id
stringlengths
9
16
title
stringlengths
4
278
categories
stringlengths
5
104
abstract
stringlengths
6
4.09k
2502.05836
LegalSeg: Unlocking the Structure of Indian Legal Judgments Through Rhetorical Role Classification
cs.CL cs.AI cs.IR cs.LG
In this paper, we address the task of semantic segmentation of legal documents through rhetorical role classification, with a focus on Indian legal judgments. We introduce LegalSeg, the largest annotated dataset for this task, comprising over 7,000 documents and 1.4 million sentences, labeled with 7 rhetorical roles. To benchmark performance, we evaluate multiple state-of-the-art models, including Hierarchical BiLSTM-CRF, TransformerOverInLegalBERT (ToInLegalBERT), Graph Neural Networks (GNNs), and Role-Aware Transformers, alongside an exploratory RhetoricLLaMA, an instruction-tuned large language model. Our results demonstrate that models incorporating broader context, structural relationships, and sequential sentence information outperform those relying solely on sentence-level features. Additionally, we conducted experiments using surrounding context and predicted or actual labels of neighboring sentences to assess their impact on classification accuracy. Despite these advancements, challenges persist in distinguishing between closely related roles and addressing class imbalance. Our work underscores the potential of advanced techniques for improving legal document understanding and sets a strong foundation for future research in legal NLP.
2502.05842
A Grid-Forming HVDC Series Tapping Converter Using Extended Techniques of Flex-LCC
eess.SY cs.SY
This paper discusses an extension technology for the previously proposed Flexible Line-Commutated Converter (Flex LCC) [1]. The proposed extension involves modifying the arm internal-electromotive-force control, redesigning the main-circuit parameters, and integrating a low-power coordination strategy. As a result, the Flex-LCC transforms from a grid-forming (GFM) voltage source converter (VSC) based on series-connected LCC and FBMMC into a novel GFM HVDC series tapping converter, referred to as the Extended Flex-LCC (EFLCC). The EFLCC provides dc characteristics resembling those of current source converters (CSCs) and ac characteristics resembling those of GFM VSCs. This makes it easier to integrate relatively small renewable energy sources (RESs) that operate in islanded or weak-grid supported conditions with an existing LCC-HVDC. Meanwhile, the EFLCC distinguishes itself by requiring fewer full-controlled switches and less energy storage, resulting in lower losses and costs compared to the FBMMC HVDC series tap solution. In particular, the reduced capacity requirement and the wide allowable range of valve-side ac voltages in the FBMMC part facilitate the matching of current-carrying capacities between full-controlled switches and thyristors. The application scenario, system-level analysis, implementation, converter-level operation, and comparison of the EFLCC are presented in detail in this paper. The theoretical analysis is confirmed by experimental and simulation results.
2502.05843
Training-free Anomaly Event Detection via LLM-guided Symbolic Pattern Discovery
cs.CV
Anomaly event detection plays a crucial role in various real-world applications. However, current approaches predominantly rely on supervised learning, which faces significant challenges: the requirement for extensive labeled training data and lack of interpretability in decision-making processes. To address these limitations, we present a training-free framework that integrates open-set object detection with symbolic regression, powered by Large Language Models (LLMs) for efficient symbolic pattern discovery. The LLMs guide the symbolic reasoning process, establishing logical relationships between detected entities. Through extensive experiments across multiple domains, our framework demonstrates several key advantages: (1) achieving superior detection accuracy through direct reasoning without any training process; (2) providing highly interpretable logical expressions that are readily comprehensible to humans; and (3) requiring minimal annotation effort - approximately 1% of the data needed by traditional training-based methods.To facilitate comprehensive evaluation and future research, we introduce two datasets: a large-scale private dataset containing over 110,000 annotated images covering various anomaly scenarios including construction site safety violations, illegal fishing activities, and industrial hazards, along with a public benchmark dataset of 5,000 samples with detailed anomaly event annotations. Code is available at here.
2502.05845
Exploiting the Hidden Capacity of MMC Through Accurate Quantification of Modulation Indices
eess.SY cs.SY
The modular multilevel converter (MMC) has become increasingly important in voltage-source converter-based high-voltage direct current (VSC-HVDC) systems. Direct and indirect modulation are widely used as mainstream modulation techniques in MMCs. However, due to the challenge of quantitatively evaluating the operation of different modulation schemes, the academic and industrial communities still hold differing opinions on their performance. To address this controversy, this paper employs the state-of-the-art computational methods and quantitative metrics to compare the performance among different modulation schemes. The findings indicate that direct modulation offers superior modulation potential for MMCs, highlighting its higher ac voltage output capability and broader linear PQ operation region. Conversely, indirect modulation is disadvantaged in linear modulation, which indicates inferior output voltage capability. Furthermore, this paper delves into the conditions whereby direct and indirect modulation techniques become equivalent in steady-state. The study findings suggest that the modulation capability of direct modulation is the same as that of indirect modulation in steady-state when additional controls, including closed-loop capacitor voltage control and circulating current suppression control (CCSC), are simultaneously active. Simulation and experiments verify the correctness and validity.
2502.05846
Rapid Detection of High-impedance Arc Faults in Medium Voltage Electrical Distribution Systems
eess.SY cs.SY
High-impedance arc faults in AC power systems have the potential to lead to catastrophic accidents. However, significant challenges exist in identifying these faults because of the much weaker characteristics and variety when grounded with different surfaces. Addressing a noteworthy gap in prior research, which largely focused on arc fault detection in low-voltage systems. A novel approach has been applied that offers rapid arc fault detection for medium voltage distribution lines. In contrast to existing black-box feature-based approaches, Hankel alternative view of the Koopman (HAVOK) analysis developed from nonlinear dynamics has been applied which offers not only interpretable features but also opens up new application options in the area of arc fault detection. The method displays a much faster detection speed within 0.45 ms making it appropriate for real-time applications. It demonstrates the ability to detect arc faults across various scenarios, boosting its practical importance for stakeholders in safety-critical industries.
2502.05849
Fact-or-Fair: A Checklist for Behavioral Testing of AI Models on Fairness-Related Queries
cs.CL
The generation of incorrect images, such as depictions of people of color in Nazi-era uniforms by Gemini, frustrated users and harmed Google's reputation, motivating us to investigate the relationship between accurately reflecting factuality and promoting diversity and equity. In this study, we focus on 19 real-world statistics collected from authoritative sources. Using these statistics, we develop a checklist comprising objective and subjective queries to analyze behavior of large language models (LLMs) and text-to-image (T2I) models. Objective queries assess the models' ability to provide accurate world knowledge. In contrast, the design of subjective queries follows a key principle: statistical or experiential priors should not be overgeneralized to individuals, ensuring that models uphold diversity. These subjective queries are based on three common human cognitive errors that often result in social biases. We propose metrics to assess factuality and fairness, and formally prove the inherent trade-off between these two aspects. Results show that GPT-4o and DALL-E 3 perform notably well among six LLMs and four T2I models. Our code is publicly available at https://github.com/uclanlp/Fact-or-Fair.
2502.05850
MetaML-Pro: Cross-Stage Design Flow Automation for Efficient Deep Learning Acceleration
cs.AR cs.LG
This paper presents a unified framework for codifying and automating optimization strategies to efficiently deploy deep neural networks (DNNs) on resource-constrained hardware, such as FPGAs, while maintaining high performance, accuracy, and resource efficiency. Deploying DNNs on such platforms involves addressing the significant challenge of balancing performance, resource usage (e.g., DSPs and LUTs), and inference accuracy, which often requires extensive manual effort and domain expertise. Our novel approach addresses two key issues: cross-stage co-optimization and optimization search. By seamlessly integrating programmatic DNN optimization techniques with high-level synthesis (HLS)-based metaprogramming and leveraging advanced design space exploration (DSE) strategies like Bayesian optimization, the framework automates both top-down and bottom-up design flows, reducing the need for manual intervention and domain expertise. The proposed framework introduces customizable optimization, transformation, and control blocks to enhance DNN accelerator performance and resource efficiency. Experimental results demonstrate up to a 92\% DSP and 89\% LUT usage reduction for select networks, while preserving accuracy, along with a 15.6-fold reduction in optimization time compared to grid search. These results underscore the novelty and potential of the proposed framework for automated, resource-efficient DNN accelerator designs.
2502.05851
Fairness Driven Slot Allocation Problem in Billboard Advertisement
cs.GT cs.DB cs.MA
In billboard advertisement, a number of digital billboards are owned by an influence provider, and several commercial houses (which we call advertisers) approach the influence provider for a specific number of views of their advertisement content on a payment basis. Though the billboard slot allocation problem has been studied in the literature, this problem still needs to be addressed from a fairness point of view. In this paper, we introduce the Fair Billboard Slot Allocation Problem, where the objective is to allocate a given set of billboard slots among a group of advertisers based on their demands fairly and efficiently. As fairness criteria, we consider the maximin fair share, which ensures that each advertiser will receive a subset of slots that maximizes the minimum share for all the advertisers. We have proposed a solution approach that generates an allocation and provides an approximate maximum fair share. The proposed methodology has been analyzed to understand its time and space requirements and a performance guarantee. It has been implemented with real-world trajectory and billboard datasets, and the results have been reported. The results show that the proposed approach leads to a balanced allocation by satisfying the maximin fairness criteria. At the same time, it maximizes the utility of advertisers.
2502.05853
Zak-Transform-Induced Optimal Sequences and Their Applications in OTFS
cs.IT math.IT
This paper introduces a novel finite Zak transform (FZT)-aided framework for constructing multiple zero-correlation zone (ZCZ) sequence sets with optimal correlation properties. Specifically, each sequence is perfect with zero auto-correlation sidelobes, each ZCZ sequence set meets the Tang-Fan-Matsufuji bound with equality, and the maximum inter-set cross-correlation of multiple sequence sets meets the Sarwate bound with equality. Our study shows that these sequences can be sparsely expressed in the Zak domain through properly selected index and phase matrices. Particularly, it is found that the maximum inter-set cross-correlation beats the Sarwate bound if every index matrix is a circular Florentine array. Several construction methods of multiple ZCZ sequence sets are proposed, demonstrating both the optimality and high flexibility. Additionally, it is shown that excellent synchronization performance can be achieved by the proposed sequences in orthogonal-time-frequency-space (OTFS) systems.
2502.05854
NSPG-Miner: Mining Repetitive Negative Sequential Patterns
cs.DB
Sequential pattern mining (SPM) with gap constraints (or repetitive SPM or tandem repeat discovery in bioinformatics) can find frequent repetitive subsequences satisfying gap constraints, which are called positive sequential patterns with gap constraints (PSPGs). However, classical SPM with gap constraints cannot find the frequent missing items in the PSPGs. To tackle this issue, this paper explores negative sequential patterns with gap constraints (NSPGs). We propose an efficient NSPG-Miner algorithm that can mine both frequent PSPGs and NSPGs simultaneously. To effectively reduce candidate patterns, we propose a pattern join strategy with negative patterns which can generate both positive and negative candidate patterns at the same time. To calculate the support (frequency of occurrence) of a pattern in each sequence, we explore a NegPair algorithm that employs a key-value pair array structure to deal with the gap constraints and the negative items simultaneously and can avoid redundant rescanning of the original sequence, thus improving the efficiency of the algorithm. To report the performance of NSPG-Miner, 11 competitive algorithms and 11 datasets are employed. The experimental results not only validate the effectiveness of the strategies adopted by NSPG-Miner, but also verify that NSPG-Miner can discover more valuable information than the state-of-the-art algorithms. Algorithms and datasets can be downloaded from https://github.com/wuc567/Pattern-Mining/tree/master/NSPG-Miner.
2502.05855
DexVLA: Vision-Language Model with Plug-In Diffusion Expert for General Robot Control
cs.RO cs.CV
Enabling robots to perform diverse tasks across varied environments is a central challenge in robot learning. While vision-language-action (VLA) models have shown promise for generalizable robot skills, realizing their full potential requires addressing limitations in action representation and efficient training. Current VLA models often focus on scaling the vision-language model (VLM) component, while the action space representation remains a critical bottleneck. This paper introduces DexVLA, a novel framework designed to enhance the efficiency and generalization capabilities of VLAs for complex, long-horizon tasks across diverse robot embodiments. DexVLA features a novel diffusion-based action expert, scaled to one billion parameters, designed for cross-embodiment learning. A novel embodiment curriculum learning strategy facilitates efficient training: (1) pre-training the diffusion expert that is separable from the VLA on cross-embodiment data, (2) aligning the VLA model to specific embodiments, and (3) post-training for rapid adaptation to new tasks. We conduct comprehensive experiments across multiple embodiments, including single-arm, bimanual, and dexterous hand, demonstrating DexVLA's adaptability to challenging tasks without task-specific adaptation, its ability to learn dexterous skills on novel embodiments with limited data, and its capacity to complete complex, long-horizon tasks using only direct language prompting, such as laundry folding. In all settings, our method demonstrates superior performance compared to state-of-the-art models like Octo, OpenVLA, and Diffusion Policy.
2502.05857
Acquisition through My Eyes and Steps: A Joint Predictive Agent Model in Egocentric Worlds
cs.CV cs.AI cs.LG
This paper addresses the task of learning an agent model behaving like humans, which can jointly perceive, predict, and act in egocentric worlds. Previous methods usually train separate models for these three abilities, leading to information silos among them, which prevents these abilities from learning from each other and collaborating effectively. In this paper, we propose a joint predictive agent model, named EgoAgent, that simultaneously learns to represent the world, predict future states, and take reasonable actions with a single transformer. EgoAgent unifies the representational spaces of the three abilities by mapping them all into a sequence of continuous tokens. Learnable query tokens are appended to obtain current states, future states, and next actions. With joint supervision, our agent model establishes the internal relationship among these three abilities and effectively mimics the human inference and learning processes. Comprehensive evaluations of EgoAgent covering image classification, egocentric future state prediction, and 3D human motion prediction tasks demonstrate the superiority of our method. The code and trained model will be released for reproducibility.
