id string | title string | abstract string | text string | label string | primary_category string | published string |
|---|---|---|---|---|---|---|
2512.19990v1 | A Dual-Branch Local-Global Framework for Cross-Resolution Land Cover Mapping | Cross-resolution land cover mapping aims to produce high-resolution semantic predictions from coarse or low-resolution supervision, yet the severe resolution mismatch makes effective learning highly challenging. Existing weakly supervised approaches often struggle to align fine-grained spatial structures with coarse la... | A Dual-Branch Local-Global Framework for Cross-Resolution Land Cover Mapping. Cross-resolution land cover mapping aims to produce high-resolution semantic predictions from coarse or low-resolution supervision, yet the severe resolution mismatch makes effective learning highly challenging. Existing weakly supervised app... | cv_vision | cs.CV | 2025-12-23T02:32:02+00:00 |
2512.19020v1 | CETCAM: Camera-Controllable Video Generation via Consistent and Extensible Tokenization | Achieving precise camera control in video generation remains challenging, as existing methods often rely on camera pose annotations that are difficult to scale to large and dynamic datasets and are frequently inconsistent with depth estimation, leading to train-test discrepancies. We introduce CETCAM, a camera-controll... | CETCAM: Camera-Controllable Video Generation via Consistent and Extensible Tokenization. Achieving precise camera control in video generation remains challenging, as existing methods often rely on camera pose annotations that are difficult to scale to large and dynamic datasets and are frequently inconsistent with dept... | cv_vision | cs.CV | 2025-12-22T04:21:39+00:00 |
2510.04556v2 | Model Monitoring: A General Framework with an Application to Non-life Insurance Pricing | Maintaining the predictive performance of pricing models is challenging when insurance portfolios and data-generating mechanisms evolve over time. Focusing on non-life insurance, we adopt the concept-drift terminology from machine learning and distinguish virtual drift from real concept drift in an actuarial setting. M... | Model Monitoring: A General Framework with an Application to Non-life Insurance Pricing. Maintaining the predictive performance of pricing models is challenging when insurance portfolios and data-generating mechanisms evolve over time. Focusing on non-life insurance, we adopt the concept-drift terminology from machine ... | stat_ml | stat.ML | 2025-10-06T07:41:09+00:00 |
2512.04228v1 | Addressing Logical Fallacies In Scientific Reasoning From Large Language Models: Towards a Dual-Inference Training Framework | Large Language Models (LLMs) have transformed natural language processing and hold growing promise for advancing science, healthcare, and decision-making. Yet their training paradigms remain dominated by affirmation-based inference, akin to \textit{modus ponens}, where accepted premises yield predicted consequents. Whi... | Addressing Logical Fallacies In Scientific Reasoning From Large Language Models: Towards a Dual-Inference Training Framework. Large Language Models (LLMs) have transformed natural language processing and hold growing promise for advancing science, healthcare, and decision-making. Yet their training paradigms remain dom... | ai_general | cs.AI | 2025-12-03T19:50:39+00:00 |
2512.20988v1 | PUFM++: Point Cloud Upsampling via Enhanced Flow Matching | Recent advances in generative modeling have demonstrated strong promise for high-quality point cloud upsampling. In this work, we present PUFM++, an enhanced flow-matching framework for reconstructing dense and accurate point clouds from sparse, noisy, and partial observations. PUFM++ improves flow matching along three... | PUFM++: Point Cloud Upsampling via Enhanced Flow Matching. Recent advances in generative modeling have demonstrated strong promise for high-quality point cloud upsampling. In this work, we present PUFM++, an enhanced flow-matching framework for reconstructing dense and accurate point clouds from sparse, noisy, and part... | cv_vision | cs.CV | 2025-12-24T06:30:42+00:00 |
2509.20587v3 | Unsupervised Domain Adaptation for Binary Classification with an Unobservable Source Subpopulation | We study an unsupervised domain adaptation problem where the source domain consists of subpopulations defined by the binary label $Y$ and a binary background (or environment) $A$. We focus on a challenging setting in which one such subpopulation in the source domain is unobservable. Naively ignoring this unobserved gro... | Unsupervised Domain Adaptation for Binary Classification with an Unobservable Source Subpopulation. We study an unsupervised domain adaptation problem where the source domain consists of subpopulations defined by the binary label $Y$ and a binary background (or environment) $A$. We focus on a challenging setting in whi... | stat_ml | stat.ML | 2025-09-24T22:00:49+00:00 |
