical Blahut-Arimoto (BA) algorithm, the Wasserstein distance is introduced to the proximal term for the continuous case, resulting in an iterative algorithm related to the Wasserstein gradient descent. Practical implementation involves moving particles along the negative gradient direction of the objective function's first variation in the Wasserstein space and approximating integrals by the importance sampling (IS) technique. Such formulation is also applied to the rate-distortion (R-D) function for continuous source spaces and thus provides a unified computation framework for both problems.
2501.10672
Homotopical Entropy
math.CT cs.IT math-ph math.IT math.MP
We present a "homotopification" of fundamental concepts from information theory. Using homotopy type theory, we define homotopy types that behave analogously to probability spaces, random variables, and the exponentials of Shannon entropy and relative entropy. The original analytic theories emerge through homotopy cardinality, which maps homotopy types to real numbers and generalizes the cardinality of sets.
2501.10673
Hybrid-Quantum Neural Architecture Search for The Proximal Policy Optimization Algorithm
quant-ph cs.LG cs.NE
Recent studies in quantum machine learning advocated the use of hybrid models to assist with the limitations of the currently existing Noisy Intermediate Scale Quantum (NISQ) devices, but what was missing from most of them was the explanations and interpretations of the choices that were made to pick those exact architectures and the differentiation between good and bad hybrid architectures, this research attempts to tackle that gap in the literature by using the Regularized Evolution algorithm to search for the optimal hybrid classical-quantum architecture for the Proximal Policy Optimization (PPO) algorithm, a well-known reinforcement learning algorithm, ultimately the classical models dominated the leaderboard with the best hybrid model coming in eleventh place among all unique models, while we also try to explain the factors that contributed to such results,and for some models to behave better than others in hope to grasp a better intuition about what we should consider good practices for designing an efficient hybrid architecture.
2501.10674
Can Multimodal LLMs do Visual Temporal Understanding and Reasoning? The answer is No!
cs.CV cs.CL
Multimodal Large Language Models (MLLMs) have achieved significant advancements in tasks like Visual Question Answering (VQA) by leveraging foundational Large Language Models (LLMs). However, their abilities in specific areas such as visual temporal understanding, which is crucial for comprehending real-world dynamics, remain underexplored. To address this, we propose a challenging evaluation benchmark named TemporalVQA, consisting of two parts: 1) Temporal Order Understanding and 2) Time-lapse Estimation. The first part requires MLLMs to determine the sequence of events by analyzing temporally consecutive video frames. The second part presents image pairs with varying time differences, framed as multiple-choice questions, asking MLLMs to estimate the time-lapse between images with options ranging from seconds to years. Our evaluations of advanced MLLMs, including models like GPT-4o and Gemini-1.5-Pro, reveal significant challenges: GPT-4o achieved only 49.1% average consistent accuracy in temporal order task and 70% in time-lapse estimation, with open-source models performing even poorly. These findings underscore the limitations of current MLLMs in visual temporal understanding and reasoning, highlighting the need for further improvements for their temporal capability. Our dataset can be found at https://huggingface.co/datasets/fazliimam/temporal-vqa.
2501.10677
Class-Imbalanced-Aware Adaptive Dataset Distillation for Scalable Pretrained Model on Credit Scoring
cs.LG cs.AI q-fin.RM
The advent of artificial intelligence has significantly enhanced credit scoring technologies. Despite the remarkable efficacy of advanced deep learning models, mainstream adoption continues to favor tree-structured models due to their robust predictive performance on tabular data. Although pretrained models have seen considerable development, their application within the financial realm predominantly revolves around question-answering tasks and the use of such models for tabular-structured credit scoring datasets remains largely unexplored. Tabular-oriented large models, such as TabPFN, has made the application of large models in credit scoring feasible, albeit can only processing with limited sample sizes. This paper provides a novel framework to combine tabular-tailored dataset distillation technique with the pretrained model, empowers the scalability for TabPFN. Furthermore, though class imbalance distribution is the common nature in financial datasets, its influence during dataset distillation has not been explored. We thus integrate the imbalance-aware techniques during dataset distillation, resulting in improved performance in financial datasets (e.g., a 2.5% enhancement in AUC). This study presents a novel framework for scaling up the application of large pretrained models on financial tabular datasets and offers a comparative analysis of the influence of class imbalance on the dataset distillation process. We believe this approach can broaden the applications and downstream tasks of large models in the financial domain.
2501.10684
Deep Operator Networks for Bayesian Parameter Estimation in PDEs
cs.LG cs.CE stat.ML
We present a novel framework combining Deep Operator Networks (DeepONets) with Physics-Informed Neural Networks (PINNs) to solve partial differential equations (PDEs) and estimate their unknown parameters. By integrating data-driven learning with physical constraints, our method achieves robust and accurate solutions across diverse scenarios. Bayesian training is implemented through variational inference, allowing for comprehensive uncertainty quantification for both aleatoric and epistemic uncertainties. This ensures reliable predictions and parameter estimates even in noisy conditions or when some of the physical equations governing the problem are missing. The framework demonstrates its efficacy in solving forward and inverse problems, including the 1D unsteady heat equation and 2D reaction-diffusion equations, as well as regression tasks with sparse, noisy observations. This approach provides a computationally efficient and generalizable method for addressing uncertainty quantification in PDE surrogate modeling.
2501.10685
Harnessing the Potential of Large Language Models in Modern Marketing Management: Applications, Future Directions, and Strategic Recommendations
cs.CL
Large Language Models (LLMs) have revolutionized the process of customer engagement, campaign optimization, and content generation, in marketing management. In this paper, we explore the transformative potential of LLMs along with the current applications, future directions, and strategic recommendations for marketers. In particular, we focus on LLMs major business drivers such as personalization, real-time-interactive customer insights, and content automation, and how they enable customers and business outcomes. For instance, the ethical aspects of AI with respect to data privacy, transparency, and mitigation of bias are also covered, with the goal of promoting responsible use of the technology through best practices and the use of new technologies businesses can tap into the LLM potential, which help growth and stay one step ahead in the turmoil of digital marketing. This article is designed to give marketers the necessary guidance by using best industry practices to integrate these powerful LLMs into their marketing strategy and innovation without compromising on the ethos of their brand.
2501.10687
EMO2: End-Effector Guided Audio-Driven Avatar Video Generation
cs.CV
In this paper, we propose a novel audio-driven talking head method capable of simultaneously generating highly expressive facial expressions and hand gestures. Unlike existing methods that focus on generating full-body or half-body poses, we investigate the challenges of co-speech gesture generation and identify the weak correspondence between audio features and full-body gestures as a key limitation. To address this, we redefine the task as a two-stage process. In the first stage, we generate hand poses directly from audio input, leveraging the strong correlation between audio signals and hand movements. In the second stage, we employ a diffusion model to synthesize video frames, incorporating the hand poses generated in the first stage to produce realistic facial expressions and body movements. Our experimental results demonstrate that the proposed method outperforms state-of-the-art approaches, such as CyberHost and Vlogger, in terms of both visual quality and synchronization accuracy. This work provides a new perspective on audio-driven gesture generation and a robust framework for creating expressive and natural talking head animations.
2501.10688
Neural Algorithmic Reasoning for Hypergraphs with Looped Transformers
cs.LG cs.AI cs.CC cs.CL
Looped Transformers have shown exceptional neural algorithmic reasoning capability in simulating traditional graph algorithms, but their application to more complex structures like hypergraphs remains underexplored. Hypergraphs generalize graphs by modeling higher-order relationships among multiple entities, enabling richer representations but introducing significant computational challenges. In this work, we extend the Loop Transformer architecture's neural algorithmic reasoning capability to simulate hypergraph algorithms, addressing the gap between neural networks and combinatorial optimization over hypergraphs. Specifically, we propose a novel degradation mechanism for reducing hypergraphs to graph representations, enabling the simulation of graph-based algorithms, such as Dijkstra's shortest path. Furthermore, we introduce a hyperedge-aware encoding scheme to simulate hypergraph-specific algorithms, exemplified by Helly's algorithm. We establish theoretical guarantees for these simulations, demonstrating the feasibility of processing high-dimensional and combinatorial data using Loop Transformers. This work highlights the potential of Transformers as general-purpose algorithmic solvers for structured data.
2501.10690
Insights from the application of nonlinear model predictive control to a cart-pendulum
eess.SY cs.SY
Inspired greatly by Mills et al. (2009) and the solution within, this paper aims to more clearly explain the mathematics and implementation details of such a powerful control algorithm. While the aforementioned paper is well written and of sound mathematics, it is extreamly dense and requires some time and patience to decipher, especially as it draws on many other sources to complete the algorithm. This dense property is a clear result of the paper being restricted to the brief form and important details being ommited as a result. We provide the much needed elaboration here for the benifit of the reader.
2501.10692
Multi-modal Fusion and Query Refinement Network for Video Moment Retrieval and Highlight Detection
cs.CV
Given a video and a linguistic query, video moment retrieval and highlight detection (MR&HD) aim to locate all the relevant spans while simultaneously predicting saliency scores. Most existing methods utilize RGB images as input, overlooking the inherent multi-modal visual signals like optical flow and depth. In this paper, we propose a Multi-modal Fusion and Query Refinement Network (MRNet) to learn complementary information from multi-modal cues. Specifically, we design a multi-modal fusion module to dynamically combine RGB, optical flow, and depth map. Furthermore, to simulate human understanding of sentences, we introduce a query refinement module that merges text at different granularities, containing word-, phrase-, and sentence-wise levels. Comprehensive experiments on QVHighlights and Charades datasets indicate that MRNet outperforms current state-of-the-art methods, achieving notable improvements in MR-mAP@Avg (+3.41) and HD-HIT@1 (+3.46) on QVHighlights.
2501.10693
Distributionally Robust Policy Evaluation and Learning for Continuous Treatment with Observational Data
cs.AI cs.LG
Using offline observational data for policy evaluation and learning allows decision-makers to evaluate and learn a policy that connects characteristics and interventions. Most existing literature has focused on either discrete treatment spaces or assumed no difference in the distributions between the policy-learning and policy-deployed environments. These restrict applications in many real-world scenarios where distribution shifts are present with continuous treatment. To overcome these challenges, this paper focuses on developing a distributionally robust policy under a continuous treatment setting. The proposed distributionally robust estimators are established using the Inverse Probability Weighting (IPW) method extended from the discrete one for policy evaluation and learning under continuous treatments. Specifically, we introduce a kernel function into the proposed IPW estimator to mitigate the exclusion of observations that can occur in the standard IPW method to continuous treatments. We then provide finite-sample analysis that guarantees the convergence of the proposed distributionally robust policy evaluation and learning estimators. The comprehensive experiments further verify the effectiveness of our approach when distribution shifts are present.
2501.10694
Energy Efficiency Maximization for Movable Antenna-Enhanced System Based on Statistical CSI
cs.IT eess.SP math.IT
This paper investigates an innovative movable antenna (MA)-enhanced multiple-input multiple-output (MIMO) system designed to enhance communication performance. We aim to maximize the energy efficiency (EE) under statistical channel state information (S-CSI) through a joint optimization of the transmit covariance matrix and the antenna position vectors (APVs). To solve the stochastic problem, we consider the large number of antennas scenario and resort to deterministic equivalent (DE) technology to reformulate the system EE w.r.t. the transmit variables, i.e., the transmit covariance matrix and APV, and the receive variables, i.e., the receive APV, respectively. Then, we propose an alternative optimization (AO) algorithm to update the transmit variables and the receive variables to maximize the system EE, respectively. Our numerical results reveal that, the proposed MA-enhanced system can significantly improve EE compared to several benchmark schemes and the optimal performance can be achieved with a finite size of movement regions for MAs.