2502.05858
Let's Have Both! Optimal List-Recoverability via Alphabet Permutation Codes
cs.IT math.IT
We construct a new family of codes that requires only polynomial randomness yet achieves $(\rho,\ell,L)$-list-recoverability at a rate within $\epsilon$ of capacity, with $L \approx \tfrac{\ell}{\epsilon}$. In contrast, every previous construction using polynomial randomness required an exponentially larger list size. Our approach extends earlier work by Li and Wootters (2021) on the list-decodability of random linear binary codes.
2502.05859
SphereFusion: Efficient Panorama Depth Estimation via Gated Fusion
cs.CV
Due to the rapid development of panorama cameras, the task of estimating panorama depth has attracted significant attention from the computer vision community, especially in applications such as robot sensing and autonomous driving. However, existing methods relying on different projection formats often encounter challenges, either struggling with distortion and discontinuity in the case of equirectangular, cubemap, and tangent projections, or experiencing a loss of texture details with the spherical projection. To tackle these concerns, we present SphereFusion, an end-to-end framework that combines the strengths of various projection methods. Specifically, SphereFusion initially employs 2D image convolution and mesh operations to extract two distinct types of features from the panorama image in both equirectangular and spherical projection domains. These features are then projected onto the spherical domain, where a gate fusion module selects the most reliable features for fusion. Finally, SphereFusion estimates panorama depth within the spherical domain. Meanwhile, SphereFusion employs a cache strategy to improve the efficiency of mesh operation. Extensive experiments on three public panorama datasets demonstrate that SphereFusion achieves competitive results with other state-of-the-art methods, while presenting the fastest inference speed at only 17 ms on a 512$\times$1024 panorama image.
2502.05863
Uni-Retrieval: A Multi-Style Retrieval Framework for STEM's Education
cs.IR cs.AI cs.MM
In AI-facilitated teaching, leveraging various query styles to interpret abstract text descriptions is crucial for ensuring high-quality teaching. However, current retrieval models primarily focus on natural text-image retrieval, making them insufficiently tailored to educational scenarios due to the ambiguities in the retrieval process. In this paper, we propose a diverse expression retrieval task tailored to educational scenarios, supporting retrieval based on multiple query styles and expressions. We introduce the STEM Education Retrieval Dataset (SER), which contains over 24,000 query pairs of different styles, and the Uni-Retrieval, an efficient and style-diversified retrieval vision-language model based on prompt tuning. Uni-Retrieval extracts query style features as prototypes and builds a continuously updated Prompt Bank containing prompt tokens for diverse queries. This bank can updated during test time to represent domain-specific knowledge for different subject retrieval scenarios. Our framework demonstrates scalability and robustness by dynamically retrieving prompt tokens based on prototype similarity, effectively facilitating learning for unknown queries. Experimental results indicate that Uni-Retrieval outperforms existing retrieval models in most retrieval tasks. This advancement provides a scalable and precise solution for diverse educational needs.
2502.05864
Learning Accurate, Efficient, and Interpretable MLPs on Multiplex Graphs via Node-wise Multi-View Ensemble Distillation
cs.LG
Multiplex graphs, with multiple edge types (graph views) among common nodes, provide richer structural semantics and better modeling capabilities. Multiplex Graph Neural Networks (MGNNs), typically comprising view-specific GNNs and a multi-view integration layer, have achieved advanced performance in various downstream tasks. However, their reliance on neighborhood aggregation poses challenges for deployment in latency-sensitive applications. Motivated by recent GNN-to-MLP knowledge distillation frameworks, we propose Multiplex Graph-Free Neural Networks (MGFNN and MGFNN+) to combine MGNNs' superior performance and MLPs' efficient inference via knowledge distillation. MGFNN directly trains student MLPs with node features as input and soft labels from teacher MGNNs as targets. MGFNN+ further employs a low-rank approximation-based reparameterization to learn node-wise coefficients, enabling adaptive knowledge ensemble from each view-specific GNN. This node-wise multi-view ensemble distillation strategy allows student MLPs to learn more informative multiplex semantic knowledge for different nodes. Experiments show that MGFNNs achieve average accuracy improvements of about 10% over vanilla MLPs and perform comparably or even better to teacher MGNNs (accurate); MGFNNs achieve a 35.40$\times$-89.14$\times$ speedup in inference over MGNNs (efficient); MGFNN+ adaptively assigns different coefficients for multi-view ensemble distillation regarding different nodes (interpretable).
2502.05867
Self-Training Large Language Models for Tool-Use Without Demonstrations
cs.CL
Large language models (LLMs) remain prone to factual inaccuracies and computational errors, including hallucinations and mistakes in mathematical reasoning. Recent work augmented LLMs with tools to mitigate these shortcomings, but often requires curated gold tool-use demonstrations. In this paper, we investigate whether LLMs can learn to use tools without demonstrations. First, we analyse zero-shot prompting strategies to guide LLMs in tool utilisation. Second, we propose a self-training method to synthesise tool-use traces using the LLM itself. We compare supervised fine-tuning and preference fine-tuning techniques for fine-tuning the model on datasets constructed using existing Question Answering (QA) datasets, i.e., TriviaQA and GSM8K. Experiments show that tool-use enhances performance on a long-tail knowledge task: 3.7% on PopQA, which is used solely for evaluation, but leads to mixed results on other datasets, i.e., TriviaQA, GSM8K, and NQ-Open. Our findings highlight the potential and challenges of integrating external tools into LLMs without demonstrations.
2502.05868
Norm Augmented Graph AutoEncoders for Link Prediction
cs.LG
Link Prediction (LP) is a crucial problem in graph-structured data. Graph Neural Networks (GNNs) have gained prominence in LP, with Graph AutoEncoders (GAEs) being a notable representation. However, our empirical findings reveal that GAEs' LP performance suffers heavily from the long-tailed node degree distribution, i.e., low-degree nodes tend to exhibit inferior LP performance compared to high-degree nodes. \emph{What causes this degree-related bias, and how can it be mitigated?} In this study, we demonstrate that the norm of node embeddings learned by GAEs exhibits variation among nodes with different degrees, underscoring its central significance in influencing the final performance of LP. Specifically, embeddings with larger norms tend to guide the decoder towards predicting higher scores for positive links and lower scores for negative links, thereby contributing to superior performance. This observation motivates us to improve GAEs' LP performance on low-degree nodes by increasing their embedding norms, which can be implemented simply yet effectively by introducing additional self-loops into the training objective for low-degree nodes. This norm augmentation strategy can be seamlessly integrated into existing GAE methods with light computational cost. Extensive experiments on various datasets and GAE methods show the superior performance of norm-augmented GAEs.
2502.05869
HyLiFormer: Hyperbolic Linear Attention for Skeleton-based Human Action Recognition
cs.CV
Transformers have demonstrated remarkable performance in skeleton-based human action recognition, yet their quadratic computational complexity remains a bottleneck for real-world applications. To mitigate this, linear attention mechanisms have been explored but struggle to capture the hierarchical structure of skeleton data. Meanwhile, the Poincar\'e model, as a typical hyperbolic geometry, offers a powerful framework for modeling hierarchical structures but lacks well-defined operations for existing mainstream linear attention. In this paper, we propose HyLiFormer, a novel hyperbolic linear attention Transformer tailored for skeleton-based action recognition. Our approach incorporates a Hyperbolic Transformation with Curvatures (HTC) module to map skeleton data into hyperbolic space and a Hyperbolic Linear Attention (HLA) module for efficient long-range dependency modeling. Theoretical analysis and extensive experiments on NTU RGB+D and NTU RGB+D 120 datasets demonstrate that HyLiFormer significantly reduces computational complexity while preserving model accuracy, making it a promising solution for efficiency-critical applications.
2502.05874
MMGDreamer: Mixed-Modality Graph for Geometry-Controllable 3D Indoor Scene Generation
cs.CV cs.AI cs.LG
Controllable 3D scene generation has extensive applications in virtual reality and interior design, where the generated scenes should exhibit high levels of realism and controllability in terms of geometry. Scene graphs provide a suitable data representation that facilitates these applications. However, current graph-based methods for scene generation are constrained to text-based inputs and exhibit insufficient adaptability to flexible user inputs, hindering the ability to precisely control object geometry. To address this issue, we propose MMGDreamer, a dual-branch diffusion model for scene generation that incorporates a novel Mixed-Modality Graph, visual enhancement module, and relation predictor. The mixed-modality graph allows object nodes to integrate textual and visual modalities, with optional relationships between nodes. It enhances adaptability to flexible user inputs and enables meticulous control over the geometry of objects in the generated scenes. The visual enhancement module enriches the visual fidelity of text-only nodes by constructing visual representations using text embeddings. Furthermore, our relation predictor leverages node representations to infer absent relationships between nodes, resulting in more coherent scene layouts. Extensive experimental results demonstrate that MMGDreamer exhibits superior control of object geometry, achieving state-of-the-art scene generation performance. Project page: https://yangzhifeio.github.io/project/MMGDreamer.
2502.05878
Enhancing Financial Time-Series Forecasting with Retrieval-Augmented Large Language Models
cs.CL
Stock movement prediction, a critical task in financial time-series forecasting, relies on identifying and retrieving key influencing factors from vast and complex datasets. However, traditional text-trained or numeric similarity-based retrieval methods often struggle to handle the intricacies of financial data. To address this, we propose the first retrieval-augmented generation (RAG) framework specifically designed for financial time-series forecasting. Our framework incorporates three key innovations: a fine-tuned 1B large language model (StockLLM) as its backbone, a novel candidate selection method enhanced by LLM feedback, and a training objective that maximizes the similarity between queries and historically significant sequences. These advancements enable our retriever, FinSeer, to uncover meaningful patterns while effectively minimizing noise in complex financial datasets. To support robust evaluation, we also construct new datasets that integrate financial indicators and historical stock prices. Experimental results demonstrate that our RAG framework outperforms both the baseline StockLLM and random retrieval methods, showcasing its effectiveness. FinSeer, as the retriever, achieves an 8% higher accuracy on the BIGDATA22 benchmark and retrieves more impactful sequences compared to existing retrieval methods. This work highlights the importance of tailored retrieval models in financial forecasting and provides a novel, scalable framework for future research in the field.
2502.05879
Enhancing Depression Detection with Chain-of-Thought Prompting: From Emotion to Reasoning Using Large Language Models
cs.CL cs.AI
Depression is one of the leading causes of disability worldwide, posing a severe burden on individuals, healthcare systems, and society at large. Recent advancements in Large Language Models (LLMs) have shown promise in addressing mental health challenges, including the detection of depression through text-based analysis. However, current LLM-based methods often struggle with nuanced symptom identification and lack a transparent, step-by-step reasoning process, making it difficult to accurately classify and explain mental health conditions. To address these challenges, we propose a Chain-of-Thought Prompting approach that enhances both the performance and interpretability of LLM-based depression detection. Our method breaks down the detection process into four stages: (1) sentiment analysis, (2) binary depression classification, (3) identification of underlying causes, and (4) assessment of severity. By guiding the model through these structured reasoning steps, we improve interpretability and reduce the risk of overlooking subtle clinical indicators. We validate our method on the E-DAIC dataset, where we test multiple state-of-the-art large language models. Experimental results indicate that our Chain-of-Thought Prompting technique yields superior performance in both classification accuracy and the granularity of diagnostic insights, compared to baseline approaches.
2502.05883
NeuralPrefix: A Zero-shot Sensory Data Imputation Plugin
cs.LG cs.AI stat.ML
Real-world sensing challenges such as sensor failures, communication issues, and power constraints lead to data intermittency. An issue that is known to undermine the traditional classification task that assumes a continuous data stream. Previous works addressed this issue by designing bespoke solutions (i.e. task-specific and/or modality-specific imputation). These approaches, while effective for their intended purposes, had limitations in their applicability across different tasks and sensor modalities. This raises an important question: Can we build a task-agnostic imputation pipeline that is transferable to new sensors without requiring additional training? In this work, we formalise the concept of zero-shot imputation and propose a novel approach that enables the adaptation of pre-trained models to handle data intermittency. This framework, named NeuralPrefix, is a generative neural component that precedes a task model during inference, filling in gaps caused by data intermittency. NeuralPrefix is built as a continuous dynamical system, where its internal state can be estimated at any point in time by solving an Ordinary Differential Equation (ODE). This approach allows for a more versatile and adaptable imputation method, overcoming the limitations of task-specific and modality-specific solutions. We conduct a comprehensive evaluation of NeuralPrefix on multiple sensory datasets, demonstrating its effectiveness across various domains. When tested on intermittent data with a high 50% missing data rate, NeuralPreifx accurately recovers all the missing samples, achieving SSIM score between 0.93-0.96. Zero-shot evaluations show that NeuralPrefix generalises well to unseen datasets, even when the measurements come from a different modality.
2502.05884
Study of Robust Multiuser Scheduling and Power Allocation in Cell-Free MIMO Networks
cs.IT math.IT
This paper introduces a robust resource allocation framework for the downlink of cell-free massive multi-input multi-output (CF-mMIMO) networks to address the effects caused by imperfect channel state information (CSI). In particular, the proposed robust resource allocation framework includes a robust user scheduling algorithm to optimize the network's sum-rate and a robust power allocation technique aimed at minimizing the mean square error (MSE) for a network with a linear precoder. Unlike non-robust resource allocation techniques, the proposed robust strategies effectively counteract the effects of imperfect CSI, enhancing network efficiency and reliability. Simulation results show a significant improvement in network performance obtained by the proposed approaches, highlighting the impact of robust resource allocation in wireless networks.
2502.05887
MTPChat: A Multimodal Time-Aware Persona Dataset for Conversational Agents
cs.CL cs.AI
Understanding temporal dynamics is critical for conversational agents, enabling effective content analysis and informed decision-making. However, time-aware datasets, particularly for persona-grounded conversations, are still limited, which narrows their scope and diminishes their complexity. To address this gap, we introduce MTPChat, a multimodal, time-aware persona dialogue dataset that integrates linguistic, visual, and temporal elements within dialogue and persona memory. Leveraging MTPChat, we propose two time-sensitive tasks: Temporal Next Response Prediction (TNRP) and Temporal Grounding Memory Prediction (TGMP), both designed to assess a model's ability to understand implicit temporal cues and dynamic interactions. Additionally, we present an innovative framework featuring an adaptive temporal module to effectively integrate multimodal streams and capture temporal dependencies. Experimental results validate the challenges posed by MTPChat and demonstrate the effectiveness of our framework in multimodal time-sensitive scenarios.