2512.07384v1 | On the Impact of Graph Neural Networks in Recommender Systems: A Topological Perspective | In recommender systems, user-item interactions can be modeled as a bipartite graph, where user and item nodes are connected by undirected edges. This graph-based view has motivated the rapid adoption of graph neural networks (GNNs), which often outperform collaborative filtering (CF) methods such as latent factor model... | On the Impact of Graph Neural Networks in Recommender Systems: A Topological Perspective. In recommender systems, user-item interactions can be modeled as a bipartite graph, where user and item nodes are connected by undirected edges. This graph-based view has motivated the rapid adoption of graph neural networks (GNNs... | ir_retrieval | cs.IR | 2025-12-08T10:19:43+00:00 |
2511.03606v2 | Vector-valued self-normalized concentration inequalities beyond sub-Gaussianity | The study of self-normalized processes plays a crucial role in a wide range of applications, from sequential decision-making to econometrics. While the behavior of self-normalized concentration has been widely investigated for scalar-valued processes, vector-valued processes remain comparatively underexplored, especial... | Vector-valued self-normalized concentration inequalities beyond sub-Gaussianity. The study of self-normalized processes plays a crucial role in a wide range of applications, from sequential decision-making to econometrics. While the behavior of self-normalized concentration has been widely investigated for scalar-value... | stat_ml | stat.ML | 2025-11-05T16:27:02+00:00 |
2510.26095v1 | ORBIT -- Open Recommendation Benchmark for Reproducible Research with Hidden Tests | Recommender systems are among the most impactful AI applications, interacting with billions of users every day, guiding them to relevant products, services, or information tailored to their preferences. However, the research and development of recommender systems are hindered by existing datasets that fail to capture r... | ORBIT -- Open Recommendation Benchmark for Reproducible Research with Hidden Tests. Recommender systems are among the most impactful AI applications, interacting with billions of users every day, guiding them to relevant products, services, or information tailored to their preferences. However, the research and develop... | ir_retrieval | cs.IR | 2025-10-30T03:10:45+00:00 |
2509.20928v3 | Conditionally Whitened Generative Models for Probabilistic Time Series Forecasting | Probabilistic forecasting of multivariate time series is challenging due to non-stationarity, inter-variable dependencies, and distribution shifts. While recent diffusion and flow matching models have shown promise, they often ignore informative priors such as conditional means and covariances. In this work, we propose... | Conditionally Whitened Generative Models for Probabilistic Time Series Forecasting. Probabilistic forecasting of multivariate time series is challenging due to non-stationarity, inter-variable dependencies, and distribution shifts. While recent diffusion and flow matching models have shown promise, they often ignore in... | stat_ml | stat.ML | 2025-09-25T09:13:37+00:00 |
2512.16902v2 | In-Context Algebra | We investigate the mechanisms that arise when transformers are trained to solve arithmetic on sequences where tokens are variables whose meaning is determined only through their interactions in-context. While prior work has studied transformers in settings where the answer relies on fixed parametric or geometric inform... | In-Context Algebra. We investigate the mechanisms that arise when transformers are trained to solve arithmetic on sequences where tokens are variables whose meaning is determined only through their interactions in-context. While prior work has studied transformers in settings where the answer relies on fixed parametric... | cl_nlp | cs.CL | 2025-12-18T18:56:50+00:00 |
2512.15011v2 | Epistemic diversity across language models mitigates knowledge collapse | As artificial intelligence (AI) becomes more widely used, concerns are growing that model collapse could lead to knowledge collapse, i.e. a degradation to a narrow and inaccurate set of ideas. Prior work has demonstrated single-model collapse, defined as performance decay in an AI model trained on its own outputs. Insp... | Epistemic diversity across language models mitigates knowledge collapse. As artificial intelligence (AI) becomes more widely used, concerns are growing that model collapse could lead to knowledge collapse, i.e. a degradation to a narrow and inaccurate set of ideas. Prior work has demonstrated single-model collapse, def... | lg_ml | cs.LG | 2025-12-17T02:03:28+00:00 |