2501.10695
Exploring Transferable Homogeneous Groups for Compositional Zero-Shot Learning
cs.CV
Conditional dependency present one of the trickiest problems in Compositional Zero-Shot Learning, leading to significant property variations of the same state (object) across different objects (states). To address this problem, existing approaches often adopt either all-to-one or one-to-one representation paradigms. However, these extremes create an imbalance in the seesaw between transferability and discriminability, favoring one at the expense of the other. Comparatively, humans are adept at analogizing and reasoning in a hierarchical clustering manner, intuitively grouping categories with similar properties to form cohesive concepts. Motivated by this, we propose Homogeneous Group Representation Learning (HGRL), a new perspective formulates state (object) representation learning as multiple homogeneous sub-group representation learning. HGRL seeks to achieve a balance between semantic transferability and discriminability by adaptively discovering and aggregating categories with shared properties, learning distributed group centers that retain group-specific discriminative features. Our method integrates three core components designed to simultaneously enhance both the visual and prompt representation capabilities of the model. Extensive experiments on three benchmark datasets validate the effectiveness of our method.
2501.10696
Algorithmic Derivation of Human Spatial Navigation Indices From Eye Movement Data
cs.HC cs.AI
Spatial navigation is a complex cognitive function involving sensory inputs, such as visual, auditory, and proprioceptive information, to understand and move within space. This ability allows humans to create mental maps, navigate through environments, and process directional cues, crucial for exploring new places and finding one's way in unfamiliar surroundings. This study takes an algorithmic approach to extract indices relevant to human spatial navigation using eye movement data. Leveraging electrooculography signals, we analyzed statistical features and applied feature engineering techniques to study eye movements during navigation tasks. The proposed work combines signal processing and machine learning approaches to develop indices for navigation and orientation, spatial anxiety, landmark recognition, path survey, and path route. The analysis yielded five subscore indices with notable accuracy. Among these, the navigation and orientation subscore achieved an R2 score of 0.72, while the landmark recognition subscore attained an R2 score of 0.50. Additionally, statistical features highly correlated with eye movement metrics, including blinks, saccades, and fixations, were identified. The findings of this study can lead to more cognitive assessments and enable early detection of spatial navigation impairments, particularly among individuals at risk of cognitive decline.
2501.10698
An Interpretable Neural Control Network with Adaptable Online Learning for Sample Efficient Robot Locomotion Learning
cs.RO cs.LG
Robot locomotion learning using reinforcement learning suffers from training sample inefficiency and exhibits the non-understandable/black-box nature. Thus, this work presents a novel SME-AGOL to address such problems. Firstly, Sequential Motion Executor (SME) is a three-layer interpretable neural network, where the first produces the sequentially propagating hidden states, the second constructs the corresponding triangular bases with minor non-neighbor interference, and the third maps the bases to the motor commands. Secondly, the Adaptable Gradient-weighting Online Learning (AGOL) algorithm prioritizes the update of the parameters with high relevance score, allowing the learning to focus more on the highly relevant ones. Thus, these two components lead to an analyzable framework, where each sequential hidden state/basis represents the learned key poses/robot configuration. Compared to state-of-the-art methods, the SME-AGOL requires 40% fewer samples and receives 150% higher final reward/locomotion performance on a simulated hexapod robot, while taking merely 10 minutes of learning time from scratch on a physical hexapod robot. Taken together, this work not only proposes the SME-AGOL for sample efficient and understandable locomotion learning but also emphasizes the potential exploitation of interpretability for improving sample efficiency and learning performance.
2501.10700
Subcodes of Second-Order Reed-Muller Codes via Recursive Subproducts
cs.IT math.IT
We use a simple construction called `recursive subproducts' (that is known to yield good codes of lengths $n^m$, $n \geq 3$) to identify a family of codes sandwiched between first-order and second-order Reed-Muller (RM) codes. These codes are subcodes of multidimensional product codes that use first-order RM codes as components. We identify the minimum weight codewords of all the codes in this family, and numerically determine the weight distribution of some of them. While these codes have the same minimum distance and a smaller rate than second-order RM codes, they have significantly fewer minimum weight codewords. Further, these codes can be decoded via modifications to known RM decoders which yield codeword error rates within 0.25 dB of second-order RM codes and better than CRC-aided Polar codes (in terms of $E_b/N_o$ for lengths $256, 512, 1024$), thereby offering rate adaptation options for RM codes in low-capacity scenarios.
2501.10705
Secure Communication in Dynamic RDARS-Driven Systems
cs.IT eess.SP math.IT
In this letter, we investigate a dynamic reconfigurable distributed antenna and reflection surface (RDARS)-driven secure communication system, where the working mode of the RDARS can be flexibly configured. We aim to maximize the secrecy rate by jointly designing the active beamforming vectors, reflection coefficients, and the channel-aware mode selection matrix. To address the non-convex binary and cardinality constraints introduced by dynamic mode selection, we propose an efficient alternating optimization (AO) framework that employs penalty-based fractional programming (FP) and successive convex approximation (SCA) transformations. Simulation results demonstrate the potential of RDARS in enhancing the secrecy rate and show its superiority compared to existing reflection surface-based schemes.
2501.10709
Revisiting Ensemble Methods for Stock Trading and Crypto Trading Tasks at ACM ICAIF FinRL Contest 2023-2024
cs.CE cs.AI stat.ML
Reinforcement learning has demonstrated great potential for performing financial tasks. However, it faces two major challenges: policy instability and sampling bottlenecks. In this paper, we revisit ensemble methods with massively parallel simulations on graphics processing units (GPUs), significantly enhancing the computational efficiency and robustness of trained models in volatile financial markets. Our approach leverages the parallel processing capability of GPUs to significantly improve the sampling speed for training ensemble models. The ensemble models combine the strengths of component agents to improve the robustness of financial decision-making strategies. We conduct experiments in both stock and cryptocurrency trading tasks to evaluate the effectiveness of our approach. Massively parallel simulation on a single GPU improves the sampling speed by up to $1,746\times$ using $2,048$ parallel environments compared to a single environment. The ensemble models have high cumulative returns and outperform some individual agents, reducing maximum drawdown by up to $4.17\%$ and improving the Sharpe ratio by up to $0.21$. This paper describes trading tasks at ACM ICAIF FinRL Contests in 2023 and 2024.
2501.10711
How Should We Build A Benchmark? Revisiting 274 Code-Related Benchmarks For LLMs
cs.SE cs.AI cs.CL
Various benchmarks have been proposed to assess the performance of large language models (LLMs) in different coding scenarios. We refer to them as code-related benchmarks. However, there are no systematic guidelines by which such a benchmark should be developed to ensure its quality, reliability, and reproducibility. We propose How2Bench, which is comprised of a 55-criteria checklist as a set of guidelines to govern the development of code-related benchmarks comprehensively. Using HOW2BENCH, we profiled 274 benchmarks released within the past decade and found concerning issues. Nearly 70% of the benchmarks did not take measures for data quality assurance; over 10% did not even open source or only partially open source. Many highly cited benchmarks have loopholes, including duplicated samples, incorrect reference codes/tests/prompts, and unremoved sensitive/confidential information. Finally, we conducted a human study involving 49 participants, which revealed significant gaps in awareness of the importance of data quality, reproducibility, and transparency.
2501.10712
Poisson Hail on a Wireless Ground
cs.IT cs.NI math.IT
This paper defines a new model which incorporates three key ingredients of a large class of wireless communication systems: (1) spatial interactions through interference, (2) dynamics of the queueing type, with users joining and leaving, and (3) carrier sensing and collision avoidance as used in, e.g., WiFi. In systems using (3), rather than directly accessing the shared resources upon arrival, a customer is considerate and waits to access them until nearby users in service have left. This new model can be seen as a missing piece of a larger puzzle that contains such dynamics as spatial birth-and-death processes, the Poisson-Hail model, and wireless dynamics as key other pieces. It is shown that, under natural assumptions, this model can be represented as a Markov process on the space of counting measures. The main results are then two-fold. The first is on the shape of the stability region and, more precisely, on the characterization of the critical value of the arrival rate that separates stability from instability. The second is of a more qualitative or perhaps even ethical nature. There is evidence that for natural values of the system parameters, the implementation of sensing and collision avoidance stabilizes a system that would be unstable if immediate access to the shared resources would be granted. In other words, for these parameters, renouncing greedy access makes sharing sustainable, whereas indulging in greedy access kills the system.
2501.10713
Human-like Nonverbal Behavior with MetaHumans in Real-World Interaction Studies: An Architecture Using Generative Methods and Motion Capture
cs.HC cs.RO
Socially interactive agents are gaining prominence in domains like healthcare, education, and service contexts, particularly virtual agents due to their inherent scalability. To facilitate authentic interactions, these systems require verbal and nonverbal communication through e.g., facial expressions and gestures. While natural language processing technologies have rapidly advanced, incorporating human-like nonverbal behavior into real-world interaction contexts is crucial for enhancing the success of communication, yet this area remains underexplored. One barrier is creating autonomous systems with sophisticated conversational abilities that integrate human-like nonverbal behavior. This paper presents a distributed architecture using Epic Games MetaHuman, combined with advanced conversational AI and camera-based user management, that supports methods like motion capture, handcrafted animation, and generative approaches for nonverbal behavior. We share insights into a system architecture designed to investigate nonverbal behavior in socially interactive agents, deployed in a three-week field study in the Deutsches Museum Bonn, showcasing its potential in realistic nonverbal behavior research.
2501.10714
FSMoE: A Flexible and Scalable Training System for Sparse Mixture-of-Experts Models
cs.LG
Recent large language models (LLMs) have tended to leverage sparsity to reduce computations, employing the sparsely activated mixture-of-experts (MoE) technique. MoE introduces four modules, including token routing, token communication, expert computation, and expert parallelism, that impact model quality and training efficiency. To enable versatile usage of MoE models, we introduce FSMoE, a flexible training system optimizing task scheduling with three novel techniques: 1) Unified abstraction and online profiling of MoE modules for task scheduling across various MoE implementations. 2) Co-scheduling intra-node and inter-node communications with computations to minimize communication overheads. 3) To support near-optimal task scheduling, we design an adaptive gradient partitioning method for gradient aggregation and a schedule to adaptively pipeline communications and computations. We conduct extensive experiments with configured MoE layers and real-world MoE models on two GPU clusters. Experimental results show that 1) our FSMoE supports four popular types of MoE routing functions and is more efficient than existing implementations (with up to a 1.42$\times$ speedup), and 2) FSMoE outperforms the state-of-the-art MoE training systems (DeepSpeed-MoE and Tutel) by 1.18$\times$-1.22$\times$ on 1458 MoE layers and 1.19$\times$-3.01$\times$ on real-world MoE models based on GPT-2 and Mixtral using a popular routing function.
2501.10722
A Unified Regularization Approach to High-Dimensional Generalized Tensor Bandits
cs.LG stat.ML
Modern decision-making scenarios often involve data that is both high-dimensional and rich in higher-order contextual information, where existing bandits algorithms fail to generate effective policies. In response, we propose in this paper a generalized linear tensor bandits algorithm designed to tackle these challenges by incorporating low-dimensional tensor structures, and further derive a unified analytical framework of the proposed algorithm. Specifically, our framework introduces a convex optimization approach with the weakly decomposable regularizers, enabling it to not only achieve better results based on the tensor low-rankness structure assumption but also extend to cases involving other low-dimensional structures such as slice sparsity and low-rankness. The theoretical analysis shows that, compared to existing low-rankness tensor result, our framework not only provides better bounds but also has a broader applicability. Notably, in the special case of degenerating to low-rank matrices, our bounds still offer advantages in certain scenarios.
2501.10727
In the Picture: Medical Imaging Datasets, Artifacts, and their Living Review
cs.CV cs.AI eess.IV
Datasets play a critical role in medical imaging research, yet issues such as label quality, shortcuts, and metadata are often overlooked. This lack of attention may harm the generalizability of algorithms and, consequently, negatively impact patient outcomes. While existing medical imaging literature reviews mostly focus on machine learning (ML) methods, with only a few focusing on datasets for specific applications, these reviews remain static -- they are published once and not updated thereafter. This fails to account for emerging evidence, such as biases, shortcuts, and additional annotations that other researchers may contribute after the dataset is published. We refer to these newly discovered findings of datasets as research artifacts. To address this gap, we propose a living review that continuously tracks public datasets and their associated research artifacts across multiple medical imaging applications. Our approach includes a framework for the living review to monitor data documentation artifacts, and an SQL database to visualize the citation relationships between research artifact and dataset. Lastly, we discuss key considerations for creating medical imaging datasets, review best practices for data annotation, discuss the significance of shortcuts and demographic diversity, and emphasize the importance of managing datasets throughout their entire lifecycle. Our demo is publicly available at http://130.226.140.142.