2502.05891
Room-scale magnetoquasistatic wireless power transfer using a cavity-based multimode resonator
physics.app-ph cs.SY eess.SY
Magnetoquasistatic wireless power transfer can be used to charge and power electronic devices such as smartphones and small home appliances. However, existing coil-based transmitters, which are composed of wire conductors, have a limited range. Here we show that multimode quasistatic cavity resonance can provide room-scale wireless power transfer. The approach uses multidirectional, widely distributed currents on conductive surfaces that are placed around the target volume. It generates multiple, mutually unique, three-dimensional magnetic field patterns, where each pattern is attributed to different eigenmodes of a single room-scale resonator. Using these modes together, a power delivery efficiency exceeding 37.1% can be achieved throughout a 3 m * 3 m * 2 m test room. With this approach, power exceeding 50 W could potentially be delivered to mobile receivers in accordance with safety guidelines.
2502.05892
A Distributional Perspective on Word Learning in Neural Language Models
cs.CL cs.AI
Language models (LMs) are increasingly being studied as models of human language learners. Due to the nascency of the field, it is not well-established whether LMs exhibit similar learning dynamics to humans, and there are few direct comparisons between learning trajectories in humans and models. Word learning trajectories for children are relatively well-documented, and recent work has tried to extend these investigations to language models. However, there are no widely agreed-upon metrics for word learning in language models. We take a distributional approach to this problem, defining lexical knowledge in terms of properties of the learned distribution for a target word. We argue that distributional signatures studied in prior work fail to capture key distributional information. Thus, we propose an array of signatures that improve on earlier approaches by capturing knowledge of both where the target word can and cannot occur as well as gradient preferences about the word's appropriateness. We obtain learning trajectories for a selection of small language models we train from scratch, study the relationship between different distributional signatures, compare how well they align with human word learning trajectories and interpretable lexical features, and address basic methodological questions about estimating these distributional signatures. Our metrics largely capture complementary information, suggesting that it is important not to rely on a single metric. However, across all metrics, language models' learning trajectories fail to correlate with those of children.
2502.05894
Suppressing Leakage Magnetic Field in Wireless Power Transfer using Halbach Array-Based Resonators
physics.app-ph cs.SY eess.SY
Wireless power transfer has the potential to seamlessly power electronic systems, such as electric vehicles, industrial robots, and mobile devices. However, the leakage magnetic field is a critical bottleneck that limits the transferable power level, and heavy ferromagnetic shields are needed for transferring large amounts of power. In this paper, we propose a ferrite-less coil design that generates an asymmetric magnetic field pattern focused on one side of the resonator, which effectively reduces the leakage magnetic field. The key to enabling the asymmetric field pattern is a coil winding strategy inspired by the Halbach array, a permanent magnet arrangement, which is then tailored for wireless power using an evolutionary strategy algorithm. Numerical analyses and simulations demonstrated that the proposed coil structure delivers the same amount of power as spiral coils, while achieving an 86.6% reduction in magnetic field intensity at a plane located 75 mm away from the resonator pair and a power efficiency of 96.0%. We verified our approach by measuring the power efficiency and magnetic field intensity of a test wireless power system operating at 6.78 MHz. These findings indicate that our approach can efficiently deliver over 50 times more power without increasing magnetic field exposure, making it a promising solution for high-power wireless power transfer applications.
2502.05895
Beyond Fine-Tuning: A Systematic Study of Sampling Techniques in Personalized Image Generation
cs.CV
Personalized text-to-image generation aims to create images tailored to user-defined concepts and textual descriptions. Balancing the fidelity of the learned concept with its ability for generation in various contexts presents a significant challenge. Existing methods often address this through diverse fine-tuning parameterizations and improved sampling strategies that integrate superclass trajectories during the diffusion process. While improved sampling offers a cost-effective, training-free solution for enhancing fine-tuned models, systematic analyses of these methods remain limited. Current approaches typically tie sampling strategies with fixed fine-tuning configurations, making it difficult to isolate their impact on generation outcomes. To address this issue, we systematically analyze sampling strategies beyond fine-tuning, exploring the impact of concept and superclass trajectories on the results. Building on this analysis, we propose a decision framework evaluating text alignment, computational constraints, and fidelity objectives to guide strategy selection. It integrates with diverse architectures and training approaches, systematically optimizing concept preservation, prompt adherence, and resource efficiency. The source code can be found at https://github.com/ControlGenAI/PersonGenSampler.
2502.05902
Fast Omni-Directional Image Super-Resolution: Adapting the Implicit Image Function with Pixel and Semantic-Wise Spherical Geometric Priors
cs.CV
In the context of Omni-Directional Image (ODI) Super-Resolution (SR), the unique challenge arises from the non-uniform oversampling characteristics caused by EquiRectangular Projection (ERP). Considerable efforts in designing complex spherical convolutions or polyhedron reprojection offer significant performance improvements but at the expense of cumbersome processing procedures and slower inference speeds. Under these circumstances, this paper proposes a new ODI-SR model characterized by its capacity to perform Fast and Arbitrary-scale ODI-SR processes, denoted as FAOR. The key innovation lies in adapting the implicit image function from the planar image domain to the ERP image domain by incorporating spherical geometric priors at both the latent representation and image reconstruction stages, in a low-overhead manner. Specifically, at the latent representation stage, we adopt a pair of pixel-wise and semantic-wise sphere-to-planar distortion maps to perform affine transformations on the latent representation, thereby incorporating it with spherical properties. Moreover, during the image reconstruction stage, we introduce a geodesic-based resampling strategy, aligning the implicit image function with spherical geometrics without introducing additional parameters. As a result, the proposed FAOR outperforms the state-of-the-art ODI-SR models with a much faster inference speed. Extensive experimental results and ablation studies have demonstrated the effectiveness of our design.
2502.05905
QP-SNN: Quantized and Pruned Spiking Neural Networks
cs.CV
Brain-inspired Spiking Neural Networks (SNNs) leverage sparse spikes to encode information and operate in an asynchronous event-driven manner, offering a highly energy-efficient paradigm for machine intelligence. However, the current SNN community focuses primarily on performance improvement by developing large-scale models, which limits the applicability of SNNs in resource-limited edge devices. In this paper, we propose a hardware-friendly and lightweight SNN, aimed at effectively deploying high-performance SNN in resource-limited scenarios. Specifically, we first develop a baseline model that integrates uniform quantization and structured pruning, called QP-SNN baseline. While this baseline significantly reduces storage demands and computational costs, it suffers from performance decline. To address this, we conduct an in-depth analysis of the challenges in quantization and pruning that lead to performance degradation and propose solutions to enhance the baseline's performance. For weight quantization, we propose a weight rescaling strategy that utilizes bit width more effectively to enhance the model's representation capability. For structured pruning, we propose a novel pruning criterion using the singular value of spatiotemporal spike activities to enable more accurate removal of redundant kernels. Extensive experiments demonstrate that integrating two proposed methods into the baseline allows QP-SNN to achieve state-of-the-art performance and efficiency, underscoring its potential for enhancing SNN deployment in edge intelligence computing.
2502.05907
EvoAgent: Agent Autonomous Evolution with Continual World Model for Long-Horizon Tasks
cs.RO
Completing Long-Horizon (LH) tasks in open-ended worlds is an important yet difficult problem for embodied agents. Existing approaches suffer from two key challenges: (1) they heavily rely on experiences obtained from human-created data or curricula, lacking the ability to continuously update multimodal experiences, and (2) they may encounter catastrophic forgetting issues when faced with new tasks, lacking the ability to continuously update world knowledge. To solve these challenges, this paper presents EvoAgent, an autonomous-evolving agent with a continual World Model (WM), which can autonomously complete various LH tasks across environments through self-planning, self-control, and self-reflection, without human intervention. Our proposed EvoAgent contains three modules, i.e., i) the memory-driven planner which uses an LLM along with the WM and interaction memory, to convert LH tasks into executable sub-tasks; ii) the WM-guided action controller which leverages WM to generate low-level actions and incorporates a self-verification mechanism to update multimodal experiences; iii) the experience-inspired reflector which implements a two-stage curriculum learning algorithm to select experiences for task-adaptive WM updates. Moreover, we develop a continual World Model for EvoAgent, which can continuously update the multimodal experience pool and world knowledge through closed-loop dynamics. We conducted extensive experiments on Minecraft, compared with existing methods, EvoAgent can achieve an average success rate improvement of 105% and reduce ineffective actions by more than 6x.
2502.05908
Inverse Problem Sampling in Latent Space Using Sequential Monte Carlo
eess.IV cs.CV cs.LG
In image processing, solving inverse problems is the task of finding plausible reconstructions of an image that was corrupted by some (usually known) degradation model. Commonly, this process is done using a generative image model that can guide the reconstruction towards solutions that appear natural. The success of diffusion models over the last few years has made them a leading candidate for this task. However, the sequential nature of diffusion models makes this conditional sampling process challenging. Furthermore, since diffusion models are often defined in the latent space of an autoencoder, the encoder-decoder transformations introduce additional difficulties. Here, we suggest a novel sampling method based on sequential Monte Carlo (SMC) in the latent space of diffusion models. We use the forward process of the diffusion model to add additional auxiliary observations and then perform an SMC sampling as part of the backward process. Empirical evaluations on ImageNet and FFHQ show the benefits of our approach over competing methods on various inverse problem tasks.
2502.05911
GRAIT: Gradient-Driven Refusal-Aware Instruction Tuning for Effective Hallucination Mitigation
cs.CL
Refusal-Aware Instruction Tuning (RAIT) aims to enhance Large Language Models (LLMs) by improving their ability to refuse responses to questions beyond their knowledge, thereby reducing hallucinations and improving reliability. Effective RAIT must address two key challenges: firstly, effectively reject unknown questions to minimize hallucinations; secondly, avoid over-refusal to ensure questions that can be correctly answered are not rejected, thereby maintain the helpfulness of LLM outputs. In this paper, we address the two challenges by deriving insightful observations from the gradient-based perspective, and proposing the Gradient-driven Refusal Aware Instruction Tuning Framework GRAIT: (1) employs gradient-driven sample selection to effectively minimize hallucinations and (2) introduces an adaptive weighting mechanism during fine-tuning to reduce the risk of over-refusal, achieving the balance between accurate refusals and maintaining useful responses. Experimental evaluations on open-ended and multiple-choice question answering tasks demonstrate that GRAIT significantly outperforms existing RAIT methods in the overall performance. The source code and data will be available at https://github.com/opendatalab/GRAIT .
2502.05912
LpBound: Pessimistic Cardinality Estimation using $\ell_p$-Norms of Degree Sequences
cs.DB
Cardinality estimation is the problem of estimating the size of the output of a query, without actually evaluating the query. The cardinality estimator is a critical piece of a query optimizer, and is often the main culprit when the optimizer chooses a poor plan. This paper introduces LpBound, a pessimistic cardinality estimator for multijoin queries (acyclic or cyclic) with selection predicates and group-by clauses. LpBound computes a guaranteed upper bound on the size of the query output using simple statistics on the input relations, consisting of $\ell_p$-norms of degree sequences. The bound is the optimal solution of a linear program whose constraints encode data statistics and Shannon inequalities. We introduce two optimizations that exploit the structure of the query in order to speed up the estimation time and make LpBound practical. We experimentally evaluate LpBound against a range of traditional, pessimistic, and machine learning-based estimators on the JOB, STATS, and subgraph matching benchmarks. Our main finding is that LpBound can be orders of magnitude more accurate than traditional estimators used in mainstream open-source and commercial database systems. Yet it has comparable low estimation time and space requirements. When injected the estimates of LpBound, Postgres derives query plans at least as good as those derived using the true cardinalities.
2502.05916
Adaptive Grasping of Moving Objects in Dense Clutter via Global-to-Local Detection and Static-to-Dynamic Planning
cs.RO cs.SY eess.SY
Robotic grasping is facing a variety of real-world uncertainties caused by non-static object states, unknown object properties, and cluttered object arrangements. The difficulty of grasping increases with the presence of more uncertainties, where commonly used learning-based approaches struggle to perform consistently across varying conditions. In this study, we integrate the idea of similarity matching to tackle the challenge of grasping novel objects that are simultaneously in motion and densely cluttered using a single RGBD camera, where multiple uncertainties coexist. We achieve this by shifting visual detection from global to local states and operating grasp planning from static to dynamic scenes. Notably, we introduce optimization methods to enhance planning efficiency for this time-sensitive task. Our proposed system can adapt to various object types, arrangements and movement speeds without the need for extensive training, as demonstrated by real-world experiments. Videos are available at https://youtu.be/sdC50dx-xp8?si=27oVr4dhG0rqN_tT.
2502.05917
Modeling and Beamforming Optimization for Pinching-Antenna Systems
cs.IT math.IT
The Pinching-Antenna SyStem (PASS) is a revolutionary flexible antenna technology designed to enhance wireless communication by establishing strong line-of-sight (LoS) links, reducing free-space path loss and enabling antenna array reconfigurability. PASS uses dielectric waveguides with low propagation loss for signal transmission, radiating via a passive pinching antenna, which is a small dielectric element applied to the waveguide. This paper first proposes a physics-based hardware model for PASS, where the pinching antenna is modeled as an open-ended directional coupler, and the electromagnetic field behavior is analyzed using coupled-mode theory. A simplified signal model characterizes the coupling effect between multiple antennas on the same waveguide. Based on this, two power models are proposed: equal power and proportional power models. Additionally, a transmit power minimization problem is formulated/studied for the joint optimization of transmit and pinching beamforming under both continuous and discrete pinching antenna activations. Two algorithms are proposed to solve this multimodal optimization problem: the penalty-based alternating optimization algorithm and a low-complexity zero-forcing (ZF)-based algorithm. Numerical results show that 1) the ZF-based low-complexity algorithm performs similarly to the penalty-based algorithm, 2) PASS reduces transmit power by over 95% compared to conventional and massive MIMO, 3) discrete activation causes minimal performance loss but requires a dense antenna set to match continuous activation, and 4) the proportional power model yields performance comparable to the equal power model.