2510.18161v3 | Beating the Winner's Curse via Inference-Aware Policy Optimization | There has been a surge of recent interest in automatically learning policies to target treatment decisions based on rich individual covariates. In addition, practitioners want confidence that the learned policy has better performance than the incumbent policy according to downstream policy evaluation. However, due to t... | Beating the Winner's Curse via Inference-Aware Policy Optimization. There has been a surge of recent interest in automatically learning policies to target treatment decisions based on rich individual covariates. In addition, practitioners want confidence that the learned policy has better performance than the incumbent... | stat_ml | stat.ML | 2025-10-20T23:28:12+00:00 |
2510.08985v1 | Rethinking Reasoning in Document Ranking: Why Chain-of-Thought Falls Short | Document reranking is a key component in information retrieval (IR), aimed at refining initial retrieval results to improve ranking quality for downstream tasks. Recent studies--motivated by large reasoning models (LRMs)--have begun incorporating explicit chain-of-thought (CoT) reasoning into LLM-based rerankers. Howev... | Rethinking Reasoning in Document Ranking: Why Chain-of-Thought Falls Short. Document reranking is a key component in information retrieval (IR), aimed at refining initial retrieval results to improve ranking quality for downstream tasks. Recent studies--motivated by large reasoning models (LRMs)--have begun incorporati... | ir_retrieval | cs.IR | 2025-10-10T03:59:17+00:00 |
2512.21506v1 | MotionTeller: Multi-modal Integration of Wearable Time-Series with LLMs for Health and Behavioral Understanding | As wearable sensing becomes increasingly pervasive, a key challenge remains: how can we generate natural language summaries from raw physiological signals such as actigraphy - minute-level movement data collected via accelerometers? In this work, we introduce MotionTeller, a generative framework that natively integrate... | MotionTeller: Multi-modal Integration of Wearable Time-Series with LLMs for Health and Behavioral Understanding. As wearable sensing becomes increasingly pervasive, a key challenge remains: how can we generate natural language summaries from raw physiological signals such as actigraphy - minute-level movement data coll... | lg_ml | cs.LG | 2025-12-25T04:37:07+00:00 |
2512.18092v1 | Faithful and Stable Neuron Explanations for Trustworthy Mechanistic Interpretability | Neuron identification is a popular tool in mechanistic interpretability, aiming to uncover the human-interpretable concepts represented by individual neurons in deep networks. While algorithms such as Network Dissection and CLIP-Dissect achieve great empirical success, a rigorous theoretical foundation remains absent, ... | Faithful and Stable Neuron Explanations for Trustworthy Mechanistic Interpretability. Neuron identification is a popular tool in mechanistic interpretability, aiming to uncover the human-interpretable concepts represented by individual neurons in deep networks. While algorithms such as Network Dissection and CLIP-Disse... | ai_general | cs.AI | 2025-12-19T21:55:17+00:00 |
2512.23419v1 | The World Is Bigger! A Computationally-Embedded Perspective on the Big World Hypothesis | Continual learning is often motivated by the idea, known as the big world hypothesis, that "the world is bigger" than the agent. Recent problem formulations capture this idea by explicitly constraining an agent relative to the environment. These constraints lead to solutions in which the agent continually adapts to bes... | The World Is Bigger! A Computationally-Embedded Perspective on the Big World Hypothesis. Continual learning is often motivated by the idea, known as the big world hypothesis, that "the world is bigger" than the agent. Recent problem formulations capture this idea by explicitly constraining an agent relative to the envi... | ai_general | cs.AI | 2025-12-29T12:31:46+00:00 |
2512.18248v3 | On the Convergence Rate of LoRA Gradient Descent | The low-rank adaptation (LoRA) algorithm for fine-tuning large models has grown popular in recent years due to its remarkable performance and low computational requirements. LoRA trains two ``adapter" matrices that form a low-rank representation of the model parameters, thereby massively reducing the number of paramete... | On the Convergence Rate of LoRA Gradient Descent. The low-rank adaptation (LoRA) algorithm for fine-tuning large models has grown popular in recent years due to its remarkable performance and low computational requirements. LoRA trains two ``adapter" matrices that form a low-rank representation of the model parameters,... | lg_ml | cs.LG | 2025-12-20T07:20:54+00:00 |