2501.10729
Robust Local Polynomial Regression with Similarity Kernels
stat.ME cs.LG stat.ML
Local Polynomial Regression (LPR) is a widely used nonparametric method for modeling complex relationships due to its flexibility and simplicity. It estimates a regression function by fitting low-degree polynomials to localized subsets of the data, weighted by proximity. However, traditional LPR is sensitive to outliers and high-leverage points, which can significantly affect estimation accuracy. This paper revisits the kernel function used to compute regression weights and proposes a novel framework that incorporates both predictor and response variables in the weighting mechanism. By introducing two positive definite kernels, the proposed method robustly estimates weights, mitigating the influence of outliers through localized density estimation. The method is implemented in Python and is publicly available at https://github.com/yaniv-shulman/rsklpr, demonstrating competitive performance in synthetic benchmark experiments. Compared to standard LPR, the proposed approach consistently improves robustness and accuracy, especially in heteroscedastic and noisy environments, without requiring multiple iterations. This advancement provides a promising extension to traditional LPR, opening new possibilities for robust regression applications.
2501.10731
Characterizing the Effects of Translation on Intertextuality using Multilingual Embedding Spaces
cs.CL
Rhetorical devices are difficult to translate, but they are crucial to the translation of literary documents. We investigate the use of multilingual embedding spaces to characterize the preservation of intertextuality, one common rhetorical device, across human and machine translation. To do so, we use Biblical texts, which are both full of intertextual references and are highly translated works. We provide a metric to characterize intertextuality at the corpus level and provide a quantitative analysis of the preservation of this rhetorical device across extant human translations and machine-generated counterparts. We go on to provide qualitative analysis of cases wherein human translations over- or underemphasize the intertextuality present in the text, whereas machine translations provide a neutral baseline. This provides support for established scholarship proposing that human translators have a propensity to amplify certain literary characteristics of the original manuscripts.
2501.10733
A CNN-Transformer for Classification of Longitudinal 3D MRI Images -- A Case Study on Hepatocellular Carcinoma Prediction
cs.CV
Longitudinal MRI analysis is crucial for predicting disease outcomes, particularly in chronic conditions like hepatocellular carcinoma (HCC), where early detection can significantly influence treatment strategies and patient prognosis. Yet, due to challenges like limited data availability, subtle parenchymal changes, and the irregular timing of medical screenings, current approaches have so far focused on cross-sectional imaging data. To address this, we propose HCCNet, a novel model architecture that integrates a 3D adaptation of the ConvNeXt CNN architecture with a Transformer encoder, capturing both the intricate spatial features of 3D MRIs and the complex temporal dependencies across different time points. HCCNet utilizes a two-stage pre-training process tailored for longitudinal MRI data. The CNN backbone is pre-trained using a self-supervised learning framework adapted for 3D MRIs, while the Transformer encoder is pre-trained with a sequence-order-prediction task to enhance its understanding of disease progression over time. We demonstrate the effectiveness of HCCNet by applying it to a cohort of liver cirrhosis patients undergoing regular MRI screenings for HCC surveillance. Our results show that HCCNet significantly improves predictive accuracy and reliability over baseline models, providing a robust tool for personalized HCC surveillance. The methodological approach presented in this paper is versatile and can be adapted to various longitudinal MRI screening applications. Its ability to handle varying patient record lengths and irregular screening intervals establishes it as an invaluable framework for monitoring chronic diseases, where timely and accurate disease prognosis is critical for effective treatment planning.
2501.10734
GEC-RAG: Improving Generative Error Correction via Retrieval-Augmented Generation for Automatic Speech Recognition Systems
eess.AS cs.AI cs.SD
Automatic Speech Recognition (ASR) systems have demonstrated remarkable performance across various applications. However, limited data and the unique language features of specific domains, such as low-resource languages, significantly degrade their performance and lead to higher Word Error Rates (WER). In this study, we propose Generative Error Correction via Retrieval-Augmented Generation (GEC-RAG), a novel approach designed to improve ASR accuracy for low-resource domains, like Persian. Our approach treats the ASR system as a black-box, a common practice in cloud-based services, and proposes a Retrieval-Augmented Generation (RAG) approach within the In-Context Learning (ICL) scheme to enhance the quality of ASR predictions. By constructing a knowledge base that pairs ASR predictions (1-best and 5-best hypotheses) with their corresponding ground truths, GEC-RAG retrieves lexically similar examples to the ASR transcription using the Term Frequency-Inverse Document Frequency (TF-IDF) measure. This process provides relevant error patterns of the system alongside the ASR transcription to the Generative Large Language Model (LLM), enabling targeted corrections. Our results demonstrate that this strategy significantly reduces WER in Persian and highlights a potential for domain adaptation and low-resource scenarios. This research underscores the effectiveness of using RAG in enhancing ASR systems without requiring direct model modification or fine-tuning, making it adaptable to any domain by simply updating the transcription knowledge base with domain-specific data.
2501.10736
Semi-supervised Semantic Segmentation for Remote Sensing Images via Multi-scale Uncertainty Consistency and Cross-Teacher-Student Attention
cs.CV cs.AI
Semi-supervised learning offers an appealing solution for remote sensing (RS) image segmentation to relieve the burden of labor-intensive pixel-level labeling. However, RS images pose unique challenges, including rich multi-scale features and high inter-class similarity. To address these problems, this paper proposes a novel semi-supervised Multi-Scale Uncertainty and Cross-Teacher-Student Attention (MUCA) model for RS image semantic segmentation tasks. Specifically, MUCA constrains the consistency among feature maps at different layers of the network by introducing a multi-scale uncertainty consistency regularization. It improves the multi-scale learning capability of semi-supervised algorithms on unlabeled data. Additionally, MUCA utilizes a Cross-Teacher-Student attention mechanism to guide the student network, guiding the student network to construct more discriminative feature representations through complementary features from the teacher network. This design effectively integrates weak and strong augmentations (WA and SA) to further boost segmentation performance. To verify the effectiveness of our model, we conduct extensive experiments on ISPRS-Potsdam and LoveDA datasets. The experimental results show the superiority of our method over state-of-the-art semi-supervised methods. Notably, our model excels in distinguishing highly similar objects, showcasing its potential for advancing semi-supervised RS image segmentation tasks.
2501.10739
Computational Discovery of Chiasmus in Ancient Religious Text
cs.CL
Chiasmus, a debated literary device in Biblical texts, has captivated mystics while sparking ongoing scholarly discussion. In this paper, we introduce the first computational approach to systematically detect chiasmus within Biblical passages. Our method leverages neural embeddings to capture lexical and semantic patterns associated with chiasmus, applied at multiple levels of textual granularity (half-verses, verses). We also involve expert annotators to review a subset of the detected patterns. Despite its computational efficiency, our method achieves robust results, with high inter-annotator agreement and system precision@k of 0.80 at the verse level and 0.60 at the half-verse level. We further provide a qualitative analysis of the distribution of detected chiasmi, along with selected examples that highlight the effectiveness of our approach.
2501.10741
Development of Application-Specific Large Language Models to Facilitate Research Ethics Review
cs.CL cs.CY
Institutional review boards (IRBs) play a crucial role in ensuring the ethical conduct of human subjects research, but face challenges including inconsistency, delays, and inefficiencies. We propose the development and implementation of application-specific large language models (LLMs) to facilitate IRB review processes. These IRB-specific LLMs would be fine-tuned on IRB-specific literature and institutional datasets, and equipped with retrieval capabilities to access up-to-date, context-relevant information. We outline potential applications, including pre-review screening, preliminary analysis, consistency checking, and decision support. While addressing concerns about accuracy, context sensitivity, and human oversight, we acknowledge remaining challenges such as over-reliance on AI and the need for transparency. By enhancing the efficiency and quality of ethical review while maintaining human judgment in critical decisions, IRB-specific LLMs offer a promising tool to improve research oversight. We call for pilot studies to evaluate the feasibility and impact of this approach.
2501.10743
Analysis of Age-Energy Trade-off in IoT Networks Using Stochastic Geometry
cs.IT math.IT
We study an internet of things (IoT) network where devices harvest energy from transmitter power. IoT devices use this harvested energy to operate and decode data packets. We propose a slot division scheme based on a parameter $\xi$, where the first phase is for energy harvesting (EH) and the second phase is for data transmission. We define the joint success probability (JSP) metric as the probability of the event that both the harvested energy and the received signal-to-interference ratio (SIR) exceed their respective thresholds. We provide lower and upper bounds of (JSP), as obtaining an exact JSP expression is challenging. Then, the peak age-of-information (PAoI) of data packets is determined using this framework. Higher slot intervals for EH reduce data transmission time, requiring higher link rates. In contrast, a lower EH slot interval will leave IoT devices without enough energy to decode the packets. We demonstrate that both non-preemptive and preemptive queuing disciplines may have the same optimal slot partitioning factor for maximizing the JSP and minimizing the PAoI. For different transmit powers and deployment areas, we recommend the optimal slot partitioning factor for the above two metrics under both queuing disciplines.
2501.10750
PEARL: Preconditioner Enhancement through Actor-critic Reinforcement Learning
cs.LG cs.NA math.NA stat.ML
We present PEARL (Preconditioner Enhancement through Actor-critic Reinforcement Learning), a novel approach to learning matrix preconditioners. Existing preconditioners such as Jacobi, Incomplete LU, and Algebraic Multigrid methods offer problem-specific advantages but rely heavily on hyperparameter tuning. Recent advances have explored using deep neural networks to learn preconditioners, though challenges such as misbehaved objective functions and costly training procedures remain. PEARL introduces a reinforcement learning approach for learning preconditioners, specifically, a contextual bandit formulation. The framework utilizes an actor-critic model, where the actor generates the incomplete Cholesky decomposition of preconditioners, and the critic evaluates them based on reward-specific feedback. To further guide the training, we design a dual-objective function, combining updates from the critic and condition number. PEARL contributes a generalizable preconditioner learning method, dynamic sparsity exploration, and cosine schedulers for improved stability and exploratory power. We compare our approach to traditional and neural preconditioners, demonstrating improved flexibility and iterative solving speed.
2501.10752
Quadcopter Position Hold Function using Optical Flow in a Smartphone-based Flight Computer
cs.CV
Purpose. This paper explores the capability of smartphones as computing devices for a quadcopter, specifically in terms of the ability of drones to maintain a position known as the position hold function. Image processing can be performed with the phone's sensors and powerful built-in camera. Method. Using Shi-Tomasi corner detection and the Lucas-Kanade sparse optical flow algorithms, ground features are recognized and tracked using the downward-facing camera. The position is maintained by computing quadcopter displacement from the center of the image using Euclidian distance, and the corresponding pitch and roll estimate is calculated using the PID controller. Results. Actual flights show a double standard deviation of 18.66 cm from the center for outdoor tests. With a quadcopter size of 58cm x 58cm used, it implies that 95% of the time, the quadcopter is within a diameter of 96 cm. For indoor tests, a double standard deviation of 10.55 cm means that 95% of the time, the quadcopter is within a diameter of 79 cm. Conclusion. Smartphone sensors and cameras can be used to perform optical flow position hold functions, proving their potential as computing devices for drones. Recommendations. To further improve the positioning system of the phone-based quadcopter system, it is suggested that potential sensor fusion be explored with the phone's GNSS sensor, which gives absolute positioning information for outdoor applications. Research Implications. As different devices and gadgets are integrated into the smartphone, this paper presents an opportunity for phone manufacturers and researchers to explore the potential of smartphones for a drone use-case.