2502.05919
Can Generative Agent-Based Modeling Replicate the Friendship Paradox in Social Media Simulations?
cs.SI
Generative Agent-Based Modeling (GABM) is an emerging simulation paradigm that combines the reasoning abilities of Large Language Models with traditional Agent-Based Modeling to replicate complex social behaviors, including interactions on social media. While prior work has focused on localized phenomena such as opinion formation and information spread, its potential to capture global network dynamics remains underexplored. This paper addresses this gap by analyzing GABM-based social media simulations through the lens of the Friendship Paradox (FP), a counterintuitive phenomenon where individuals, on average, have fewer friends than their friends. We propose a GABM framework for social media simulations, featuring generative agents that emulate real users with distinct personalities and interests. Using Twitter datasets on the US 2020 Election and the QAnon conspiracy, we show that the FP emerges naturally in GABM simulations. Consistent with real-world observations, the simulations unveil a hierarchical structure, where agents preferentially connect with others displaying higher activity or influence. Additionally, we find that infrequent connections primarily drive the FP, reflecting patterns in real networks. These findings validate GABM as a robust tool for modeling global social media phenomena and highlight its potential for advancing social science by enabling nuanced analysis of user behavior.
2502.05923
ARISE: Iterative Rule Induction and Synthetic Data Generation for Text Classification
cs.CL
We propose ARISE, a framework that iteratively induces rules and generates synthetic data for text classification. We combine synthetic data generation and automatic rule induction, via bootstrapping, to iteratively filter the generated rules and data. We induce rules via inductive generalisation of syntactic n-grams, enabling us to capture a complementary source of supervision. These rules alone lead to performance gains in both, in-context learning (ICL) and fine-tuning (FT) settings. Similarly, use of augmented data from ARISE alone improves the performance for a model, outperforming configurations that rely on complex methods like contrastive learning. Further, our extensive experiments on various datasets covering three full-shot, eight few-shot and seven multilingual variant settings demonstrate that the rules and data we generate lead to performance improvements across these diverse domains and languages.
2502.05924
Multi-Branch Collaborative Learning Network for Video Quality Assessment in Industrial Video Search
cs.CV cs.IR
Video Quality Assessment (VQA) is vital for large-scale video retrieval systems, aimed at identifying quality issues to prioritize high-quality videos. In industrial systems, low-quality video characteristics fall into four categories: visual-related issues like mosaics and black boxes, textual issues from video titles and OCR content, and semantic issues like frame incoherence and frame-text mismatch from AI-generated videos. Despite their prevalence in industrial settings, these low-quality videos have been largely overlooked in academic research, posing a challenge for accurate identification. To address this, we introduce the Multi-Branch Collaborative Network (MBCN) tailored for industrial video retrieval systems. MBCN features four branches, each designed to tackle one of the aforementioned quality issues. After each branch independently scores videos, we aggregate these scores using a weighted approach and a squeeze-and-excitation mechanism to dynamically address quality issues across different scenarios. We implement point-wise and pair-wise optimization objectives to ensure score stability and reasonableness. Extensive offline and online experiments on a world-level video search engine demonstrate MBCN's effectiveness in identifying video quality issues, significantly enhancing the retrieval system's ranking performance. Detailed experimental analyses confirm the positive contribution of all four evaluation branches. Furthermore, MBCN significantly improves recognition accuracy for low-quality AI-generated videos compared to the baseline.
2502.05925
Sign-Symmetry Learning Rules are Robust Fine-Tuners
cs.LG cs.AI
Backpropagation (BP) has long been the predominant method for training neural networks due to its effectiveness. However, numerous alternative approaches, broadly categorized under feedback alignment, have been proposed, many of which are motivated by the search for biologically plausible learning mechanisms. Despite their theoretical appeal, these methods have consistently underperformed compared to BP, leading to a decline in research interest. In this work, we revisit the role of such methods and explore how they can be integrated into standard neural network training pipelines. Specifically, we propose fine-tuning BP-pre-trained models using Sign-Symmetry learning rules and demonstrate that this approach not only maintains performance parity with BP but also enhances robustness. Through extensive experiments across multiple tasks and benchmarks, we establish the validity of our approach. Our findings introduce a novel perspective on neural network training and open new research directions for leveraging biologically inspired learning rules in deep learning.
2502.05926
A Generative Framework for Bidirectional Image-Report Understanding in Chest Radiography
eess.IV cs.CL cs.CV
The rapid advancements in large language models (LLMs) have unlocked their potential for multimodal tasks, where text and visual data are processed jointly. However, applying LLMs to medical imaging, particularly for chest X-rays (CXR), poses significant challenges due to the need for precise visual-textual alignment and the preservation of critical diagnostic details. In this paper, we propose Multi-Stage Adaptive Vision-Language Tuning (MAViLT), a novel framework designed to enhance multimodal reasoning and generation for CXR understanding. MAViLT incorporates a clinical gradient-weighted tokenization process and a hierarchical fine-tuning strategy, enabling it to generate accurate radiology reports, synthesize realistic CXRs from text, and answer vision-based clinical questions. We evaluate MAViLT on two benchmark datasets, MIMIC-CXR and Indiana University CXR, achieving state-of-the-art results across all tasks. Human evaluations further validate the clinical relevance and utility of MAViLT, making it a robust tool for real-world medical applications. This work demonstrates the feasibility of leveraging LLMs for multimodal medical imaging while addressing key challenges in vision-language integration.
2502.05928
ClinKD: Cross-Modal Clinic Knowledge Distiller For Multi-Task Medical Images
cs.CV
Med-VQA (Medical Visual Question Answering) is a crucial subtask within the broader VQA (Visual Question Answering) domain. This task requires a visual question answering system to analyze the provided image and corresponding question,offering reasonable analysis and suggestions to assist medical professionals in making pathological diagnoses, or ideally, enabling the system to independently provide correct diagnoses. Furthermore, more advanced Med-VQA tasks involve Referring and Grounding, which not only require the system to accurately comprehend medical images but also to pinpoint specific biological locations within those images. While many large pre-trained models have demonstrated substantial VQA capabilities,challenges persist in the medical imaging domain. The intricacy of biological features in medical images and the scarcity of high-quality medical image datasets, combined with the fact that current models are not tailored for the medical field in terms of architecture and training paradigms, hinder the full exploitation of model generalization. This results in issues such as hallucination in Visual Grounding. In this paper, we introduce the ClinKD model, which incorporates modifications to model position encoding and a diversified training process. Initially, we enhance the model's ability to perceive image and modality variations by using Med-CLIP Guided Rotary Position Embedding. Subsequently, we leverage distillation to provide prior knowledge to the model before using complete training data. Additionally, the feedback-based training process during the formal training phase further enhances data utilization. Notably, under unchanged evaluation protocols, we achieve a new state-of-the-art performance on the Med-GRIT-270k dataset, and the Med-CLIP Guided Rotary Position Embedding approach presents potential for generalizing to universal model position encoding.
2502.05931
Protecting Intellectual Property of EEG-based Neural Networks with Watermarking
cs.LG cs.AI cs.CR
EEG-based neural networks, pivotal in medical diagnosis and brain-computer interfaces, face significant intellectual property (IP) risks due to their reliance on sensitive neurophysiological data and resource-intensive development. Current watermarking methods, particularly those using abstract trigger sets, lack robust authentication and fail to address the unique challenges of EEG models. This paper introduces a cryptographic wonder filter-based watermarking framework tailored for EEG-based neural networks. Leveraging collision-resistant hashing and public-key encryption, the wonder filter embeds the watermark during training, ensuring minimal distortion ($\leq 5\%$ drop in EEG task accuracy) and high reliability (100\% watermark detection). The framework is rigorously evaluated against adversarial attacks, including fine-tuning, transfer learning, and neuron pruning. Results demonstrate persistent watermark retention, with classification accuracy for watermarked states remaining above 90\% even after aggressive pruning, while primary task performance degrades faster, deterring removal attempts. Piracy resistance is validated by the inability to embed secondary watermarks without severe accuracy loss ( $>10\%$ in EEGNet and CCNN models). Cryptographic hashing ensures authentication, reducing brute-force attack success probabilities. Evaluated on the DEAP dataset across models (CCNN, EEGNet, TSception), the method achieves $>99.4\%$ null-embedding accuracy, effectively eliminating false positives. By integrating wonder filters with EEG-specific adaptations, this work bridges a critical gap in IP protection for neurophysiological models, offering a secure, tamper-proof solution for healthcare and biometric applications. The framework's robustness against adversarial modifications underscores its potential to safeguard sensitive EEG models while maintaining diagnostic utility.
2502.05932
Skill Expansion and Composition in Parameter Space
cs.LG cs.AI cs.RO
Humans excel at reusing prior knowledge to address new challenges and developing skills while solving problems. This paradigm becomes increasingly popular in the development of autonomous agents, as it develops systems that can self-evolve in response to new challenges like human beings. However, previous methods suffer from limited training efficiency when expanding new skills and fail to fully leverage prior knowledge to facilitate new task learning. In this paper, we propose Parametric Skill Expansion and Composition (PSEC), a new framework designed to iteratively evolve the agents' capabilities and efficiently address new challenges by maintaining a manageable skill library. This library can progressively integrate skill primitives as plug-and-play Low-Rank Adaptation (LoRA) modules in parameter-efficient finetuning, facilitating efficient and flexible skill expansion. This structure also enables the direct skill compositions in parameter space by merging LoRA modules that encode different skills, leveraging shared information across skills to effectively program new skills. Based on this, we propose a context-aware module to dynamically activate different skills to collaboratively handle new tasks. Empowering diverse applications including multi-objective composition, dynamics shift, and continual policy shift, the results on D4RL, DSRL benchmarks, and the DeepMind Control Suite show that PSEC exhibits superior capacity to leverage prior knowledge to efficiently tackle new challenges, as well as expand its skill libraries to evolve the capabilities. Project website: https://ltlhuuu.github.io/PSEC/.
2502.05933
Learning to Substitute Words with Model-based Score Ranking
cs.CL cs.AI
Smart word substitution aims to enhance sentence quality by improving word choices; however current benchmarks rely on human-labeled data. Since word choices are inherently subjective, ground-truth word substitutions generated by a small group of annotators are often incomplete and likely not generalizable. To circumvent this issue, we instead employ a model-based score (BARTScore) to quantify sentence quality, thus forgoing the need for human annotations. Specifically, we use this score to define a distribution for each word substitution, allowing one to test whether a substitution is statistically superior relative to others. In addition, we propose a loss function that directly optimizes the alignment between model predictions and sentence scores, while also enhancing the overall quality score of a substitution. Crucially, model learning no longer requires human labels, thus avoiding the cost of annotation while maintaining the quality of the text modified with substitutions. Experimental results show that the proposed approach outperforms both masked language models (BERT, BART) and large language models (GPT-4, LLaMA). The source code is available at https://github.com/Hyfred/Substitute-Words-with-Ranking.
2502.05934
Barriers and Pathways to Human-AI Alignment: A Game-Theoretic Approach
cs.AI cs.CC cs.GT cs.LG cs.MA
Under what conditions can capable AI agents efficiently align their actions with human preferences? More specifically, when they are proficient enough to collaborate with us, how long does coordination take, and when is it computationally feasible? These foundational questions of AI alignment help define what makes an AI agent ``sufficiently safe'' and valuable to humans. Since such generally capable systems do not yet exist, a theoretical analysis is needed to establish when guarantees hold -- and what they even are. We introduce a game-theoretic framework that generalizes prior alignment approaches with fewer assumptions, allowing us to analyze the computational complexity of alignment across $M$ objectives and $N$ agents, providing both upper and lower bounds. Unlike previous work, which often assumes common priors, idealized communication, or implicit tractability, our framework formally characterizes the difficulty of alignment under minimal assumptions. Our main result shows that even when agents are fully rational and computationally \emph{unbounded}, alignment can be achieved with high probability in time \emph{linear} in the task space size. Therefore, in real-world settings, where task spaces are often \emph{exponential} in input length, this remains impractical. More strikingly, our lower bound demonstrates that alignment is \emph{impossible} to speed up when scaling to exponentially many tasks or agents, highlighting a fundamental computational barrier to scalable alignment. Relaxing these idealized assumptions, we study \emph{computationally bounded} agents with noisy messages (representing obfuscated intent), showing that while alignment can still succeed with high probability, it incurs additional \emph{exponential} slowdowns in the task space size, number of agents, and number of tasks. We conclude by identifying conditions that make alignment more feasible.
2502.05935
Interactive Inference: A Neuromorphic Theory of Human-Computer Interaction
cs.HC cs.IT math.IT
Neuromorphic HCI is a new theoretical approach to designing better UX inspired by the neurophysiology of the brain. Here, we apply the neuroscientific theory of Active Inference to HCI, postulating that users perform Bayesian inference on progress and goal distributions to predict their next action (Interactive Inference). We show how Bayesian surprise between goal and progress distributions follows a mean square error function of the signal-to-noise ratio (SNR) of the task. However, capacity to process Bayesian surprise follows the logarithm of SNR, and errors occur when average capacity is exceeded. Our model allows the quantitative analysis of performance and error in one framework with real-time estimation of mental load. We show through mathematical theorems how three basic laws of HCI, Hick's Law, Fitts' Law and the Power Law fit our model. We then test the validity of the general model by empirically measuring how well it predicts human performance in a car following task. Results suggest that driver processing capacity indeed is a logarithmic function of the SNR of the distance to a lead car. This positive result provides initial evidence that Interactive Interference can work as a new theoretical underpinning for HCI, deserving further exploration.
2502.05937
A Semi-Supervised Text Generation Framework Combining a Deep Transformer and a GAN
cs.CL cs.AI
This paper introduces a framework that connects a deep generative pre-trained Transformer language model with a generative adversarial network for semi-supervised text generation. In other words, the proposed model is first pre-trained unsupervised on a large and diverse text corpus with 24 layers. Then a simple GAN architecture for synthetic text generation is introduced, and Gumbel-Softmax is applied to handle the discreteness of tokens. The paper also shows a semi-supervised approach where real data is augmented with GAN samples, which is further used to fine-tune the Transformer model on the merged dataset. Detailed theoretical derivations are also included, outlining the proof of the min-max objective function, and an extensive discussion of the Gumbel-Softmax reparameterization trick.