2512.18635v1 | Uni-Neur2Img: Unified Neural Signal-Guided Image Generation, Editing, and Stylization via Diffusion Transformers | Generating or editing images directly from Neural signals has immense potential at the intersection of neuroscience, vision, and Brain-computer interaction. In this paper, We present Uni-Neur2Img, a unified framework for neural signal-driven image generation and editing. The framework introduces a parameter-efficient L... | Uni-Neur2Img: Unified Neural Signal-Guided Image Generation, Editing, and Stylization via Diffusion Transformers. Generating or editing images directly from Neural signals has immense potential at the intersection of neuroscience, vision, and Brain-computer interaction. In this paper, We present Uni-Neur2Img, a unified... | cv_vision | cs.CV | 2025-12-21T08:12:38+00:00 |
2512.20213v1 | JDPNet: A Network Based on Joint Degradation Processing for Underwater Image Enhancement | Given the complexity of underwater environments and the variability of water as a medium, underwater images are inevitably subject to various types of degradation. The degradations present nonlinear coupling rather than simple superposition, which renders the effective processing of such coupled degradations particular... | JDPNet: A Network Based on Joint Degradation Processing for Underwater Image Enhancement. Given the complexity of underwater environments and the variability of water as a medium, underwater images are inevitably subject to various types of degradation. The degradations present nonlinear coupling rather than simple sup... | cv_vision | cs.CV | 2025-12-23T10:12:35+00:00 |
2512.24504v2 | Thinking on Maps: How Foundation Model Agents Explore, Remember, and Reason Map Environments | Map environments provide a fundamental medium for representing spatial structure. Understanding how foundation model (FM) agents understand and act in such environments is therefore critical for enabling reliable map-based reasoning and applications. However, most existing evaluations of spatial ability in FMs rely on ... | Thinking on Maps: How Foundation Model Agents Explore, Remember, and Reason Map Environments. Map environments provide a fundamental medium for representing spatial structure. Understanding how foundation model (FM) agents understand and act in such environments is therefore critical for enabling reliable map-based rea... | ai_general | cs.AI | 2025-12-30T23:04:29+00:00 |
2512.10772v1 | Grow Up and Merge: Scaling Strategies for Efficient Language Adaptation | Achieving high-performing language models which include medium- and lower-resource languages remains a challenge. Massively multilingual models still underperform compared to language-specific adaptations, especially at smaller model scales. In this work, we investigate scaling as an efficient strategy for adapting pre... | Grow Up and Merge: Scaling Strategies for Efficient Language Adaptation. Achieving high-performing language models which include medium- and lower-resource languages remains a challenge. Massively multilingual models still underperform compared to language-specific adaptations, especially at smaller model scales. In th... | cl_nlp | cs.CL | 2025-12-11T16:09:54+00:00 |
2512.25014v1 | Diffusion Language Models are Provably Optimal Parallel Samplers | Diffusion language models (DLMs) have emerged as a promising alternative to autoregressive models for faster inference via parallel token generation. We provide a rigorous foundation for this advantage by formalizing a model of parallel sampling and showing that DLMs augmented with polynomial-length chain-of-thought (C... | Diffusion Language Models are Provably Optimal Parallel Samplers. Diffusion language models (DLMs) have emerged as a promising alternative to autoregressive models for faster inference via parallel token generation. We provide a rigorous foundation for this advantage by formalizing a model of parallel sampling and show... | lg_ml | cs.LG | 2025-12-31T18:03:05+00:00 |
2512.06296v1 | How Sharp and Bias-Robust is a Model? Dual Evaluation Perspectives on Knowledge Graph Completion | Knowledge graph completion (KGC) aims to predict missing facts from the observed KG. While a number of KGC models have been studied, the evaluation of KGC still remain underexplored. In this paper, we observe that existing metrics overlook two key perspectives for KGC evaluation: (A1) predictive sharpness -- the degree... | How Sharp and Bias-Robust is a Model? Dual Evaluation Perspectives on Knowledge Graph Completion. Knowledge graph completion (KGC) aims to predict missing facts from the observed KG. While a number of KGC models have been studied, the evaluation of KGC still remain underexplored. In this paper, we observe that existing... | ai_general | cs.AI | 2025-12-06T04:49:29+00:00 |