2501.10753
Pinching Antennas: Principles, Applications and Challenges
cs.IT eess.SP math.IT
Flexible-antenna systems, such as fluid antennas and movable antennas, have been recognized as key enabling technologies for sixth-generation (6G) wireless networks, as they can intelligently reconfigure the effective channel gains of the users and hence significantly improve their data transmission capabilities. However, existing flexible-antenna systems have been designed to combat small-scale fading in non-line-of-sight (NLoS) conditions. As a result, they lack the ability to establish line-of-sight links, which are typically 100 times stronger than NLoS links. In addition, existing flexible-antenna systems have limited flexibility, where adding/removing an antenna is not straightforward. This article introduces an innovative flexible-antenna system called pinching antennas, which are realized by applying small dielectric particles to waveguides. We first describe the basics of pinching-antenna systems and their ability to provide strong LoS links by deploying pinching antennas close to the users as well as their capability to scale up/down the antenna system. We then focus on communication scenarios with different numbers of waveguides and pinching antennas, where innovative approaches to implement multiple-input multiple-output and non-orthogonal multiple access are discussed. In addition, promising 6G-related applications of pinching antennas, including integrated sensing and communication and next-generation multiple access, are presented. Finally, important directions for future research, such as waveguide deployment and channel estimation, are highlighted.
2501.10755
An Experimental Study on Joint Modeling for Sound Event Localization and Detection with Source Distance Estimation
cs.SD cs.LG eess.AS
In traditional sound event localization and detection (SELD) tasks, the focus is typically on sound event detection (SED) and direction-of-arrival (DOA) estimation, but they fall short of providing full spatial information about the sound source. The 3D SELD task addresses this limitation by integrating source distance estimation (SDE), allowing for complete spatial localization. We propose three approaches to tackle this challenge: a novel method with independent training and joint prediction, which firstly treats DOA and distance estimation as separate tasks and then combines them to solve 3D SELD; a dual-branch representation with source Cartesian coordinate used for simultaneous DOA and distance estimation; and a three-branch structure that jointly models SED, DOA, and SDE within a unified framework. Our proposed method ranked first in the DCASE 2024 Challenge Task 3, demonstrating the effectiveness of joint modeling for addressing the 3D SELD task. The relevant code for this paper will be open-sourced in the future.
2501.10756
D2D Coded Caching Schemes for Multiaccess Networks with Combinatorial Access Topology
cs.IT math.IT
This paper considers wireless device-to-device (D2D) coded caching in a multiaccess network, where the users communicate with each other and each user can access multiple cache nodes. Access topologies derived from two combinatorial designs known as the $t$-design and $t$-group divisible design ($t$-GDD), referred to as the $t$-design and $t$-GDD topologies respectively, which subsume a few other known topologies, have been studied for the multiaccess coded caching (MACC) network by Cheng \textit{et al.} in \cite{MACC_des}. These access topologies are extended to a multiaccess D2D coded caching (MADCC) network and novel MADCC schemes are proposed. MADCC network has been studied so far only for the cyclic wrap-around topology. Apart from the proposed novel MADCC schemes, MADCC schemes are also derived from the existing MACC schemes in \cite{MACC_des}. To compare the performance of different MADCC schemes, the metrics of load per user and subpacketization level are used while keeping the number of caches and cache memory size same. The proposed MADCC scheme with $t$-design topology performs better in terms of subpacketization level while achieving the same load per user compared to the MADCC scheme derived from the MACC scheme with $t$-design topology in \cite{MACC_des}. The proposed MADCC scheme with $t$-GDD topology performs better in terms of load per user while achieving the same subpacketization level compared to the MADCC scheme derived from the MACC scheme with $t$-GDD topology in \cite{MACC_des} in some cases. Compared to the existing MADCC scheme with cyclic wrap-around topology, the proposed MADCC scheme with $t$-design topology performs better in terms of load per user, and the proposed MADCC scheme with $t$-GDD topology performs better in terms of subpacketization level at the expense of an increase in load per user.
2501.10757
Deformable Image Registration of Dark-Field Chest Radiographs for Local Lung Signal Change Assessment
eess.IV cs.CV physics.med-ph
Dark-field radiography of the human chest has been demonstrated to have promising potential for the analysis of the lung microstructure and the diagnosis of respiratory diseases. However, previous studies of dark-field chest radiographs evaluated the lung signal only in the inspiratory breathing state. Our work aims to add a new perspective to these previous assessments by locally comparing dark-field lung information between different respiratory states. To this end, we discuss suitable image registration methods for dark-field chest radiographs to enable consistent spatial alignment of the lung in distinct breathing states. Utilizing full inspiration and expiration scans from a clinical chronic obstructive pulmonary disease study, we assess the performance of the proposed registration framework and outline applicable evaluation approaches. Our regional characterization of lung dark-field signal changes between the breathing states provides a proof-of-principle that dynamic radiography-based lung function assessment approaches may benefit from considering registered dark-field images in addition to standard plain chest radiographs.
2501.10761
Infrared and Visible Image Fusion: From Data Compatibility to Task Adaption
cs.CV
Infrared-visible image fusion (IVIF) is a critical task in computer vision, aimed at integrating the unique features of both infrared and visible spectra into a unified representation. Since 2018, the field has entered the deep learning era, with an increasing variety of approaches introducing a range of networks and loss functions to enhance visual performance. However, challenges such as data compatibility, perception accuracy, and efficiency remain. Unfortunately, there is a lack of recent comprehensive surveys that address this rapidly expanding domain. This paper fills that gap by providing a thorough survey covering a broad range of topics. We introduce a multi-dimensional framework to elucidate common learning-based IVIF methods, from visual enhancement strategies to data compatibility and task adaptability. We also present a detailed analysis of these approaches, accompanied by a lookup table clarifying their core ideas. Furthermore, we summarize performance comparisons, both quantitatively and qualitatively, focusing on registration, fusion, and subsequent high-level tasks. Beyond technical analysis, we discuss potential future directions and open issues in this area. For further details, visit our GitHub repository: https://github.com/RollingPlain/IVIF_ZOO.
2501.10768
MAPS: Advancing Multi-Modal Reasoning in Expert-Level Physical Science
cs.AI
Pre-trained on extensive text and image corpora, current Multi-Modal Large Language Models (MLLM) have shown strong capabilities in general visual reasoning tasks. However, their performance is still lacking in physical domains that require understanding diagrams with complex physical structures and quantitative analysis based on multi-modal information. To address this, we develop a new framework, named Multi-Modal Scientific Reasoning with Physics Perception and Simulation (MAPS) based on an MLLM. MAPS decomposes expert-level multi-modal reasoning task into physical diagram understanding via a Physical Perception Model (PPM) and reasoning with physical knowledge via a simulator. The PPM module is obtained by fine-tuning a visual language model using carefully designed synthetic data with paired physical diagrams and corresponding simulation language descriptions. At the inference stage, MAPS integrates the simulation language description of the input diagram provided by PPM and results obtained through a Chain-of-Simulation process with MLLM to derive the underlying rationale and the final answer. Validated using our collected college-level circuit analysis problems, MAPS significantly improves reasoning accuracy of MLLM and outperforms all existing models. The results confirm MAPS offers a promising direction for enhancing multi-modal scientific reasoning ability of MLLMs. We will release our code, model and dataset used for our experiments upon publishing of this paper.
2501.10770
Enhancing Diagnostic in 3D COVID-19 Pneumonia CT-scans through Explainable Uncertainty Bayesian Quantification
eess.IV cs.AI cs.CV cs.LG
Accurately classifying COVID-19 pneumonia in 3D CT scans remains a significant challenge in the field of medical image analysis. Although deterministic neural networks have shown promising results in this area, they provide only point estimates outputs yielding poor diagnostic in clinical decision-making. In this paper, we explore the use of Bayesian neural networks for classifying COVID-19 pneumonia in 3D CT scans providing uncertainties in their predictions. We compare deterministic networks and their Bayesian counterpart, enhancing the decision-making accuracy under uncertainty information. Remarkably, our findings reveal that lightweight architectures achieve the highest accuracy of 96\% after developing extensive hyperparameter tuning. Furthermore, the Bayesian counterpart of these architectures via Multiplied Normalizing Flow technique kept a similar performance along with calibrated uncertainty estimates. Finally, we have developed a 3D-visualization approach to explain the neural network outcomes based on SHAP values. We conclude that explainability along with uncertainty quantification will offer better clinical decisions in medical image analysis, contributing to ongoing efforts for improving the diagnosis and treatment of COVID-19 pneumonia.
2501.10774
Model Monitoring in the Absence of Labeled Data via Feature Attributions Distributions
cs.LG
Model monitoring involves analyzing AI algorithms once they have been deployed and detecting changes in their behaviour. This thesis explores machine learning model monitoring ML before the predictions impact real-world decisions or users. This step is characterized by one particular condition: the absence of labelled data at test time, which makes it challenging, even often impossible, to calculate performance metrics. The thesis is structured around two main themes: (i) AI alignment, measuring if AI models behave in a manner consistent with human values and (ii) performance monitoring, measuring if the models achieve specific accuracy goals or desires. The thesis uses a common methodology that unifies all its sections. It explores feature attribution distributions for both monitoring dimensions. Using these feature attribution explanations, we can exploit their theoretical properties to derive and establish certain guarantees and insights into model monitoring.
2501.10775
MedFILIP: Medical Fine-grained Language-Image Pre-training
cs.CV cs.AI
Medical vision-language pretraining (VLP) that leverages naturally-paired medical image-report data is crucial for medical image analysis. However, existing methods struggle to accurately characterize associations between images and diseases, leading to inaccurate or incomplete diagnostic results. In this work, we propose MedFILIP, a fine-grained VLP model, introduces medical image-specific knowledge through contrastive learning, specifically: 1) An information extractor based on a large language model is proposed to decouple comprehensive disease details from reports, which excels in extracting disease deals through flexible prompt engineering, thereby effectively reducing text complexity while retaining rich information at a tiny cost. 2) A knowledge injector is proposed to construct relationships between categories and visual attributes, which help the model to make judgments based on image features, and fosters knowledge extrapolation to unfamiliar disease categories. 3) A semantic similarity matrix based on fine-grained annotations is proposed, providing smoother, information-richer labels, thus allowing fine-grained image-text alignment. 4) We validate MedFILIP on numerous datasets, e.g., RSNA-Pneumonia, NIH ChestX-ray14, VinBigData, and COVID-19. For single-label, multi-label, and fine-grained classification, our model achieves state-of-the-art performance, the classification accuracy has increased by a maximum of 6.69\%. The code is available in https://github.com/PerceptionComputingLab/MedFILIP.
2501.10777
The working principles of model-based GAs fall within the PAC framework: A mathematical theory of problem decomposition
cs.NE
The concepts of linkage, building blocks, and problem decomposition have long existed in the genetic algorithm (GA) field and have guided the development of model-based GAs for decades. However, their definitions are usually vague, making it difficult to develop theoretical support. This paper provides an algorithm-independent definition to describe the concept of linkage. With this definition, the paper proves that any problems with a bounded degree of linkage are decomposable and that proper problem decomposition is possible via linkage learning. The way of decomposition given in this paper also offers a new perspective on nearly decomposable problems with bounded difficulty and building blocks from the theoretical aspect. Finally, this paper relates problem decomposition to PAC learning and proves that the global optima of these problems and the minimum decomposition blocks are PAC learnable under certain conditions.
2501.10781
Simultaneous Computation with Multiple Prioritizations in Multi-Agent Motion Planning
cs.MA cs.AI cs.RO
Multi-agent path finding (MAPF) in large networks is computationally challenging. An approach for MAPF is prioritized planning (PP), in which agents plan sequentially according to their priority. Albeit a computationally efficient approach for MAPF, the solution quality strongly depends on the prioritization. Most prioritizations rely either on heuristics, which do not generalize well, or iterate to find adequate priorities, which costs computational effort. In this work, we show how agents can compute with multiple prioritizations simultaneously. Our approach is general as it does not rely on domain-specific knowledge. The context of this work is multi-agent motion planning (MAMP) with a receding horizon subject to computation time constraints. MAMP considers the system dynamics in more detail compared to MAPF. In numerical experiments on MAMP, we demonstrate that our approach to prioritization comes close to optimal prioritization and outperforms state-of-the-art methods with only a minor increase in computation time. We show real-time capability in an experiment on a road network with ten vehicles in our Cyber-Physical Mobility Lab.