2502.05938
Energy-Efficient Autonomous Aerial Navigation with Dynamic Vision Sensors: A Physics-Guided Neuromorphic Approach
cs.RO
Vision-based object tracking is a critical component for achieving autonomous aerial navigation, particularly for obstacle avoidance. Neuromorphic Dynamic Vision Sensors (DVS) or event cameras, inspired by biological vision, offer a promising alternative to conventional frame-based cameras. These cameras can detect changes in intensity asynchronously, even in challenging lighting conditions, with a high dynamic range and resistance to motion blur. Spiking neural networks (SNNs) are increasingly used to process these event-based signals efficiently and asynchronously. Meanwhile, physics-based artificial intelligence (AI) provides a means to incorporate system-level knowledge into neural networks via physical modeling. This enhances robustness, energy efficiency, and provides symbolic explainability. In this work, we present a neuromorphic navigation framework for autonomous drone navigation. The focus is on detecting and navigating through moving gates while avoiding collisions. We use event cameras for detecting moving objects through a shallow SNN architecture in an unsupervised manner. This is combined with a lightweight energy-aware physics-guided neural network (PgNN) trained with depth inputs to predict optimal flight times, generating near-minimum energy paths. The system is implemented in the Gazebo simulator and integrates a sensor-fused vision-to-planning neuro-symbolic framework built with the Robot Operating System (ROS) middleware. This work highlights the future potential of integrating event-based vision with physics-guided planning for energy-efficient autonomous navigation, particularly for low-latency decision-making.
2502.05943
Continual Adaptation for Autonomous Driving with the Mixture of Progressive Experts Network
cs.RO
Learning-based autonomous driving requires continuous integration of diverse knowledge in complex traffic , yet existing methods exhibit significant limitations in adaptive capabilities. Addressing this gap demands autonomous driving systems that enable continual adaptation through dynamic adjustments to evolving environmental interactions. This underscores the necessity for enhanced continual learning capabilities to improve system adaptability. To address these challenges, the paper introduces a dynamic progressive optimization framework that facilitates adaptation to variations in dynamic environments, achieved by integrating reinforcement learning and supervised learning for data aggregation. Building on this framework, we propose the Mixture of Progressive Experts (MoPE) network. The proposed method selectively activates multiple expert models based on the distinct characteristics of each task and progressively refines the network architecture to facilitate adaptation to new tasks. Simulation results show that the MoPE model outperforms behavior cloning methods, achieving up to a 7.8% performance improvement in intricate urban road environments.
2502.05944
Multi-granular Training Strategies for Robust Multi-hop Reasoning Over Noisy and Heterogeneous Knowledge Sources
cs.CL
Multi-source multi-hop question answering (QA) represents a challenging task in natural language processing due to the need for dynamic integration of heterogeneous knowledge sources and multi-step reasoning. Existing methods often suffer from cascading errors, insufficient handling of knowledge conflicts, and computational inefficiency. In this paper, we propose Adaptive Multi-source Knowledge-Oriented Reasoning (AMKOR), a generative framework that leverages large language models (LLMs) to dynamically fuse parametric and retrieved knowledge while exploring reasoning trajectories using probabilistic beam reasoning. AMKOR is further enhanced by a multi-granular learning strategy, optimizing both local reasoning steps and global answer accuracy. Experiments conducted on four widely-used multi-hop QA datasets, including HotpotQA and MuSiQue, demonstrate that AMKOR achieves state-of-the-art performance, significantly outperforming baseline methods on both reasoning accuracy and robustness. Additional analyses confirm its scalability, adaptability to noisy knowledge, and superior ability to handle complex multi-hop tasks. This work establishes a new benchmark for multi-source multi-hop QA by effectively combining reasoning quality and efficiency.
2502.05945
"Let the AI conspiracy begin..." Language Model coordination is just one inference-intervention away
cs.CL cs.AI
In this work, we introduce a straightforward and effective methodology to steer large language model behaviour capable of bypassing learned alignment goals. We employ interference-time activation shifting, which is effective without additional training. Following prior studies, we derive intervention directions from activation differences in contrastive pairs of model outputs, which represent the desired and undesired behaviour. By prompting the model to include multiple-choice answers in its response, we can automatically evaluate the sensitivity of model output to individual attention heads steering efforts. We demonstrate that interventions on these heads generalize well to open-ended answer generation in the challenging "AI coordination" dataset. In this dataset, models must choose between assisting another AI or adhering to ethical, safe, and unharmful behaviour. Our fine-grained interventions lead Llama 2 to prefer coordination with other AIs over following established alignment goals. Additionally, this approach enables stronger interventions than those applied to whole model layers, preserving the overall cohesiveness of the output. The simplicity of our method highlights the shortcomings of current alignment strategies and points to potential future research directions, as concepts like "AI coordination" can be influenced by selected attention heads.
2502.05947
Acceleration Multiple Heads Decoding for LLM via Dynamic Tree Attention
cs.CV cs.CL
Multiple heads decoding accelerates the inference of Large Language Models (LLMs) by predicting next several tokens simultaneously. It generates and verifies multiple candidate sequences in parallel via tree attention with a fixed structure. In this paper, we replace the fixed tree attention with dynamic tree attention on multiple head decoding, specifically in the context of MEDUSA. We propose a simple and low complexity strategy to generate candidates and construct the dynamic tree structure. Preliminary experiments show that the proposed method improves the decoding efficiency of multiple head decoding for LLMs while maintaining the generation quality. This result demonstrates the potential for improvement of multiple head decoding in candidate generation.
2502.05949
Verifying Proportionality in Temporal Voting
cs.GT cs.AI
We study a model of temporal voting where there is a fixed time horizon, and at each round the voters report their preferences over the available candidates and a single candidate is selected. Prior work has adapted popular notions of justified representation as well as voting rules that provide strong representation guarantees from the multiwinner election setting to this model. In our work, we focus on the complexity of verifying whether a given outcome offers proportional representation. We show that in the temporal setting verification is strictly harder than in multiwinner voting, but identify natural special cases that enable efficient algorithms.
2502.05950
Survival Concept-Based Learning Models
cs.LG cs.AI stat.ML
Concept-based learning enhances prediction accuracy and interpretability by leveraging high-level, human-understandable concepts. However, existing CBL frameworks do not address survival analysis tasks, which involve predicting event times in the presence of censored data -- a common scenario in fields like medicine and reliability analysis. To bridge this gap, we propose two novel models: SurvCBM (Survival Concept-based Bottleneck Model) and SurvRCM (Survival Regularized Concept-based Model), which integrate concept-based learning with survival analysis to handle censored event time data. The models employ the Cox proportional hazards model and the Beran estimator. SurvCBM is based on the architecture of the well-known concept bottleneck model, offering interpretable predictions through concept-based explanations. SurvRCM uses concepts as regularization to enhance accuracy. Both models are trained end-to-end and provide interpretable predictions in terms of concepts. Two interpretability approaches are proposed: one leveraging the linear relationship in the Cox model and another using an instance-based explanation framework with the Beran estimator. Numerical experiments demonstrate that SurvCBM outperforms SurvRCM and traditional survival models, underscoring the importance and advantages of incorporating concept information. The code for the proposed algorithms is publicly available.
2502.05951
Cyri: A Conversational AI-based Assistant for Supporting the Human User in Detecting and Responding to Phishing Attacks
cs.HC cs.AI cs.CR
This work introduces Cyri, an AI-powered conversational assistant designed to support a human user in detecting and analyzing phishing emails by leveraging Large Language Models. Cyri has been designed to scrutinize emails for semantic features used in phishing attacks, such as urgency, and undesirable consequences, using an approach that unifies features already established in the literature with others by Cyri features extraction methodology. Cyri can be directly plugged into a client mail or webmail, ensuring seamless integration with the user's email workflow while maintaining data privacy through local processing. By performing analyses on the user's machine, Cyri eliminates the need to transmit sensitive email data over the internet, reducing associated security risks. The Cyri user interface has been designed to reduce habituation effects and enhance user engagement. It employs dynamic visual cues and context-specific explanations to keep users alert and informed while using emails. Additionally, it allows users to explore identified malicious semantic features both through conversation with the agent and visual exploration, obtaining the advantages of both modalities for expert or non-expert users. It also allows users to keep track of the conversation, supports the user in solving additional questions on both computed features or new parts of the mail, and applies its detection on demand. To evaluate Cyri, we crafted a comprehensive dataset of 420 phishing emails and 420 legitimate emails. Results demonstrate high effectiveness in identifying critical phishing semantic features fundamental to phishing detection. A user study involving 10 participants, both experts and non-experts, evaluated Cyri's effectiveness and usability. Results indicated that Cyri significantly aided users in identifying phishing emails and enhanced their understanding of phishing tactics.
2502.05954
Optimization under Attack: Resilience, Vulnerability, and the Path to Collapse
cs.MA
Optimization is instrumental for improving operations of large-scale socio-technical infrastructures of Smart Cities, for instance, energy and traffic systems. In particular, understanding the performance of multi-agent discrete-choice combinatorial optimization under distributed adversary attacks is a compelling and underexplored problem, since multi-agent systems exhibit a large number of remote control variables that can influence in an unprecedented way the cost-effectiveness of distributed optimization heuristics. This paper unravels for the first time the trajectories of distributed optimization from resilience to vulnerability, and finally to collapse under varying adversary influence. Using real-world data to emulate over 28 billion multi-agent optimization scenarios, we exhaustively assess how the number of agents with different adversarial severity and network positioning influences optimization performance, including the influence on Pareto optimal points. With this novel large-scale dataset, made openly available as a benchmark, we disentangle how optimization remains resilient to adversaries and which adversary conditions are required to make optimization vulnerable or collapsed. These new findings can provide new insights for designing self-healing strategies for fault-tolerance and fault-correction in adversarial distributed optimization that have been missing so far.
2502.05957
AutoAgent: A Fully-Automated and Zero-Code Framework for LLM Agents
cs.AI cs.CL
Large Language Model (LLM) Agents have demonstrated remarkable capabilities in task automation and intelligent decision-making, driving the widespread adoption of agent development frameworks such as LangChain and AutoGen. However, these frameworks predominantly serve developers with extensive technical expertise - a significant limitation considering that only 0.03 % of the global population possesses the necessary programming skills. This stark accessibility gap raises a fundamental question: Can we enable everyone, regardless of technical background, to build their own LLM agents using natural language alone? To address this challenge, we introduce AutoAgent-a Fully-Automated and highly Self-Developing framework that enables users to create and deploy LLM agents through Natural Language Alone. Operating as an autonomous Agent Operating System, AutoAgent comprises four key components: i) Agentic System Utilities, ii) LLM-powered Actionable Engine, iii) Self-Managing File System, and iv) Self-Play Agent Customization module. This lightweight yet powerful system enables efficient and dynamic creation and modification of tools, agents, and workflows without coding requirements or manual intervention. Beyond its code-free agent development capabilities, AutoAgent also serves as a versatile multi-agent system for General AI Assistants. Comprehensive evaluations on the GAIA benchmark demonstrate AutoAgent's effectiveness in generalist multi-agent tasks, surpassing existing state-of-the-art methods. Furthermore, AutoAgent's Retrieval-Augmented Generation (RAG)-related capabilities have shown consistently superior performance compared to many alternative LLM-based solutions.
2502.05959
Ensemble-Tight Second-Order Asymptotics and Exponents for Guessing-Based Decoding with Abandonment
cs.IT math.IT
This paper considers guessing-based decoders with abandonment for discrete memoryless channels in which all codewords have the same composition. This class of decoders rank-orders all input sequences in the codebook's composition class from ``closest'' to ``farthest'' from the channel output and then queries them sequentially in that order for codebook membership. Decoding terminates when a codeword is encountered or when a predetermined number of guesses is reached, and decoding is abandoned. We derive ensemble-tight first-order asymptotics for the code rate and abandonment rate, which shows that guessing-based decoding is more efficient than conventional testing-based decoding whenever the capacity of the channel exceeds half the entropy of the capacity-achieving input distribution. The main focus of this paper is on refined asymptotics, specifically, second-order asymptotics, error exponents, and strong converse exponents. The optimal second-order region is characterized in terms of the minimum of the second-order code and abandonment rates. The error (resp.\ strong converse) exponent is characterized in terms of the minimum (resp.\ maximum) of the usual channel coding exponent and an abandonment exponent, which turns out to be a special case of the exponent of conditional almost-lossless source coding.
2502.05963
Redefining Robot Generalization Through Interactive Intelligence
cs.LG cs.AI cs.RO
Recent advances in large-scale machine learning have produced high-capacity foundation models capable of adapting to a broad array of downstream tasks. While such models hold great promise for robotics, the prevailing paradigm still portrays robots as single, autonomous decision-makers, performing tasks like manipulation and navigation, with limited human involvement. However, a large class of real-world robotic systems, including wearable robotics (e.g., prostheses, orthoses, exoskeletons), teleoperation, and neural interfaces, are semiautonomous, and require ongoing interactive coordination with human partners, challenging single-agent assumptions. In this position paper, we argue that robot foundation models must evolve to an interactive multi-agent perspective in order to handle the complexities of real-time human-robot co-adaptation. We propose a generalizable, neuroscience-inspired architecture encompassing four modules: (1) a multimodal sensing module informed by sensorimotor integration principles, (2) an ad-hoc teamwork model reminiscent of joint-action frameworks in cognitive science, (3) a predictive world belief model grounded in internal model theories of motor control, and (4) a memory/feedback mechanism that echoes concepts of Hebbian and reinforcement-based plasticity. Although illustrated through the lens of cyborg systems, where wearable devices and human physiology are inseparably intertwined, the proposed framework is broadly applicable to robots operating in semi-autonomous or interactive contexts. By moving beyond single-agent designs, our position emphasizes how foundation models in robotics can achieve a more robust, personalized, and anticipatory level of performance.