2501.10782
ML-SceGen: A Multi-level Scenario Generation Framework
cs.AI
Current scientific research witnesses various attempts at applying Large Language Models for scenario generation but is inclined only to comprehensive or dangerous scenarios. In this paper, we seek to build a three-stage framework that not only lets users regain controllability over the generated scenarios but also generates comprehensive scenarios containing danger factors in uncontrolled intersection settings. In the first stage, LLM agents will contribute to translating the key components of the description of the expected scenarios into Functional Scenarios. For the second stage, we use Answer Set Programming (ASP) solver Clingo to help us generate comprehensive logical traffic within intersections. During the last stage, we use LLM to update relevant parameters to increase the critical level of the concrete scenario.
2501.10784
Measuring Fairness in Financial Transaction Machine Learning Models
cs.LG
Mastercard, a global leader in financial services, develops and deploys machine learning models aimed at optimizing card usage and preventing attrition through advanced predictive models. These models use aggregated and anonymized card usage patterns, including cross-border transactions and industry-specific spending, to tailor bank offerings and maximize revenue opportunities. Mastercard has established an AI Governance program, based on its Data and Tech Responsibility Principles, to evaluate any built and bought AI for efficacy, fairness, and transparency. As part of this effort, Mastercard has sought expertise from the Turing Institute through a Data Study Group to better assess fairness in more complex AI/ML models. The Data Study Group challenge lies in defining, measuring, and mitigating fairness in these predictions, which can be complex due to the various interpretations of fairness, gaps in the research literature, and ML-operations challenges.
2501.10787
LD-DETR: Loop Decoder DEtection TRansformer for Video Moment Retrieval and Highlight Detection
cs.CV cs.IR cs.LG
Video Moment Retrieval and Highlight Detection aim to find corresponding content in the video based on a text query. Existing models usually first use contrastive learning methods to align video and text features, then fuse and extract multimodal information, and finally use a Transformer Decoder to decode multimodal information. However, existing methods face several issues: (1) Overlapping semantic information between different samples in the dataset hinders the model's multimodal aligning performance; (2) Existing models are not able to efficiently extract local features of the video; (3) The Transformer Decoder used by the existing model cannot adequately decode multimodal features. To address the above issues, we proposed the LD-DETR model for Video Moment Retrieval and Highlight Detection tasks. Specifically, we first distilled the similarity matrix into the identity matrix to mitigate the impact of overlapping semantic information. Then, we designed a method that enables convolutional layers to extract multimodal local features more efficiently. Finally, we fed the output of the Transformer Decoder back into itself to adequately decode multimodal information. We evaluated LD-DETR on four public benchmarks and conducted extensive experiments to demonstrate the superiority and effectiveness of our approach. Our model outperforms the State-Of-The-Art models on QVHighlight, Charades-STA and TACoS datasets. Our code is available at https://github.com/qingchen239/ld-detr.
2501.10788
Decoupling Appearance Variations with 3D Consistent Features in Gaussian Splatting
cs.CV
Gaussian Splatting has emerged as a prominent 3D representation in novel view synthesis, but it still suffers from appearance variations, which are caused by various factors, such as modern camera ISPs, different time of day, weather conditions, and local light changes. These variations can lead to floaters and color distortions in the rendered images/videos. Recent appearance modeling approaches in Gaussian Splatting are either tightly coupled with the rendering process, hindering real-time rendering, or they only account for mild global variations, performing poorly in scenes with local light changes. In this paper, we propose DAVIGS, a method that decouples appearance variations in a plug-and-play and efficient manner. By transforming the rendering results at the image level instead of the Gaussian level, our approach can model appearance variations with minimal optimization time and memory overhead. Furthermore, our method gathers appearance-related information in 3D space to transform the rendered images, thus building 3D consistency across views implicitly. We validate our method on several appearance-variant scenes, and demonstrate that it achieves state-of-the-art rendering quality with minimal training time and memory usage, without compromising rendering speeds. Additionally, it provides performance improvements for different Gaussian Splatting baselines in a plug-and-play manner.
2501.10789
CS-Net:Contribution-based Sampling Network for Point Cloud Simplification
cs.CV
Point cloud sampling plays a crucial role in reducing computation costs and storage requirements for various vision tasks. Traditional sampling methods, such as farthest point sampling, lack task-specific information and, as a result, cannot guarantee optimal performance in specific applications. Learning-based methods train a network to sample the point cloud for the targeted downstream task. However, they do not guarantee that the sampled points are the most relevant ones. Moreover, they may result in duplicate sampled points, which requires completion of the sampled point cloud through post-processing techniques. To address these limitations, we propose a contribution-based sampling network (CS-Net), where the sampling operation is formulated as a Top-k operation. To ensure that the network can be trained in an end-to-end way using gradient descent algorithms, we use a differentiable approximation to the Top-k operation via entropy regularization of an optimal transport problem. Our network consists of a feature embedding module, a cascade attention module, and a contribution scoring module. The feature embedding module includes a specifically designed spatial pooling layer to reduce parameters while preserving important features. The cascade attention module combines the outputs of three skip connected offset attention layers to emphasize the attractive features and suppress less important ones. The contribution scoring module generates a contribution score for each point and guides the sampling process to prioritize the most important ones. Experiments on the ModelNet40 and PU147 showed that CS-Net achieved state-of-the-art performance in two semantic-based downstream tasks (classification and registration) and two reconstruction-based tasks (compression and surface reconstruction).
2501.10791
A Novel Precoder for Peak-to-Average Power Ratio Reduction in OTFS Systems
cs.IT eess.SP math.IT
We consider the issue of high peak-to-average-power ratio (PAPR) of Orthogonal time frequency space (OTFS) modulated signals. This paper proposes a low-complexity novel iterative PAPR reduction method which achieves a PAPR reduction of roughly 5 dB when compared to a OTFS modulated signal without any PAPR compensation. Simulations reveal that the PAPR achieved by the proposed method is significantly better than that achieved by other state-of-art methods. Simulations also reveal that the error rate performance of OTFS based systems with the proposed PAPR reduction is similar to that achieved with the other state-of-art methods.
2501.10796
Dynamic Trend Fusion Module for Traffic Flow Prediction
cs.LG
Accurate traffic flow prediction is essential for applications like transport logistics but remains challenging due to complex spatio-temporal correlations and non-linear traffic patterns. Existing methods often model spatial and temporal dependencies separately, failing to effectively fuse them. To overcome this limitation, the Dynamic Spatial-Temporal Trend Transformer DST2former is proposed to capture spatio-temporal correlations through adaptive embedding and to fuse dynamic and static information for learning multi-view dynamic features of traffic networks. The approach employs the Dynamic Trend Representation Transformer (DTRformer) to generate dynamic trends using encoders for both temporal and spatial dimensions, fused via Cross Spatial-Temporal Attention. Predefined graphs are compressed into a representation graph to extract static attributes and reduce redundancy. Experiments on four real-world traffic datasets demonstrate that our framework achieves state-of-the-art performance.
2501.10799
Step-KTO: Optimizing Mathematical Reasoning through Stepwise Binary Feedback
cs.LG cs.AI
Large language models (LLMs) have recently demonstrated remarkable success in mathematical reasoning. Despite progress in methods like chain-of-thought prompting and self-consistency sampling, these advances often focus on final correctness without ensuring that the underlying reasoning process is coherent and reliable. This paper introduces Step-KTO, a training framework that combines process-level and outcome-level binary feedback to guide LLMs toward more trustworthy reasoning trajectories. By providing binary evaluations for both the intermediate reasoning steps and the final answer, Step-KTO encourages the model to adhere to logical progressions rather than relying on superficial shortcuts. Our experiments on challenging mathematical benchmarks show that Step-KTO significantly improves both final answer accuracy and the quality of intermediate reasoning steps. For example, on the MATH-500 dataset, Step-KTO achieves a notable improvement in Pass@1 accuracy over strong baselines. These results highlight the promise of integrating stepwise process feedback into LLM training, paving the way toward more interpretable and dependable reasoning capabilities.
2501.10800
Jailbreaking Large Language Models in Infinitely Many Ways
cs.LG cs.CR
We discuss the "Infinitely Many Meanings" attacks (IMM), a category of jailbreaks that leverages the increasing capabilities of a model to handle paraphrases and encoded communications to bypass their defensive mechanisms. IMMs' viability pairs and grows with a model's capabilities to handle and bind the semantics of simple mappings between tokens and work extremely well in practice, posing a concrete threat to the users of the most powerful LLMs in commerce. We show how one can bypass the safeguards of the most powerful open- and closed-source LLMs and generate content that explicitly violates their safety policies. One can protect against IMMs by improving the guardrails and making them scale with the LLMs' capabilities. For two categories of attacks that are straightforward to implement, i.e., bijection and encoding, we discuss two defensive strategies, one in token and the other in embedding space. We conclude with some research questions we believe should be prioritised to enhance the defensive mechanisms of LLMs and our understanding of their safety.
2501.10806
Non-Expansive Mappings in Two-Time-Scale Stochastic Approximation: Finite-Time Analysis
math.OC cs.LG cs.SY eess.SY stat.ML
Two-time-scale stochastic approximation is an iterative algorithm used in applications such as optimization, reinforcement learning, and control. Finite-time analysis of these algorithms has primarily focused on fixed point iterations where both time-scales have contractive mappings. In this paper, we study two-time-scale iterations, where the slower time-scale has a non-expansive mapping. For such algorithms, the slower time-scale can be considered a stochastic inexact Krasnoselskii-Mann iteration. We show that the mean square error decays at a rate $O(1/k^{1/4-\epsilon})$, where $\epsilon>0$ is arbitrarily small. We also show almost sure convergence of iterates to the set of fixed points. We show the applicability of our framework by applying our results to minimax optimization, linear stochastic approximation, and Lagrangian optimization.
2501.10808
Optimizing MACD Trading Strategies A Dance of Finance, Wavelets, and Genetics
cs.CE
In today's financial markets, quantitative trading has become an essential trading method, with the MACD indicator widely employed in quantitative trading strategies. This paper begins by screening and cleaning the dataset, establishing a model that adheres to the basic buy and sell rules of the MACD, and calculating key metrics such as the win rate, return, Sharpe ratio, and maximum drawdown for each stock. However, the MACD often generates erroneous signals in highly volatile markets. To address this, wavelet transform is applied to reduce noise, smoothing the DIF image, and a model is developed based on this to optimize the identification of buy and sell points. The results show that the annualized return has increased by 5%, verifying the feasibility of the method. Subsequently, the divergence principle is used to further optimize the trading strategy, enhancing the model's performance. Additionally, a genetic algorithm is employed to optimize the MACD parameters, tailoring the strategy to the characteristics of different stocks. To improve computational efficiency, the MindSpore framework is used for resource management and parallel computing. The optimized strategy demonstrates improved win rates, returns, Sharpe ratios, and a reduction in maximum drawdown in backtesting.