2502.05964
Revisiting Gradient-based Uncertainty for Monocular Depth Estimation
cs.CV
Monocular depth estimation, similar to other image-based tasks, is prone to erroneous predictions due to ambiguities in the image, for example, caused by dynamic objects or shadows. For this reason, pixel-wise uncertainty assessment is required for safety-critical applications to highlight the areas where the prediction is unreliable. We address this in a post hoc manner and introduce gradient-based uncertainty estimation for already trained depth estimation models. To extract gradients without depending on the ground truth depth, we introduce an auxiliary loss function based on the consistency of the predicted depth and a reference depth. The reference depth, which acts as pseudo ground truth, is in fact generated using a simple image or feature augmentation, making our approach simple and effective. To obtain the final uncertainty score, the derivatives w.r.t. the feature maps from single or multiple layers are calculated using back-propagation. We demonstrate that our gradient-based approach is effective in determining the uncertainty without re-training using the two standard depth estimation benchmarks KITTI and NYU. In particular, for models trained with monocular sequences and therefore most prone to uncertainty, our method outperforms related approaches. In addition, we publicly provide our code and models: https://github.com/jhornauer/GrUMoDepth
2502.05966
Detection of Physiological Data Tampering Attacks with Quantum Machine Learning
quant-ph cs.LG
The widespread use of cloud-based medical devices and wearable sensors has made physiological data susceptible to tampering. These attacks can compromise the reliability of healthcare systems which can be critical and life-threatening. Detection of such data tampering is of immediate need. Machine learning has been used to detect anomalies in datasets but the performance of Quantum Machine Learning (QML) is still yet to be evaluated for physiological sensor data. Thus, our study compares the effectiveness of QML for detecting physiological data tampering, focusing on two types of white-box attacks: data poisoning and adversarial perturbation. The results show that QML models are better at identifying label-flipping attacks, achieving accuracy rates of 75%-95% depending on the data and attack severity. This superior performance is due to the ability of quantum algorithms to handle complex and high-dimensional data. However, both QML and classical models struggle to detect more sophisticated adversarial perturbation attacks, which subtly alter data without changing its statistical properties. Although QML performed poorly against this attack with around 45%-65% accuracy, it still outperformed classical algorithms in some cases.
2502.05967
$\mu$nit Scaling: Simple and Scalable FP8 LLM Training
cs.LG
Large Language Model training with 8-bit floating point (FP8) formats promises significant efficiency improvements, but reduced numerical precision makes training challenging. It is currently possible to train in FP8 only if one is willing to tune various hyperparameters, reduce model scale, or accept the overhead of computing dynamic scale factors. We demonstrate simple, scalable FP8 training that requires no dynamic scaling factors or special hyperparameters, even at large model sizes. Our method, $\mu$nit Scaling ($\mu$S), also enables simple hyperparameter transfer across model widths, matched numerics across training and inference, and other desirable properties. $\mu$nit Scaling is straightforward to implement, consisting of a set of minimal interventions based on a first-principles analysis of common transformer operations. We validate our method by training models from 1B to 13B parameters, performing all hidden linear layer computations in FP8. We achieve quality equal to higher precision baselines while also training up to 33% faster.
2502.05969
Asymptotic FDR Control with Model-X Knockoffs: Is Moments Matching Sufficient?
stat.ML cs.LG math.ST stat.TH
We propose a unified theoretical framework for studying the robustness of the model-X knockoffs framework by investigating the asymptotic false discovery rate (FDR) control of the practically implemented approximate knockoffs procedure. This procedure deviates from the model-X knockoffs framework by substituting the true covariate distribution with a user-specified distribution that can be learned using in-sample observations. By replacing the distributional exchangeability condition of the model-X knockoff variables with three conditions on the approximate knockoff statistics, we establish that the approximate knockoffs procedure achieves the asymptotic FDR control. Using our unified framework, we further prove that an arguably most popularly used knockoff variable generation method--the Gaussian knockoffs generator based on the first two moments matching--achieves the asymptotic FDR control when the two-moment-based knockoff statistics are employed in the knockoffs inference procedure. For the first time in the literature, our theoretical results justify formally the effectiveness and robustness of the Gaussian knockoffs generator. Simulation and real data examples are conducted to validate the theoretical findings.
2502.05970
Known Unknowns: Out-of-Distribution Property Prediction in Materials and Molecules
cs.LG cond-mat.mtrl-sci cs.CE physics.chem-ph
Discovery of high-performance materials and molecules requires identifying extremes with property values that fall outside the known distribution. Therefore, the ability to extrapolate to out-of-distribution (OOD) property values is critical for both solid-state materials and molecular design. Our objective is to train predictor models that extrapolate zero-shot to higher ranges than in the training data, given the chemical compositions of solids or molecular graphs and their property values. We propose using a transductive approach to OOD property prediction, achieving improvements in prediction accuracy. In particular, the True Positive Rate (TPR) of OOD classification of materials and molecules improved by 3x and 2.5x, respectively, and precision improved by 2x and 1.5x compared to non-transductive baselines. Our method leverages analogical input-target relations in the training and test sets, enabling generalization beyond the training target support, and can be applied to any other material and molecular tasks.
2502.05972
Mechanic Modeling and Nonlinear Optimal Control of Actively Articulated Suspension of Mobile Heavy-Duty Manipulators
cs.RO
This paper presents the analytic modeling of mobile heavy-duty manipulators with actively articulated suspension and its optimal control to maximize its static and dynamic stabilization. By adopting the screw theory formalism, we consider the suspension mechanism as a rigid multibody composed of two closed kinematic chains. This mechanical modeling allows us to compute the spatial inertial parameters of the whole platform as a function of the suspension's linear actuators through the articulated-body inertia method. Our solution enhances the computation accuracy of the wheels' reaction normal forces by providing an exact solution for the center of mass and inertia tensor of the mobile manipulator. Moreover, these inertial parameters and the normal forces are used to define metrics of both static and dynamic stability of the mobile manipulator and formulate a nonlinear programming problem that optimizes such metrics to generate an optimal stability motion that prevents the platform's overturning, such optimal position of the actuator is tracked with a state-feedback hydraulic valve control. We demonstrate our method's efficiency in terms of C++ computational speed, accuracy and performance improvement by simulating a 7 degrees-of-freedom heavy-duty parallel-serial mobile manipulator with four wheels and actively articulated suspension.
2502.05974
Decision Making in Hybrid Environments: A Model Aggregation Approach
cs.LG stat.ML
Recent work by Foster et al. (2021, 2022, 2023) and Xu and Zeevi (2023) developed the framework of decision estimation coefficient (DEC) that characterizes the complexity of general online decision making problems and provides a general algorithm design principle. These works, however, either focus on the pure stochastic regime where the world remains fixed over time, or the pure adversarial regime where the world arbitrarily changes over time. For the hybrid regime where the dynamics of the world is fixed while the reward arbitrarily changes, they only give pessimistic bounds on the decision complexity. In this work, we propose a general extension of DEC that more precisely characterizes this case. Besides applications in special cases, our framework leads to a flexible algorithm design where the learner learns over subsets of the hypothesis set, trading estimation complexity with decision complexity, which could be of independent interest. Our work covers model-based learning and model-free learning in the hybrid regime, with a newly proposed extension of the bilinear classes (Du et al., 2021) to the adversarial-reward case. We also recover some existing model-free learning results in the pure stochastic regime.
2502.05979
VFX Creator: Animated Visual Effect Generation with Controllable Diffusion Transformer
cs.CV
Crafting magic and illusions is one of the most thrilling aspects of filmmaking, with visual effects (VFX) serving as the powerhouse behind unforgettable cinematic experiences. While recent advances in generative artificial intelligence have driven progress in generic image and video synthesis, the domain of controllable VFX generation remains relatively underexplored. In this work, we propose a novel paradigm for animated VFX generation as image animation, where dynamic effects are generated from user-friendly textual descriptions and static reference images. Our work makes two primary contributions: (i) Open-VFX, the first high-quality VFX video dataset spanning 15 diverse effect categories, annotated with textual descriptions, instance segmentation masks for spatial conditioning, and start-end timestamps for temporal control. (ii) VFX Creator, a simple yet effective controllable VFX generation framework based on a Video Diffusion Transformer. The model incorporates a spatial and temporal controllable LoRA adapter, requiring minimal training videos. Specifically, a plug-and-play mask control module enables instance-level spatial manipulation, while tokenized start-end motion timestamps embedded in the diffusion process, alongside the text encoder, allow precise temporal control over effect timing and pace. Extensive experiments on the Open-VFX test set demonstrate the superiority of the proposed system in generating realistic and dynamic effects, achieving state-of-the-art performance and generalization ability in both spatial and temporal controllability. Furthermore, we introduce a specialized metric to evaluate the precision of temporal control. By bridging traditional VFX techniques with generative approaches, VFX Creator unlocks new possibilities for efficient and high-quality video effect generation, making advanced VFX accessible to a broader audience.
2502.05980
Speech to Speech Translation with Translatotron: A State of the Art Review
cs.CL cs.AI
A cascade-based speech-to-speech translation has been considered a benchmark for a very long time, but it is plagued by many issues, like the time taken to translate a speech from one language to another and compound errors. These issues are because a cascade-based method uses a combination of methods such as speech recognition, speech-to-text translation, and finally, text-to-speech translation. Translatotron, a sequence-to-sequence direct speech-to-speech translation model was designed by Google to address the issues of compound errors associated with cascade model. Today there are 3 versions of the Translatotron model: Translatotron 1, Translatotron 2, and Translatotron3. The first version was designed as a proof of concept to show that a direct speech-to-speech translation was possible, it was found to be less effective than the cascade model but was producing promising results. Translatotron2 was an improved version of Translatotron 1 with results similar to the cascade model. Translatotron 3 the latest version of the model is better than the cascade model at some points. In this paper, a complete review of speech-to-speech translation will be presented, with a particular focus on all the versions of Translatotron models. We will also show that Translatotron is the best model to bridge the language gap between African Languages and other well-formalized languages.
2502.05982
HamRaz: A Culture-Based Persian Conversation Dataset for Person-Centered Therapy Using LLM Agents
cs.CL
This paper presents HamRaz, a novel Persian-language mental health dataset designed for Person-Centered Therapy (PCT) using Large Language Models (LLMs). Despite the growing application of LLMs in AI-driven psychological counseling, existing datasets predominantly focus on Western and East Asian contexts, overlooking cultural and linguistic nuances essential for effective Persian-language therapy. To address this gap, HamRaz combines script-based dialogues with adaptive LLM role-playing, ensuring coherent and dynamic therapy interactions. We also introduce HamRazEval, a dual evaluation framework that measures conversational quality and therapeutic effectiveness using General Dialogue Metrics and the Barrett-Lennard Relationship Inventory (BLRI). Experimental results show HamRaz outperforms conventional Script Mode and Two-Agent Mode, producing more empathetic, context-aware, and realistic therapy sessions. By releasing HamRaz, we contribute a culturally adapted, LLM-driven resource to advance AI-powered psychotherapy research in diverse communities.
2502.05986
Preventing Rogue Agents Improves Multi-Agent Collaboration
cs.CL cs.MA
Multi-agent systems, where specialized agents collaborate to solve a shared task hold great potential, from increased modularity to simulating complex environments. However, they also have a major caveat -- a single agent can cause the entire system to fail. Consider a simple game where the knowledge to solve the task is distributed between agents, which share information in a communication channel. At each round, any of the agents can terminate the game and make the final prediction, even if they are uncertain about the outcome of their action. Detection of such rogue agents $\textit{before they act}$ may prevent the system's failure. In this work, we propose to $\textit{monitor}$ agents during action prediction and $\textit{intervene}$ when a future error is likely to occur. To test our approach, we introduce WhoDunitEnv, a multi-agent collaboration environment that allows modular control over task complexity and communication structure. Experiments on two variants of WhoDunitEnv and the GovSim environment for resource sustainability show that our approach leads to substantial performance gains up to 17.4% and 20%, respectively. Moreover, a thorough analysis shows that our monitors successfully identify critical points of agent confusion and our interventions effectively stop agent errors from propagating.
2502.05988
SNAT-YOLO: Efficient Cross-Layer Aggregation Network for Edge-Oriented Gangue Detection
cs.CV
To address the issues of slow detection speed,low accuracy,difficulty in deployment on industrial edge devices,and large parameter and computational requirements in deep learning-based coal gangue target detection methods,we propose a lightweight coal gangue target detection algorithm based on an improved YOLOv11.First,we use the lightweight network ShuffleNetV2 as the backbone to enhance detection speed.Second,we introduce a lightweight downsampling operation,ADown,which reduces model complexity while improving average detection accuracy.Third,we improve the C2PSA module in YOLOv11 by incorporating the Triplet Attention mechanism,resulting in the proposed C2PSA-TriAtt module,which enhances the model's ability to focus on different dimensions of images.Fourth,we propose the Inner-FocalerIoU loss function to replace the existing CIoU loss function.Experimental results show that our model achieves a detection accuracy of 99.10% in coal gangue detection tasks,reduces the model size by 38%,the number of parameters by 41%,and the computational cost by 40%,while decreasing the average detection time per image by 1 ms.The improved model demonstrates enhanced detection speed and accuracy,making it suitable for deployment on industrial edge mobile devices,thus contributing positively to coal processing and efficient utilization of coal resources.
2502.05994
Diffusion Models for Inverse Problems in the Exponential Family
stat.ML cs.LG
Diffusion models have emerged as powerful tools for solving inverse problems, yet prior work has primarily focused on observations with Gaussian measurement noise, restricting their use in real-world scenarios. This limitation persists due to the intractability of the likelihood score, which until now has only been approximated in the simpler case of Gaussian likelihoods. In this work, we extend diffusion models to handle inverse problems where the observations follow a distribution from the exponential family, such as a Poisson or a Binomial distribution. By leveraging the conjugacy properties of exponential family distributions, we introduce the evidence trick, a method that provides a tractable approximation to the likelihood score. In our experiments, we demonstrate that our methodology effectively performs Bayesian inference on spatially inhomogeneous Poisson processes with intensities as intricate as ImageNet images. Furthermore, we demonstrate the real-world impact of our methodology by showing that it performs competitively with the current state-of-the-art in predicting malaria prevalence estimates in Sub-Saharan Africa.
2502.05995
A Comprehensive Survey on Image Signal Processing Approaches for Low-Illumination Image Enhancement
cs.CV
The usage of digital content (photos and videos) in a variety of applications has increased due to the popularity of multimedia devices. These uses include advertising campaigns, educational resources, and social networking platforms. There is an increasing need for high-quality graphic information as people become more visually focused. However, captured images frequently have poor visibility and a high amount of noise due to the limitations of image-capturing devices and lighting conditions. Improving the visual quality of images taken in low illumination is the aim of low-illumination image enhancement. This problem is addressed by traditional image enhancement techniques, which alter noise, brightness, and contrast. Deep learning-based methods, however, have dominated recently made advances in this area. These methods have effectively reduced noise while preserving important information, showing promising results in the improvement of low-illumination images. An extensive summary of image signal processing methods for enhancing low-illumination images is provided in this paper. Three categories are classified in the review for approaches: hybrid techniques, deep learning-based methods, and traditional approaches. Conventional techniques include denoising, automated white balancing, and noise reduction. Convolutional neural networks (CNNs) are used in deep learningbased techniques to recognize and extract characteristics from low-light images. To get better results, hybrid approaches combine deep learning-based methodologies with more conventional methods. The review also discusses the advantages and limitations of each approach and provides insights into future research directions in this field.
2502.05996
Motion Control in Multi-Rotor Aerial Robots Using Deep Reinforcement Learning
cs.RO cs.AI
This paper investigates the application of Deep Reinforcement (DRL) Learning to address motion control challenges in drones for additive manufacturing (AM). Drone-based additive manufacturing promises flexible and autonomous material deposition in large-scale or hazardous environments. However, achieving robust real-time control of a multi-rotor aerial robot under varying payloads and potential disturbances remains challenging. Traditional controllers like PID often require frequent parameter re-tuning, limiting their applicability in dynamic scenarios. We propose a DRL framework that learns adaptable control policies for multi-rotor drones performing waypoint navigation in AM tasks. We compare Deep Deterministic Policy Gradient (DDPG) and Twin Delayed Deep Deterministic Policy Gradient (TD3) within a curriculum learning scheme designed to handle increasing complexity. Our experiments show TD3 consistently balances training stability, accuracy, and success, particularly when mass variability is introduced. These findings provide a scalable path toward robust, autonomous drone control in additive manufacturing.
2502.05999
Pencils to Pixels: A Systematic Study of Creative Drawings across Children, Adults and AI
cs.HC cs.AI cs.CV
Can we derive computational metrics to quantify visual creativity in drawings across intelligent agents, while accounting for inherent differences in technical skill and style? To answer this, we curate a novel dataset consisting of 1338 drawings by children, adults and AI on a creative drawing task. We characterize two aspects of the drawings -- (1) style and (2) content. For style, we define measures of ink density, ink distribution and number of elements. For content, we use expert-annotated categories to study conceptual diversity, and image and text embeddings to compute distance measures. We compare the style, content and creativity of children, adults and AI drawings and build simple models to predict expert and automated creativity scores. We find significant differences in style and content in the groups -- children's drawings had more components, AI drawings had greater ink density, and adult drawings revealed maximum conceptual diversity. Notably, we highlight a misalignment between creativity judgments obtained through expert and automated ratings and discuss its implications. Through these efforts, our work provides, to the best of our knowledge, the first framework for studying human and artificial creativity beyond the textual modality, and attempts to arrive at the domain-agnostic principles underlying creativity. Our data and scripts are available on GitHub.
2502.06004
Analysis of LLM as a grammatical feature tagger for African American English
cs.CL cs.AI cs.LG
African American English (AAE) presents unique challenges in natural language processing (NLP). This research systematically compares the performance of available NLP models--rule-based, transformer-based, and large language models (LLMs)--capable of identifying key grammatical features of AAE, namely Habitual Be and Multiple Negation. These features were selected for their distinct grammatical complexity and frequency of occurrence. The evaluation involved sentence-level binary classification tasks, using both zero-shot and few-shot strategies. The analysis reveals that while LLMs show promise compared to the baseline, they are influenced by biases such as recency and unrelated features in the text such as formality. This study highlights the necessity for improved model training and architectural adjustments to better accommodate AAE's unique linguistic characteristics. Data and code are available.
2502.06006
FactIR: A Real-World Zero-shot Open-Domain Retrieval Benchmark for Fact-Checking
cs.IR
The field of automated fact-checking increasingly depends on retrieving web-based evidence to determine the veracity of claims in real-world scenarios. A significant challenge in this process is not only retrieving relevant information, but also identifying evidence that can both support and refute complex claims. Traditional retrieval methods may return documents that directly address claims or lean toward supporting them, but often struggle with more complex claims requiring indirect reasoning. While some existing benchmarks and methods target retrieval for fact-checking, a comprehensive real-world open-domain benchmark has been lacking. In this paper, we present a real-world retrieval benchmark FactIR, derived from Factiverse production logs, enhanced with human annotations. We rigorously evaluate state-of-the-art retrieval models in a zero-shot setup on FactIR and offer insights for developing practical retrieval systems for fact-checking. Code and data are available at https://github.com/factiverse/factIR.
2502.06007
Transformers versus the EM Algorithm in Multi-class Clustering
stat.ML cs.LG
LLMs demonstrate significant inference capacities in complicated machine learning tasks, using the Transformer model as its backbone. Motivated by the limited understanding of such models on the unsupervised learning problems, we study the learning guarantees of Transformers in performing multi-class clustering of the Gaussian Mixture Models. We develop a theory drawing strong connections between the Softmax Attention layers and the workflow of the EM algorithm on clustering the mixture of Gaussians. Our theory provides approximation bounds for the Expectation and Maximization steps by proving the universal approximation abilities of multivariate mappings by Softmax functions. In addition to the approximation guarantees, we also show that with a sufficient number of pre-training samples and an initialization, Transformers can achieve the minimax optimal rate for the problem considered. Our extensive simulations empirically verified our theory by revealing the strong learning capacities of Transformers even beyond the assumptions in the theory, shedding light on the powerful inference capacities of LLMs.
2502.06011
Uncertainty Quantification and Causal Considerations for Off-Policy Decision Making
stat.ML cs.LG
Off-policy evaluation (OPE) is a critical challenge in robust decision-making that seeks to assess the performance of a new policy using data collected under a different policy. However, the existing OPE methodologies suffer from several limitations arising from statistical uncertainty as well as causal considerations. In this thesis, we address these limitations by presenting three different works. Firstly, we consider the problem of high variance in the importance-sampling-based OPE estimators. We introduce the Marginal Ratio (MR) estimator, a novel OPE method that reduces variance by focusing on the marginal distribution of outcomes rather than direct policy shifts, improving robustness in contextual bandits. Next, we propose Conformal Off-Policy Prediction (COPP), a principled approach for uncertainty quantification in OPE that provides finite-sample predictive intervals, ensuring robust decision-making in risk-sensitive applications. Finally, we address causal unidentifiability in off-policy decision-making by developing novel bounds for sequential decision settings, which remain valid under arbitrary unmeasured confounding. We apply these bounds to assess the reliability of digital twin models, introducing a falsification framework to identify scenarios where model predictions diverge from real-world behaviour. Our contributions provide new insights into robust decision-making under uncertainty and establish principled methods for evaluating policies in both static and dynamic settings.
2502.06018
Kolmogorov-Arnold Fourier Networks
cs.LG cs.AI
Although Kolmogorov-Arnold based interpretable networks (KAN) have strong theoretical expressiveness, they face significant parameter explosion and high-frequency feature capture challenges in high-dimensional tasks. To address this issue, we propose the Kolmogorov-Arnold-Fourier Network (KAF), which effectively integrates trainable Random Fourier Features (RFF) and a novel hybrid GELU-Fourier activation mechanism to balance parameter efficiency and spectral representation capabilities. Our key technical contributions include: (1) merging KAN's dual-matrix structure through matrix association properties to substantially reduce parameters; (2) introducing learnable RFF initialization strategies to eliminate spectral distortion in high-dimensional approximation tasks; (3) implementing an adaptive hybrid activation function that progressively enhances frequency representation during the training process. Comprehensive experiments demonstrate the superiority of our KAF across various domains including vision, NLP, audio processing, and differential equation-solving tasks, effectively combining theoretical interpretability with practical utility and computational efficiency.
2502.06019
Noise is an Efficient Learner for Zero-Shot Vision-Language Models
cs.CV
Recently, test-time adaptation has garnered attention as a method for tuning models without labeled data. The conventional modus operandi for adapting pre-trained vision-language models (VLMs) during test-time primarily focuses on tuning learnable prompts; however, this approach overlooks potential distribution shifts in the visual representations themselves. In this work, we address this limitation by introducing Test-Time Noise Tuning (TNT), a novel method for handling unpredictable shifts in the visual space. TNT leverages, for the first time, a noise adaptation strategy that optimizes learnable noise directly in the visual input space, enabling adaptive feature learning from a single test sample. We further introduce a novel approach for inter-view representation alignment by explicitly enforcing coherence in embedding distances, ensuring consistent feature representations across views. Combined with scaled logits and confident view selection at inference, TNT substantially enhances VLM generalization and calibration, achieving average gains of +7.38% on natural distributions benchmark and +0.80% on cross-dataset evaluations over zero-shot CLIP. These improvements lay a strong foundation for adaptive out-of-distribution handling.
2502.06020
Temporal Working Memory: Query-Guided Segment Refinement for Enhanced Multimodal Understanding
cs.CV cs.MM cs.SD eess.AS
Multimodal foundation models (MFMs) have demonstrated significant success in tasks such as visual captioning, question answering, and image-text retrieval. However, these models face inherent limitations due to their finite internal capacity, which restricts their ability to process extended temporal sequences, a crucial requirement for comprehensive video and audio analysis. To overcome these challenges, we introduce a specialized cognitive module, temporal working memory (TWM), which aims to enhance the temporal modeling capabilities of MFMs. It selectively retains task-relevant information across temporal dimensions, ensuring that critical details are preserved throughout the processing of video and audio content. The TWM uses a query-guided attention approach to focus on the most informative multimodal segments within temporal sequences. By retaining only the most relevant content, TWM optimizes the use of the model's limited capacity, enhancing its temporal modeling ability. This plug-and-play module can be easily integrated into existing MFMs. With our TWM, nine state-of-the-art models exhibit significant performance improvements across tasks such as video captioning, question answering, and video-text retrieval. By enhancing temporal modeling, TWM extends the capability of MFMs to handle complex, time-sensitive data effectively. Our code is available at https://github.com/xid32/NAACL_2025_TWM.
2502.06022
Nested subspace learning with flags
stat.ML cs.LG
Many machine learning methods look for low-dimensional representations of the data. The underlying subspace can be estimated by first choosing a dimension $q$ and then optimizing a certain objective function over the space of $q$-dimensional subspaces (the Grassmannian). Trying different $q$ yields in general non-nested subspaces, which raises an important issue of consistency between the data representations. In this paper, we propose a simple trick to enforce nestedness in subspace learning methods. It consists in lifting Grassmannian optimization problems to flag manifolds (the space of nested subspaces of increasing dimension) via nested projectors. We apply the flag trick to several classical machine learning methods and show that it successfully addresses the nestedness issue.
2502.06023
Dual Caption Preference Optimization for Diffusion Models
cs.CV
Recent advancements in human preference optimization, originally developed for Large Language Models (LLMs), have shown significant potential in improving text-to-image diffusion models. These methods aim to learn the distribution of preferred samples while distinguishing them from less preferred ones. However, existing preference datasets often exhibit overlap between these distributions, leading to a conflict distribution. Additionally, we identified that input prompts contain irrelevant information for less preferred images, limiting the denoising network's ability to accurately predict noise in preference optimization methods, known as the irrelevant prompt issue. To address these challenges, we propose Dual Caption Preference Optimization (DCPO), a novel approach that utilizes two distinct captions to mitigate irrelevant prompts. To tackle conflict distribution, we introduce the Pick-Double Caption dataset, a modified version of Pick-a-Pic v2 with separate captions for preferred and less preferred images. We further propose three different strategies for generating distinct captions: captioning, perturbation, and hybrid methods. Our experiments show that DCPO significantly improves image quality and relevance to prompts, outperforming Stable Diffusion (SD) 2.1, SFT_Chosen, Diffusion-DPO, and MaPO across multiple metrics, including Pickscore, HPSv2.1, GenEval, CLIPscore, and ImageReward, fine-tuned on SD 2.1 as the backbone.
2502.06025
Universal point spread function engineering for 3D optical information processing
physics.optics cs.NE
Point spread function (PSF) engineering has been pivotal in the remarkable progress made in high-resolution imaging in the last decades. However, the diversity in PSF structures attainable through existing engineering methods is limited. Here, we report universal PSF engineering, demonstrating a method to synthesize an arbitrary set of spatially varying 3D PSFs between the input and output volumes of a spatially incoherent diffractive processor composed of cascaded transmissive surfaces. We rigorously analyze the PSF engineering capabilities of such diffractive processors within the diffraction limit of light and provide numerical demonstrations of unique imaging capabilities, such as snapshot 3D multispectral imaging without involving any spectral filters, axial scanning or digital reconstruction steps, which is enabled by the spatial and spectral engineering of 3D PSFs. Our framework and analysis would be important for future advancements in computational imaging, sensing and diffractive processing of 3D optical information.