2501.10809
Efficient Auto-Labeling of Large-Scale Poultry Datasets (ALPD) Using Semi-Supervised Models, Active Learning, and Prompt-then-Detect Approach
cs.CV cs.AI
The rapid growth of AI in poultry farming has highlighted the challenge of efficiently labeling large, diverse datasets. Manual annotation is time-consuming, making it impractical for modern systems that continuously generate data. This study explores semi-supervised auto-labeling methods, integrating active learning, and prompt-then-detect paradigm to develop an efficient framework for auto-labeling of large poultry datasets aimed at advancing AI-driven behavior and health monitoring. Viideo data were collected from broilers and laying hens housed at the University of Arkansas and the University of Georgia. The collected videos were converted into images, pre-processed, augmented, and labeled. Various machine learning models, including zero-shot models like Grounding DINO, YOLO-World, and CLIP, and supervised models like YOLO and Faster-RCNN, were utilized for broilers, hens, and behavior detection. The results showed that YOLOv8s-World and YOLOv9s performed better when compared performance metrics for broiler and hen detection under supervised learning, while among the semi-supervised model, YOLOv8s-ALPD achieved the highest precision (96.1%) and recall (99.0%) with an RMSE of 1.9. The hybrid YOLO-World model, incorporating the optimal YOLOv8s backbone, demonstrated the highest overall performance. It achieved a precision of 99.2%, recall of 99.4%, and an F1 score of 98.7% for breed detection, alongside a precision of 88.4%, recall of 83.1%, and an F1 score of 84.5% for individual behavior detection. Additionally, semi-supervised models showed significant improvements in behavior detection, achieving up to 31% improvement in precision and 16% in F1-score. The semi-supervised models with minimal active learning reduced annotation time by over 80% compared to full manual labeling. Moreover, integrating zero-shot models enhanced detection and behavior identification.
2501.10810
Convergence and Running Time of Time-dependent Ant Colony Algorithms
cs.DS cs.NE
Ant Colony Optimization (ACO) is a well-known method inspired by the foraging behavior of ants and is extensively used to solve combinatorial optimization problems. In this paper, we first consider a general framework based on the concept of a construction graph - a graph associated with an instance of the optimization problem under study, where feasible solutions are represented by walks. We analyze the running time of this ACO variant, known as the Graph-based Ant System with time-dependent evaporation rate (GBAS/tdev), and prove that the algorithm's solution converges to the optimal solution of the problem with probability 1 for a slightly stronger evaporation rate function than was previously known. We then consider two time-dependent adaptations of Attiratanasunthron and Fakcharoenphol's $n$-ANT algorithm: $n$-ANT with time-dependent evaporation rate ($n$-ANT/tdev) and $n$-ANT with time-dependent lower pheromone bound ($n$-ANT/tdlb). We analyze both variants on the single destination shortest path problem (SDSP). Our results show that $n$-ANT/tdev has a super-polynomial time lower bound on the SDSP. In contrast, we show that $n$-ANT/tdlb achieves a polynomial time upper bound on this problem.
2501.10812
Graph Coloring to Reduce Computation Time in Prioritized Planning
cs.MA cs.AI cs.RO
Distributing computations among agents in large networks reduces computational effort in multi-agent path finding (MAPF). One distribution strategy is prioritized planning (PP). In PP, we couple and prioritize interacting agents to achieve a desired behavior across all agents in the network. We characterize the interaction with a directed acyclic graph (DAG). The computation time for solving MAPF problem using PP is mainly determined through the longest path in this DAG. The longest path depends on the fixed undirected coupling graph and the variable prioritization. The approaches from literature to prioritize agents are numerous and pursue various goals. This article presents an approach for prioritization in PP to reduce the longest path length in the coupling DAG and thus the computation time for MAPF using PP. We prove that this problem can be mapped to a graph-coloring problem, in which the number of colors required corresponds to the longest path length in the coupling DAG. We propose a decentralized graph-coloring algorithm to determine priorities for the agents. We evaluate the approach by applying it to multi-agent motion planning (MAMP) for connected and automated vehicles (CAVs) on roads using, a variant of MAPF.
2501.10814
No More Sliding Window: Efficient 3D Medical Image Segmentation with Differentiable Top-k Patch Sampling
eess.IV cs.AI cs.CV cs.LG
3D models are favored over 2D for 3D medical image segmentation tasks due to their ability to leverage inter-slice relationship, yielding higher segmentation accuracy. However, 3D models demand significantly more GPU memory with increased model size and intermediate tensors. A common solution is to use patch-based training and make whole-volume predictions with sliding window (SW) inference. SW inference reduces memory usage but is slower due to equal resource allocation across patches and less accurate as it overlooks global features beyond patches. We propose NMSW-Net (No-More-Sliding-Window-Net), a novel framework that enhances efficiency and accuracy of any given 3D segmentation model by eliminating SW inference and incorporating global predictions when necessary. NMSW-Net incorporates a differentiable Top-k module to sample only the relevant patches that enhance segmentation accuracy, thereby minimizing redundant computations. Additionally, it learns to leverage coarse global predictions when patch prediction alone is insufficient. NMSW-Net is model-agnostic, making it compatible with any 3D segmentation model that previously relied on SW inference. Evaluated across 3 tasks with 3 segmentation backbones, NMSW-Net achieves competitive or sometimes superior accuracy compared to SW, while reducing computational complexity by 90% (87.5 to 7.95 TFLOPS), delivering 4x faster inference on the H100 GPU (19.0 to 4.3 sec), and 7x faster inference on the Intel Xeon Gold CPU (1710 to 230 seconds).
2501.10815
An Interpretable Measure for Quantifying Predictive Dependence between Continuous Random Variables -- Extended Version
cs.LG math.ST stat.ML stat.TH
A fundamental task in statistical learning is quantifying the joint dependence or association between two continuous random variables. We introduce a novel, fully non-parametric measure that assesses the degree of association between continuous variables $X$ and $Y$, capable of capturing a wide range of relationships, including non-functional ones. A key advantage of this measure is its interpretability: it quantifies the expected relative loss in predictive accuracy when the distribution of $X$ is ignored in predicting $Y$. This measure is bounded within the interval [0,1] and is equal to zero if and only if $X$ and $Y$ are independent. We evaluate the performance of our measure on over 90,000 real and synthetic datasets, benchmarking it against leading alternatives. Our results demonstrate that the proposed measure provides valuable insights into underlying relationships, particularly in cases where existing methods fail to capture important dependencies.
2501.10819
GAUDA: Generative Adaptive Uncertainty-guided Diffusion-based Augmentation for Surgical Segmentation
cs.CV cs.LG
Augmentation by generative modelling yields a promising alternative to the accumulation of surgical data, where ethical, organisational and regulatory aspects must be considered. Yet, the joint synthesis of (image, mask) pairs for segmentation, a major application in surgery, is rather unexplored. We propose to learn semantically comprehensive yet compact latent representations of the (image, mask) space, which we jointly model with a Latent Diffusion Model. We show that our approach can effectively synthesise unseen high-quality paired segmentation data of remarkable semantic coherence. Generative augmentation is typically applied pre-training by synthesising a fixed number of additional training samples to improve downstream task models. To enhance this approach, we further propose Generative Adaptive Uncertainty-guided Diffusion-based Augmentation (GAUDA), leveraging the epistemic uncertainty of a Bayesian downstream model for targeted online synthesis. We condition the generative model on classes with high estimated uncertainty during training to produce additional unseen samples for these classes. By adaptively utilising the generative model online, we can minimise the number of additional training samples and centre them around the currently most uncertain parts of the data distribution. GAUDA effectively improves downstream segmentation results over comparable methods by an average absolute IoU of 1.6% on CaDISv2 and 1.5% on CholecSeg8k, two prominent surgical datasets for semantic segmentation.
2501.10822
Addressing Multilabel Imbalance with an Efficiency-Focused Approach Using Diffusion Model-Generated Synthetic Samples
cs.LG cs.AI
Predictive models trained on imbalanced data tend to produce biased results. This problem is exacerbated when there is not just one output label, but a set of them. This is the case for multilabel learning (MLL) algorithms used to classify patterns, rank labels, or learn the distribution of outputs. Many solutions have been proposed in the literature. The one that can be applied universally, independent of the algorithm used to build the model, is data resampling. The generation of new instances associated with minority labels, so that empty areas of the feature space are filled, helps to improve the obtained models. The quality of these new instances depends on the algorithm used to generate them. In this paper, a diffusion model tailored to produce new instances for MLL data, called MLDM (\textit{MultiLabel Diffusion Model}), is proposed. Diffusion models have been mainly used to generate artificial images and videos. Our proposed MLDM is based on this type of models. The experiments conducted compare MLDM with several other MLL resampling algorithms. The results show that MLDM is competitive while it improves efficiency.
2501.10824
Information Content and Entropy of Finite Patterns from a Combinatorial Perspective
cs.IT cs.DM math.IT
A unified combinatorial definition of the information content and entropy of different types of patterns, compatible with the traditional concepts of information and entropy, going beyond the limitations of Shannon information interpretable for ergodic Markov processes. We compare the information content of various finite patterns and derive general properties of information quantity from these comparisons. Using these properties, we define normalized information estimation methods based on compression algorithms and Kolmogorov complexity. From a combinatorial point of view, we redefine the concept of entropy in a way that is asymptotically compatible with traditional entropy.
2501.10825
Statistical Design of Thermal Protection System Using Physics-Informed Machine learning
cs.CE
Estimating the material properties of thermal protection films is crucial for their effective design and application, particularly in high-temperature environments. This work presents a novel approach to determine the properties using uncertainty quantification simulations. We quantify uncertainty in the material properties for effective insulation by proposing a Bayesian distribution for them. Sampling from this distribution is performed using Monte Carlo simulations, which require repeatedly solving the predictive thermal model. To address the computational inefficiency of conventional numerical simulations, we develop a parametric Physics-Informed Neural Network (PINN) to solve the heat transfer problem. The proposed PINN significantly reduces computational time while maintaining accuracy, as verified against traditional numerical solutions. Additionally, we used the Sequential Monte Carlo (SMC) method to enable vectorized and parallel computations, further enhancing computational speedup. Our results demonstrate that integrating MCMC with PINN decreases computational time substantially compared to using standard numerical methods. Moreover, combining the SMC method with PINN yields multifold computational speedup, making this approach highly effective for the rapid and accurate estimation of material properties.
2501.10827
Integrating Expert and Physics Knowledge for Modeling Heat Load in District Heating Systems
eess.SY cs.SY
New residential neighborhoods are often supplied with heat via district heating systems (DHS). Improving the energy efficiency of a DHS is critical for increasing sustainability and satisfying user requirements. In this paper, we present HELIOS, a dedicated artificial intelligence (AI) model designed specifically for modeling the heat load in DHS. HELIOS leverages a combination of established physical principles and expert knowledge, resulting in superior performance compared to existing state-of-the-art models. HELIOS is explainable, enabling enhanced accountability and traceability in its predictions. We evaluate HELIOS against ten state-of-the-art data-driven models in modeling the heat load in a DHS case study in the Netherlands. HELIOS emerges as the top-performing model while maintaining complete accountability. The applications of HELIOS extend beyond the present case study, potentially supporting the adoption of AI by DHS and contributing to sustainable energy management on a larger scale.
2501.10834
Visual RAG: Expanding MLLM visual knowledge without fine-tuning
cs.CV cs.AI cs.LG
Multimodal Large Language Models (MLLMs) have achieved notable performance in computer vision tasks that require reasoning across visual and textual modalities, yet their capabilities are limited to their pre-trained data, requiring extensive fine-tuning for updates. Recent researches have explored the use of In-Context Learning (ICL) to overcome these challenges by providing a set of demonstrating examples as context to augment MLLMs performance in several tasks, showing that many-shot ICL leads to substantial improvements compared to few-shot ICL. However, the reliance on numerous demonstrating examples and the limited MLLMs context windows presents significant obstacles. This paper aims to address these challenges by introducing a novel approach, Visual RAG, that synergically combines the MLLMs capability to learn from the context, with a retrieval mechanism. The crux of this approach is to ensure to augment the MLLM knowledge by selecting only the most relevant demonstrating examples for the query, pushing it to learn by analogy. In this way, relying on the new information provided dynamically during inference time, the resulting system is not limited to the knowledge extracted from the training data, but can be updated rapidly and easily without fine-tuning. Furthermore, this greatly reduces the computational costs for improving the model image classification performance, and augments the model knowledge to new visual domains and tasks it was not trained for. Extensive experiments on eight different datasets in the state of the art spanning several domains and image classification tasks show that the proposed Visual RAG, compared to the most recent state of the art (i.e., many-shot ICL), is able to obtain an accuracy that is very close or even higher (approx. +2% improvement on average) while using a much smaller set of demonstrating examples (approx. only 23% on average).