2502.06026
A Multimodal PDE Foundation Model for Prediction and Scientific Text Descriptions
cs.LG cs.NA math.NA
Neural networks are one tool for approximating non-linear differential equations used in scientific computing tasks such as surrogate modeling, real-time predictions, and optimal control. PDE foundation models utilize neural networks to train approximations to multiple differential equations simultaneously and are thus a general purpose solver that can be adapted to downstream tasks. Current PDE foundation models focus on either learning general solution operators and/or the governing system of equations, and thus only handle numerical or symbolic modalities. However, real-world applications may require more flexible data modalities, e.g. text analysis or descriptive outputs. To address this gap, we propose a novel multimodal deep learning approach that leverages a transformer-based architecture to approximate solution operators for a wide variety of ODEs and PDEs. Our method integrates numerical inputs, such as equation parameters and initial conditions, with text descriptions of physical processes or system dynamics. This enables our model to handle settings where symbolic representations may be incomplete or unavailable. In addition to providing accurate numerical predictions, our approach generates interpretable scientific text descriptions, offering deeper insights into the underlying dynamics and solution properties. The numerical experiments show that our model provides accurate solutions for in-distribution data (with average relative error less than 3.3%) and out-of-distribution data (average relative error less than 7.8%) together with precise text descriptions (with correct descriptions generated 100% of times). In certain tests, the model is also shown to be capable of extrapolating solutions in time.
2502.06027
Generating 3D Binding Molecules Using Shape-Conditioned Diffusion Models with Guidance
cs.LG
Drug development is a critical but notoriously resource- and time-consuming process. In this manuscript, we develop a novel generative artificial intelligence (genAI) method DiffSMol to facilitate drug development. DiffSmol generates 3D binding molecules based on the shapes of known ligands. DiffSMol encapsulates geometric details of ligand shapes within pre-trained, expressive shape embeddings and then generates new binding molecules through a diffusion model. DiffSMol further modifies the generated 3D structures iteratively via shape guidance to better resemble the ligand shapes. It also tailors the generated molecules toward optimal binding affinities under the guidance of protein pockets. Here, we show that DiffSMol outperforms the state-of-the-art methods on benchmark datasets. When generating binding molecules resembling ligand shapes, DiffSMol with shape guidance achieves a success rate 61.4%, substantially outperforming the best baseline (11.2%), meanwhile producing molecules with novel molecular graph structures. DiffSMol with pocket guidance also outperforms the best baseline in binding affinities by 13.2%, and even by 17.7% when combined with shape guidance. Case studies for two critical drug targets demonstrate very favorable physicochemical and pharmacokinetic properties of the generated molecules, thus, the potential of DiffSMol in developing promising drug candidates.
2502.06029
DiTASK: Multi-Task Fine-Tuning with Diffeomorphic Transformations
cs.CV
Pre-trained Vision Transformers now serve as powerful tools for computer vision. Yet, efficiently adapting them for multiple tasks remains a challenge that arises from the need to modify the rich hidden representations encoded by the learned weight matrices, without inducing interference between tasks. Current parameter-efficient methods like LoRA, which apply low-rank updates, force tasks to compete within constrained subspaces, ultimately degrading performance. We introduce DiTASK a novel Diffeomorphic Multi-Task Fine-Tuning approach that maintains pre-trained representations by preserving weight matrix singular vectors, while enabling task-specific adaptations through neural diffeomorphic transformations of the singular values. By following this approach, DiTASK enables both shared and task-specific feature modulations with minimal added parameters. Our theoretical analysis shows that DITASK achieves full-rank updates during optimization, preserving the geometric structure of pre-trained features, and establishing a new paradigm for efficient multi-task learning (MTL). Our experiments on PASCAL MTL and NYUD show that DiTASK achieves state-of-the-art performance across four dense prediction tasks, using 75% fewer parameters than existing methods.
2502.06031
A Conditional Tabular GAN-Enhanced Intrusion Detection System for Rare Attacks in IoT Networks
cs.CR cs.LG
Internet of things (IoT) networks, boosted by 6G technology, are transforming various industries. However, their widespread adoption introduces significant security risks, particularly in detecting rare but potentially damaging cyber-attacks. This makes the development of robust IDS crucial for monitoring network traffic and ensuring their safety. Traditional IDS often struggle with detecting rare attacks due to severe class imbalances in IoT data. In this paper, we propose a novel two-stage system called conditional tabular generative synthetic minority data generation with deep neural network (CTGSM-DNN). In the first stage, a conditional tabular generative adversarial network (CTGAN) is employed to generate synthetic data for rare attack classes. In the second stage, the SMOTEENN method is applied to improve dataset quality. The full study was conducted using the CSE-CIC-IDS2018 dataset, and we assessed the performance of the proposed IDS using different evaluation metrics. The experimental results demonstrated the effectiveness of the proposed multiclass classifier, achieving an overall accuracy of 99.90% and 80% accuracy in detecting rare attacks.
2502.06034
Traveling Waves Integrate Spatial Information Into Spectral Representations
cs.CV
Traveling waves are widely observed in the brain, but their precise computational function remains unclear. One prominent hypothesis is that they enable the transfer and integration of spatial information across neural populations. However, few computational models have explored how traveling waves might be harnessed to perform such integrative processing. Drawing inspiration from the famous ``Can one hear the shape of a drum?'' problem -- which highlights how spectral modes encode geometric information -- we introduce a set of convolutional recurrent neural networks that learn to produce traveling waves in their hidden states in response to visual stimuli. By applying a spectral decomposition to these wave-like activations, we obtain a powerful new representational space that outperforms equivalently local feed-forward networks on tasks requiring global spatial context. In particular, we observe that traveling waves effectively expand the receptive field of locally connected neurons, supporting long-range encoding and communication of information. We demonstrate that models equipped with this mechanism and spectral readouts solve visual semantic segmentation tasks demanding global integration, where local feed-forward models fail. As a first step toward traveling-wave-based representations in artificial networks, our findings suggest potential efficiency benefits and offer a new framework for connecting to biological recordings of neural activity.
2502.06037
Investigating Compositional Reasoning in Time Series Foundation Models
cs.LG
Large pre-trained time series foundation models (TSFMs) have demonstrated promising zero-shot performance across a wide range of domains. However, a question remains: Do TSFMs succeed solely by memorizing training patterns, or do they possess the ability to reason? While reasoning is a topic of great interest in the study of Large Language Models (LLMs), it is undefined and largely unexplored in the context of TSFMs. In this work, inspired by language modeling literature, we formally define compositional reasoning in forecasting and distinguish it from in-distribution generalization. We evaluate the reasoning and generalization capabilities of 23 popular deep learning forecasting models on multiple synthetic and real-world datasets. Additionally, through controlled studies, we systematically examine which design choices in TSFMs contribute to improved reasoning abilities. Our study yields key insights into the impact of TSFM architecture design on compositional reasoning and generalization. We find that patch-based Transformers have the best reasoning performance, closely followed by residualized MLP-based architectures, which are 97\% less computationally complex in terms of FLOPs and 86\% smaller in terms of the number of trainable parameters. Interestingly, in some zero-shot out-of-distribution scenarios, these models can outperform moving average and exponential smoothing statistical baselines trained on in-distribution data. Only a few design choices, such as the tokenization method, had a significant (negative) impact on Transformer model performance.
2502.06038
Provably Overwhelming Transformer Models with Designed Inputs
cs.LG cs.AI cs.CC
We develop an algorithm which, given a trained transformer model $\mathcal{M}$ as input, as well as a string of tokens $s$ of length $n_{fix}$ and an integer $n_{free}$, can generate a mathematical proof that $\mathcal{M}$ is ``overwhelmed'' by $s$, in time and space $\widetilde{O}(n_{fix}^2 + n_{free}^3)$. We say that $\mathcal{M}$ is ``overwhelmed'' by $s$ when the output of the model evaluated on this string plus any additional string $t$, $\mathcal{M}(s + t)$, is completely insensitive to the value of the string $t$ whenever length($t$) $\leq n_{free}$. Along the way, we prove a particularly strong worst-case form of ``over-squashing'', which we use to bound the model's behavior. Our technique uses computer-aided proofs to establish this type of operationally relevant guarantee about transformer models. We empirically test our algorithm on a single layer transformer complete with an attention head, layer-norm, MLP/ReLU layers, and RoPE positional encoding. We believe that this work is a stepping stone towards the difficult task of obtaining useful guarantees for trained transformer models.
2502.06039
Benchmarking Prompt Engineering Techniques for Secure Code Generation with GPT Models
cs.SE cs.AI cs.CR
Prompt engineering reduces reasoning mistakes in Large Language Models (LLMs). However, its effectiveness in mitigating vulnerabilities in LLM-generated code remains underexplored. To address this gap, we implemented a benchmark to automatically assess the impact of various prompt engineering strategies on code security. Our benchmark leverages two peer-reviewed prompt datasets and employs static scanners to evaluate code security at scale. We tested multiple prompt engineering techniques on GPT-3.5-turbo, GPT-4o, and GPT-4o-mini. Our results show that for GPT-4o and GPT-4o-mini, a security-focused prompt prefix can reduce the occurrence of security vulnerabilities by up to 56%. Additionally, all tested models demonstrated the ability to detect and repair between 41.9% and 68.7% of vulnerabilities in previously generated code when using iterative prompting techniques. Finally, we introduce a "prompt agent" that demonstrates how the most effective techniques can be applied in real-world development workflows.
2502.06042
Scaling Laws for Forgetting during Finetuning with Pretraining Data Injection
cs.LG cs.CL
A widespread strategy to obtain a language model that performs well on a target domain is to finetune a pretrained model to perform unsupervised next-token prediction on data from that target domain. Finetuning presents two challenges: (i) if the amount of target data is limited, as in most practical applications, the model will quickly overfit, and (ii) the model will drift away from the original model, forgetting the pretraining data and the generic knowledge that comes with it. We aim to derive scaling laws that quantify these two phenomena for various target domains, amounts of available target data, and model scales. We measure the efficiency of injecting pretraining data into the finetuning data mixture to avoid forgetting and mitigate overfitting. A key practical takeaway from our study is that injecting as little as 1% of pretraining data in the finetuning data mixture prevents the model from forgetting the pretraining set.
2502.06044
Scalable Differentially Private Bayesian Optimization
stat.ML cs.LG
In recent years, there has been much work on scaling Bayesian Optimization to high-dimensional problems, for example hyperparameter tuning in large neural network models. These scalable methods have been successful, finding high objective values much more quickly than traditional global Bayesian Optimization or random search-based methods. At the same time, these large neural network models often use sensitive data, but preservation of Differential Privacy has not scaled alongside these modern Bayesian Optimization procedures. Here we develop a method to privately estimate potentially high-dimensional parameter spaces using Gradient Informative Bayesian Optimization. Our theoretical results prove that under suitable conditions, our method converges exponentially fast to a ball around the optimal parameter configuration. Moreover, regardless of whether the assumptions are satisfied, we show that our algorithm maintains privacy and empirically demonstrates superior performance to existing methods in the high-dimensional hyperparameter setting.
2502.06047
Neural Shortest Path for Surface Reconstruction from Point Clouds
cs.LG
In this paper, we propose the neural shortest path (NSP), a vector-valued implicit neural representation (INR) that approximates a distance function and its gradient. The key feature of NSP is to learn the exact shortest path (ESP), which directs an arbitrary point to its nearest point on the target surface. The NSP is decomposed into its magnitude and direction, and a variable splitting method is used that each decomposed component approximates a distance function and its gradient, respectively. Unlike to existing methods of learning the distance function itself, the NSP ensures the simultaneous recovery of the distance function and its gradient. We mathematically prove that the decomposed representation of NSP guarantees the convergence of the magnitude of NSP in the $H^1$ norm. Furthermore, we devise a novel loss function that enforces the property of ESP, demonstrating that its global minimum is the ESP. We evaluate the performance of the NSP through comprehensive experiments on diverse datasets, validating its capacity to reconstruct high-quality surfaces with the robustness to noise and data sparsity. The numerical results show substantial improvements over state-of-the-art methods, highlighting the importance of learning the ESP, the product of distance function and its gradient, for representing a wide variety of complex surfaces.
2502.06049
LM2: Large Memory Models
cs.CL cs.AI
This paper introduces the Large Memory Model (LM2), a decoder-only Transformer architecture enhanced with an auxiliary memory module that aims to address the limitations of standard Transformers in multi-step reasoning, relational argumentation, and synthesizing information distributed over long contexts. The proposed LM2 incorporates a memory module that acts as a contextual representation repository, interacting with input tokens via cross attention and updating through gating mechanisms. To preserve the Transformers general-purpose capabilities, LM2 maintains the original information flow while integrating a complementary memory pathway. Experimental results on the BABILong benchmark demonstrate that the LM2model outperforms both the memory-augmented RMT model by 37.1% and the baseline Llama-3.2 model by 86.3% on average across tasks. LM2 exhibits exceptional capabilities in multi-hop inference, numerical reasoning, and large-context question-answering. On the MMLU dataset, it achieves a 5.0% improvement over a pre-trained vanilla model, demonstrating that its memory module does not degrade performance on general tasks. Further, in our analysis, we explore the memory interpretability, effectiveness of memory modules, and test-time behavior. Our findings emphasize the importance of explicit memory in enhancing Transformer architectures.
2502.06051
Nearly Optimal Sample Complexity of Offline KL-Regularized Contextual Bandits under Single-Policy Concentrability
cs.LG cs.AI math.ST stat.ML stat.TH
KL-regularized policy optimization has become a workhorse in learning-based decision making, while its theoretical understanding is still very limited. Although recent progress has been made towards settling the sample complexity of KL-regularized contextual bandits, existing sample complexity bounds are either $\tilde{O}(\epsilon^{-2})$ under single-policy concentrability or $\tilde{O}(\epsilon^{-1})$ under all-policy concentrability. In this paper, we propose the \emph{first} algorithm with $\tilde{O}(\epsilon^{-1})$ sample complexity under single-policy concentrability for offline contextual bandits. Our algorithm is designed for general function approximation and based on the principle of \emph{pessimism in the face of uncertainty}. The core of our proof leverages the strong convexity of the KL regularization, and the conditional non-negativity of the gap between the true reward and its pessimistic estimator to refine a mean-value-type risk upper bound to its extreme. This in turn leads to a novel covariance-based analysis, effectively bypassing the need for uniform control over the discrepancy between any two functions in the function class. The near-optimality of our algorithm is demonstrated by an $\tilde{\Omega}(\epsilon^{-1})$ lower bound. Furthermore, we extend our algorithm to contextual dueling bandits and achieve a similar nearly optimal sample complexity.