2501.10835
Anatomy of a Historic Blackout: Decoding Spatiotemporal Dynamics of Power Outages and Disparities During Hurricane Beryl
cs.CE
This study investigates the spatial patterns and temporal variations in outage duration, intensity, and restoration/recovery following the 2024 Hurricane Beryl in Houston, Texas. This historic blackout caused widespread power disruptions across the Houston metropolitan area, leaving more than 2 million customers without power over several days, resulting in more than 143 million total customer-out hours.The findings reveal that areas with higher population density and proximity to the hurricane's path experienced more severe initial impacts. Regions with higher median income showed faster recovery, while lower-income areas exhibited prolonged restoration periods, even with favorable infrastructural conditions, suggesting disparities in restoration speed. The study also highlights how urban development features, such as road density and land elevation, explain spatial disparities in power outage impacts and recovery. This research advances the understanding of power outage dynamics in large metropolitan regions through four key contributions: (1) empirical characterization of outages from a historic hurricane, highlighting infrastructure vulnerabilities in a high-density urban context; (2) comprehensive analysis using multiple metrics to capture spatiotemporal dynamics of outages and restoration; (3) leveraging of high-resolution outage data at fine geographic scales and frequent intervals to quantify and reveal previously masked spatial disparities; and (4) systematic examination of socioeconomic, urban development, and environmental factors in shaping disparities in outage impacts and recovery timelines. These findings provide infrastructure managers, operators, utilities, and decision-makers with crucial empirical insights to quantify power outage impacts, justify resilience investments, and address vulnerability and equity issues in the power infrastructure during hazard events.
2501.10836
BAP v2: An Enhanced Task Framework for Instruction Following in Minecraft Dialogues
cs.CL cs.AI
Interactive agents capable of understanding and executing instructions in the physical world have long been a central goal in AI research. The Minecraft Collaborative Building Task (MCBT) provides one such setting to work towards this goal (Narayan-Chen, Jayannavar, and Hockenmaier 2019). It is a two-player game in which an Architect (A) instructs a Builder (B) to construct a target structure in a simulated Blocks World Environment. We focus on the challenging Builder Action Prediction (BAP) subtask of predicting correct action sequences in a given multimodal game context with limited training data (Jayannavar, Narayan-Chen, and Hockenmaier 2020). We take a closer look at evaluation and data for the BAP task, discovering key challenges and making significant improvements on both fronts to propose BAP v2, an upgraded version of the task. This will allow future work to make more efficient and meaningful progress on it. It comprises of: (1) an enhanced evaluation benchmark that includes a cleaner test set and fairer, more insightful metrics, and (2) additional synthetic training data generated from novel Minecraft dialogue and target structure simulators emulating the MCBT. We show that the synthetic data can be used to train more performant and robust neural models even with relatively simple training methods. Looking ahead, such data could also be crucial for training more sophisticated, data-hungry deep transformer models and training/fine-tuning increasingly large LLMs. Although modeling is not the primary focus of this work, we also illustrate the impact of our data and training methodologies on a simple LLM- and transformer-based model, thus validating the robustness of our approach, and setting the stage for more advanced architectures and LLMs going forward.
2501.10839
Systems Engineering for Autonomous Vehicles; Supervising AI using Large Language Models (SSuperLLM)
eess.SY cs.SY
Generative Artificial Intelligence (GAI) and the idea to use hierarchical models has been around for some years now. GAI has proved to be an extremely useful tool for Autonomous Vehicles (AVs). AVs need to perform robustly in their environment. Thus the AV behavior and short-term trajectory planning needs to be: a) designed and architected using safeguarding and supervisory systems and b) verified using proper Systems Engineering (SysEng) Principles. Can AV Systems Engineering also use Large Language Models (LLM) to help Autonomous vehicles (AV) development? This reader-friendly paper advocates the use of LLMs in 1) requirements (Reqs) development and 2) Reqs verification and 3) provides a proof-of-concept of AV supervisory control. The latter uses a simulation environment of a simple planar (bicycle) vehicle dynamics model and a Linear Quadratic Regulator (LQR) control with an LLM Application Interface (API). The Open-Source simulation SW is available from the author accessible to the readers so that they can engage into the AV stack, LLM API and rules, SysEng and Reqs and fundamental vehicle dynamics and control.
2501.10841
Practical and Ready-to-Use Methodology to Assess the re-identification Risk in Anonymized Datasets
cs.CR cs.AI cs.DB
To prove that a dataset is sufficiently anonymized, many privacy policies suggest that a re-identification risk assessment be performed, but do not provide a precise methodology for doing so, leaving the industry alone with the problem. This paper proposes a practical and ready-to-use methodology for re-identification risk assessment, the originality of which is manifold: (1) it is the first to follow well-known risk analysis methods (e.g. EBIOS) that have been used in the cybersecurity field for years, which consider not only the ability to perform an attack, but also the impact such an attack can have on an individual; (2) it is the first to qualify attributes and values of attributes with e.g. degree of exposure, as known real-world attacks mainly target certain types of attributes and not others.
2501.10842
BOOST: Microgrid Sizing using Ordinal Optimization
eess.SY cs.SY
The transition to sustainable energy systems has highlighted the critical need for efficient sizing of renewable energy resources in microgrids. In particular, designing photovoltaic (PV) and battery systems to meet residential loads is challenging due to trade-offs between cost, reliability, and environmental impact. While previous studies have employed dynamic programming and heuristic techniques for microgrid sizing, these approaches often fail to balance computational efficiency and accuracy. In this work, we propose BOOST, or Battery-solar Ordinal Optimization Sizing Technique, a novel framework for optimizing the sizing of PV and battery components in microgrids. Ordinal optimization enables computationally efficient evaluations of potential designs while preserving accuracy through robust ranking of solutions. To determine the optimal operation of the system at any given time, we introduce a mixed-integer linear programming (MILP) approach, which achieves lower costs than the commonly used dynamic programming methods. Our numerical experiments demonstrate that the proposed framework identifies optimal designs that achieve a levelized cost of energy (LCOE) as low as 8.84 cents/kWh, underscoring its potential for cost-effective microgrid design. The implications of our work are significant: BOOST provides a scalable and accurate methodology for integrating renewable energy into residential microgrids, addressing economic and environmental goals simultaneously.
2501.10848
Fake Advertisements Detection Using Automated Multimodal Learning: A Case Study for Vietnamese Real Estate Data
cs.LG cs.AI
The popularity of e-commerce has given rise to fake advertisements that can expose users to financial and data risks while damaging the reputation of these e-commerce platforms. For these reasons, detecting and removing such fake advertisements are important for the success of e-commerce websites. In this paper, we propose FADAML, a novel end-to-end machine learning system to detect and filter out fake online advertisements. Our system combines techniques in multimodal machine learning and automated machine learning to achieve a high detection rate. As a case study, we apply FADAML to detect fake advertisements on popular Vietnamese real estate websites. Our experiments show that we can achieve 91.5% detection accuracy, which significantly outperforms three different state-of-the-art fake news detection systems.
2501.10851
Exploring Siamese Networks in Self-Supervised Fast MRI Reconstruction
eess.IV cs.CV
Reconstructing MR images using deep neural networks from undersampled k-space data without using fully sampled training references offers significant value in practice, which is a self-supervised regression problem calling for effective prior knowledge and supervision. The Siamese architectures are motivated by the definition "invariance" and shows promising results in unsupervised visual representative learning. Building homologous transformed images and avoiding trivial solutions are two major challenges in Siamese-based self-supervised model. In this work, we explore Siamese architecture for MRI reconstruction in a self-supervised training fashion called SiamRecon. We show the proposed approach mimics an expectation maximization algorithm. The alternative optimization provide effective supervision signal and avoid collapse. The proposed SiamRecon achieves the state-of-the-art reconstruction accuracy in the field of self-supervised learning on both single-coil brain MRI and multi-coil knee MRI.
2501.10854
Achievable DoF Bounds for Cache-Aided Asymmetric MIMO Communications
cs.IT eess.SP math.IT
Integrating coded caching (CC) into multiple-input multiple-output (MIMO) communications can significantly enhance the achievable degrees of freedom (DoF) in wireless networks. This paper investigates a practical cache-aided asymmetric MIMO configuration with cache ratio $\gamma$, where a server equipped with $L$ transmit antennas communicates with $K$ users, each having $G_k$ receive antennas. We propose three content-aware MIMO-CC strategies: the \emph{min-G} scheme, which treats the system as symmetric by assuming all users have the same number of antennas, equal to the smallest among them; the \emph{Grouping} scheme, which maximizes spatial multiplexing gain separately within each user subset at the cost of some global caching gain; and the \emph{Phantom} scheme, which dynamically redistributes spatial resources using virtual or ``phantom'' antennas at the users, bridging the performance gains of the min-$G$ and Grouping schemes. These strategies jointly optimize the number of users, $\Omega$, and the parallel streams decoded by each user, $\beta_k$, ensuring linear decodability for all target users. Analytical and numerical results confirm that the proposed schemes achieve significant DoF improvements across various system configurations.
2501.10857
Learning Nonverbal Cues in Multiparty Social Interactions for Robotic Facilitators
cs.RO cs.LG
Conventional behavior cloning (BC) models often struggle to replicate the subtleties of human actions. Previous studies have attempted to address this issue through the development of a new BC technique: Implicit Behavior Cloning (IBC). This new technique consistently outperformed the conventional Mean Squared Error (MSE) BC models in a variety of tasks. Our goal is to replicate the performance of the IBC model by Florence [in Proceedings of the 5th Conference on Robot Learning, 164:158-168, 2022], for social interaction tasks using our custom dataset. While previous studies have explored the use of large language models (LLMs) for enhancing group conversations, they often overlook the significance of non-verbal cues, which constitute a substantial part of human communication. We propose using IBC to replicate nonverbal cues like gaze behaviors. The model is evaluated against various types of facilitator data and compared to an explicit, MSE BC model. Results show that the IBC model outperforms the MSE BC model across session types using the same metrics used in the previous IBC paper. Despite some metrics showing mixed results which are explainable for the custom dataset for social interaction, we successfully replicated the IBC model to generate nonverbal cues. Our contributions are (1) the replication and extension of the IBC model, and (2) a nonverbal cues generation model for social interaction. These advancements facilitate the integration of robots into the complex interactions between robots and humans, e.g., in the absence of a human facilitator.
2501.10858
Reliable Text-to-SQL with Adaptive Abstention
cs.DB cs.AI
Large language models (LLMs) have revolutionized natural language interfaces for databases, particularly in text-to-SQL conversion. However, current approaches often generate unreliable outputs when faced with ambiguity or insufficient context. We present Reliable Text-to-SQL (RTS), a novel framework that enhances query generation reliability by incorporating abstention and human-in-the-loop mechanisms. RTS focuses on the critical schema linking phase, which aims to identify the key database elements needed for generating SQL queries. It autonomously detects potential errors during the answer generation process and responds by either abstaining or engaging in user interaction. A vital component of RTS is the Branching Point Prediction (BPP) which utilizes statistical conformal techniques on the hidden layers of the LLM model for schema linking, providing probabilistic guarantees on schema linking accuracy. We validate our approach through comprehensive experiments on the BIRD benchmark, demonstrating significant improvements in robustness and reliability. Our findings highlight the potential of combining transparent-box LLMs with human-in-the-loop processes to create more robust natural language interfaces for databases. For the BIRD benchmark, our approach achieves near-perfect schema linking accuracy, autonomously involving a human when needed. Combined with query generation, we demonstrate that near-perfect schema linking and a small query generation model can almost match SOTA accuracy achieved with a model orders of magnitude larger than the one we use.
2501.10859
Which price to pay? Auto-tuning building MPC controller for optimal economic cost
eess.SY cs.LG cs.SY math.OC
Model predictive control (MPC) controller is considered for temperature management in buildings but its performance heavily depends on hyperparameters. Consequently, MPC necessitates meticulous hyperparameter tuning to attain optimal performance under diverse contracts. However, conventional building controller design is an open-loop process without critical hyperparameter optimization, often leading to suboptimal performance due to unexpected environmental disturbances and modeling errors. Furthermore, these hyperparameters are not adapted to different pricing schemes and may lead to non-economic operations. To address these issues, we propose an efficient performance-oriented building MPC controller tuning method based on a cutting-edge efficient constrained Bayesian optimization algorithm, CONFIG, with global optimality guarantees. We demonstrate that this technique can be applied to efficiently deal with real-world DSM program selection problems under customized black-box constraints and objectives. In this study, a simple MPC controller, which offers the advantages of reduced commissioning costs, enhanced computational efficiency, was optimized to perform on a comparable level to a delicately designed and computationally expensive MPC controller. The results also indicate that with an optimized simple MPC, the monthly electricity cost of a household can be reduced by up to 26.90% compared with the cost when controlled by a basic rule-based controller under the same constraints. Then we compared 12 real electricity contracts in Belgium for a household family with customized black-box occupant comfort constraints. The results indicate a monthly electricity bill saving up to 20.18% when the most economic contract is compared with the worst one, which again illustrates the significance of choosing a proper electricity contract.
2501.10860
Zero-shot and Few-shot Learning with Instruction-following LLMs for Claim Matching in Automated Fact-checking
cs.CL cs.AI
The claim matching (CM) task can benefit an automated fact-checking pipeline by putting together claims that can be resolved with the same fact-check. In this work, we are the first to explore zero-shot and few-shot learning approaches to the task. We consider CM as a binary classification task and experiment with a set of instruction-following large language models (GPT-3.5-turbo, Gemini-1.5-flash, Mistral-7B-Instruct, and Llama-3-8B-Instruct), investigating prompt templates. We introduce a new CM dataset, ClaimMatch, which will be released upon acceptance. We put LLMs to the test in the CM task and find that it can be tackled by leveraging more mature yet similar tasks such as natural language inference or paraphrase detection. We also propose a pipeline for CM, which we evaluate on texts of different lengths.
2501.10861
Dynamic Continual Learning: Harnessing Parameter Uncertainty for Improved Network Adaptation
cs.LG cs.AI
When fine-tuning Deep Neural Networks (DNNs) to new data, DNNs are prone to overwriting network parameters required for task-specific functionality on previously learned tasks, resulting in a loss of performance on those tasks. We propose using parameter-based uncertainty to determine which parameters are relevant to a network's learned function and regularize training to prevent change in these important parameters. We approach this regularization in two ways: (1), we constrain critical parameters from significant changes by associating more critical parameters with lower learning rates, thereby limiting alterations in those parameters; (2), important parameters are restricted from change by imposing a higher regularization weighting, causing parameters to revert to their states prior to the learning of subsequent tasks. We leverage a Bayesian Moment Propagation framework which learns network parameters concurrently with their associated uncertainties while allowing each parameter to contribute uncertainty to the network's predictive distribution, avoiding the pitfalls of existing sampling-based methods. The proposed approach is evaluated for common sequential benchmark datasets and compared to existing published approaches from the Continual Learning community. Ultimately, we show improved Continual Learning performance for Average Test Accuracy and Backward Transfer metrics compared to sampling-based methods and other non-uncertainty-based approaches.
2501.10866
QGAPHEnsemble : Combining Hybrid QLSTM Network Ensemble via Adaptive Weighting for Short Term Weather Forecasting
cs.LG
Accurate weather forecasting holds significant importance, serving as a crucial tool for decision-making in various industrial sectors. The limitations of statistical models, assuming independence among data points, highlight the need for advanced methodologies. The correlation between meteorological variables necessitate models capable of capturing complex dependencies. This research highlights the practical efficacy of employing advanced machine learning techniques proposing GenHybQLSTM and BO-QEnsemble architecture based on adaptive weight adjustment strategy. Through comprehensive hyper-parameter optimization using hybrid quantum genetic particle swarm optimisation algorithm and Bayesian Optimization, our model demonstrates a substantial improvement in the accuracy and reliability of meteorological predictions through the assessment of performance metrics such as MSE (Mean Squared Error) and MAPE (Mean Absolute Percentage Prediction Error). The paper highlights the importance of optimized ensemble techniques to improve the performance the given weather forecasting task.
2501.10868
Generating Structured Outputs from Language Models: Benchmark and Studies
cs.CL cs.AI
Reliably generating structured outputs has become a critical capability for modern language model (LM) applications. Constrained decoding has emerged as the dominant technology across sectors for enforcing structured outputs during generation. Despite its growing adoption, little has been done with the systematic evaluation of the behaviors and performance of constrained decoding. Constrained decoding frameworks have standardized around JSON Schema as a structured data format, with most uses guaranteeing constraint compliance given a schema. However, there is poor understanding of the effectiveness of the methods in practice. We present an evaluation framework to assess constrained decoding approaches across three critical dimensions: efficiency in generating constraint-compliant outputs, coverage of diverse constraint types, and quality of the generated outputs. To facilitate this evaluation, we introduce JSONSchemaBench, a benchmark for constrained decoding comprising 10K real-world JSON schemas that encompass a wide range of constraints with varying complexity. We pair the benchmark with the existing official JSON Schema Test Suite and evaluate six state-of-the-art constrained decoding frameworks, including Guidance, Outlines, Llamacpp, XGrammar, OpenAI, and Gemini. Through extensive experiments, we gain insights into the capabilities and limitations of constrained decoding on structured generation with real-world JSON schemas. Our work provides actionable insights for improving constrained decoding frameworks and structured generation tasks, setting a new standard for evaluating constrained decoding and structured generation. We release JSONSchemaBench at https://github.com/guidance-ai/jsonschemabench
2501.10869
Diffusion-Based Imitation Learning for Social Pose Generation
cs.LG cs.RO
Intelligent agents, such as robots and virtual agents, must understand the dynamics of complex social interactions to interact with humans. Effectively representing social dynamics is challenging because we require multi-modal, synchronized observations to understand a scene. We explore how using a single modality, the pose behavior, of multiple individuals in a social interaction can be used to generate nonverbal social cues for the facilitator of that interaction. The facilitator acts to make a social interaction proceed smoothly and is an essential role for intelligent agents to replicate in human-robot interactions. In this paper, we adapt an existing diffusion behavior cloning model to learn and replicate facilitator behaviors. Furthermore, we evaluate two representations of pose observations from a scene, one representation has pre-processing applied and one does not. The purpose of this paper is to introduce a new use for diffusion behavior cloning for pose generation in social interactions. The second is to understand the relationship between performance and computational load for generating social pose behavior using two different techniques for collecting scene observations. As such, we are essentially testing the effectiveness of two different types of conditioning for a diffusion model. We then evaluate the resulting generated behavior from each technique using quantitative measures such as mean per-joint position error (MPJPE), training time, and inference time. Additionally, we plot training and inference time against MPJPE to examine the trade-offs between efficiency and performance. Our results suggest that the further pre-processed data can successfully condition diffusion models to generate realistic social behavior, with reasonable trade-offs in accuracy and processing time.
2501.10870
Model-Robust and Adaptive-Optimal Transfer Learning for Tackling Concept Shifts in Nonparametric Regression
stat.ML cs.LG
When concept shifts and sample scarcity are present in the target domain of interest, nonparametric regression learners often struggle to generalize effectively. The technique of transfer learning remedies these issues by leveraging data or pre-trained models from similar source domains. While existing generalization analyses of kernel-based transfer learning typically rely on correctly specified models, we present a transfer learning procedure that is robust against model misspecification while adaptively attaining optimality. To facilitate our analysis and avoid the risk of saturation found in classical misspecified results, we establish a novel result in the misspecified single-task learning setting, showing that spectral algorithms with fixed bandwidth Gaussian kernels can attain minimax convergence rates given the true function is in a Sobolev space, which may be of independent interest. Building on this, we derive the adaptive convergence rates of the excess risk for specifying Gaussian kernels in a prevalent class of hypothesis transfer learning algorithms. Our results are minimax optimal up to logarithmic factors and elucidate the key determinants of transfer efficiency.
2501.10871
Enhancing User Intent for Recommendation Systems via Large Language Models
cs.IR
Recommendation systems play a critical role in enhancing user experience and engagement in various online platforms. Traditional methods, such as Collaborative Filtering (CF) and Content-Based Filtering (CBF), rely heavily on past user interactions or item features. However, these models often fail to capture the dynamic and evolving nature of user preferences. To address these limitations, we propose DUIP (Dynamic User Intent Prediction), a novel framework that combines LSTM networks with Large Language Models (LLMs) to dynamically capture user intent and generate personalized item recommendations. The LSTM component models the sequential and temporal dependencies of user behavior, while the LLM utilizes the LSTM-generated prompts to predict the next item of interest. Experimental results on three diverse datasets ML-1M, Games, and Bundle show that DUIP outperforms a wide range of baseline models, demonstrating its ability to handle the cold-start problem and real-time intent adaptation. The integration of dynamic prompts based on recent user interactions allows DUIP to provide more accurate, context-aware, and personalized recommendations. Our findings suggest that DUIP is a promising approach for next-generation recommendation systems, with potential for further improvements in cross-modal recommendations and scalability.
2501.10875
RIS Deployment Optimization with Iterative Detection and Decoding in Multiuser Multiple-Antenna Systems
cs.IT eess.SP math.IT
This work investigates a Reconfigurable Intelligent Surface (RIS)-assisted uplink system employing iterative detection and decoding (IDD) techniques. We analyze the impact of tuning system parameter tuning for several deployment configurations, including the number of users, access point (AP) antennas, and RIS elements on the IDD performance. Analytical results for both active and passive RIS in a single-input single-output (SISO) scenario demonstrate how deployment choices affect system performance. Numerical simulations confirm the robustness of the RIS-assisted IDD system to variations in these parameters, showing performance gains in certain configurations. Moreover, the findings indicate that the insights derived from SISO analysis extend to multiuser MIMO IDD systems.
2501.10876
Certifying Robustness via Topological Representations
stat.ML cs.CG cs.LG
We propose a neural network architecture that can learn discriminative geometric representations of data from persistence diagrams, common descriptors of Topological Data Analysis. The learned representations enjoy Lipschitz stability with a controllable Lipschitz constant. In adversarial learning, this stability can be used to certify $\epsilon$-robustness for samples in a dataset, which we demonstrate on the ORBIT5K dataset representing the orbits of a discrete dynamical system.
2501.10877
Distributed Quasi-Newton Method for Fair and Fast Federated Learning
cs.LG
Federated learning (FL) is a promising technology that enables edge devices/clients to collaboratively and iteratively train a machine learning model under the coordination of a central server. The most common approach to FL is first-order methods, where clients send their local gradients to the server in each iteration. However, these methods often suffer from slow convergence rates. As a remedy, second-order methods, such as quasi-Newton, can be employed in FL to accelerate its convergence. Unfortunately, similarly to the first-order FL methods, the application of second-order methods in FL can lead to unfair models, achieving high average accuracy while performing poorly on certain clients' local datasets. To tackle this issue, in this paper we introduce a novel second-order FL framework, dubbed \textbf{d}istributed \textbf{q}uasi-\textbf{N}ewton \textbf{fed}erated learning (DQN-Fed). This approach seeks to ensure fairness while leveraging the fast convergence properties of quasi-Newton methods in the FL context. Specifically, DQN-Fed helps the server update the global model in such a way that (i) all local loss functions decrease to promote fairness, and (ii) the rate of change in local loss functions aligns with that of the quasi-Newton method. We prove the convergence of DQN-Fed and demonstrate its \textit{linear-quadratic} convergence rate. Moreover, we validate the efficacy of DQN-Fed across a range of federated datasets, showing that it surpasses state-of-the-art fair FL methods in fairness, average accuracy and convergence speed.