2512.24885v1 | BEDA: Belief Estimation as Probabilistic Constraints for Performing Strategic Dialogue Acts | Strategic dialogue requires agents to execute distinct dialogue acts, for which belief estimation is essential. While prior work often estimates beliefs accurately, it lacks a principled mechanism to use those beliefs during generation. We bridge this gap by first formalizing two core acts Adversarial and Alignment, an... | BEDA: Belief Estimation as Probabilistic Constraints for Performing Strategic Dialogue Acts. Strategic dialogue requires agents to execute distinct dialogue acts, for which belief estimation is essential. While prior work often estimates beliefs accurately, it lacks a principled mechanism to use those beliefs during ge... | cl_nlp | cs.CL | 2025-12-31T14:26:55+00:00 |
2512.23427v1 | Towards Integrating Uncertainty for Domain-Agnostic Segmentation | Foundation models for segmentation such as the Segment Anything Model (SAM) family exhibit strong zero-shot performance, but remain vulnerable in shifted or limited-knowledge domains. This work investigates whether uncertainty quantification can mitigate such challenges and enhance model generalisability in a domain-ag... | Towards Integrating Uncertainty for Domain-Agnostic Segmentation. Foundation models for segmentation such as the Segment Anything Model (SAM) family exhibit strong zero-shot performance, but remain vulnerable in shifted or limited-knowledge domains. This work investigates whether uncertainty quantification can mitigate... | cv_vision | cs.CV | 2025-12-29T12:46:21+00:00 |
2512.23133v1 | Principled Algorithms for Optimizing Generalized Metrics in Binary Classification | In applications with significant class imbalance or asymmetric costs, metrics such as the $F_β$-measure, AM measure, Jaccard similarity coefficient, and weighted accuracy offer more suitable evaluation criteria than standard binary classification loss. However, optimizing these metrics present significant computational... | Principled Algorithms for Optimizing Generalized Metrics in Binary Classification. In applications with significant class imbalance or asymmetric costs, metrics such as the $F_β$-measure, AM measure, Jaccard similarity coefficient, and weighted accuracy offer more suitable evaluation criteria than standard binary class... | lg_ml | cs.LG | 2025-12-29T01:33:42+00:00 |
2509.19559v2 | Stochastic Path Planning in Correlated Obstacle Fields | We introduce the Stochastic Correlated Obstacle Scene (SCOS) problem, a navigation setting with spatially correlated obstacles of uncertain blockage status, realistically constrained sensors that provide noisy readings and costly disambiguation. Modeling the spatial correlation with Gaussian Random Field (GRF), we deve... | Stochastic Path Planning in Correlated Obstacle Fields. We introduce the Stochastic Correlated Obstacle Scene (SCOS) problem, a navigation setting with spatially correlated obstacles of uncertain blockage status, realistically constrained sensors that provide noisy readings and costly disambiguation. Modeling the spati... | stat_ml | stat.ML | 2025-09-23T20:30:35+00:00 |
2512.07132v1 | DART: Leveraging Multi-Agent Disagreement for Tool Recruitment in Multimodal Reasoning | Specialized visual tools can augment large language models or vision language models with expert knowledge (e.g., grounding, spatial reasoning, medical knowledge, etc.), but knowing which tools to call (and when to call them) can be challenging. We introduce DART, a multi-agent framework that uses disagreements between... | DART: Leveraging Multi-Agent Disagreement for Tool Recruitment in Multimodal Reasoning. Specialized visual tools can augment large language models or vision language models with expert knowledge (e.g., grounding, spatial reasoning, medical knowledge, etc.), but knowing which tools to call (and when to call them) can be... | cl_nlp | cs.CL | 2025-12-08T03:33:38+00:00 |
2511.16921v1 | δ-EMG: A Monotonic Graph Index for Approximate Nearest Neighbor Search | Approximate nearest neighbor (ANN) search in high-dimensional spaces is a foundational component of many modern retrieval and recommendation systems. Currently, almost all algorithms follow an $ε$-Recall-Bounded principle when comparing performance: they require the ANN search results to achieve a recall of more than $... | δ-EMG: A Monotonic Graph Index for Approximate Nearest Neighbor Search. Approximate nearest neighbor (ANN) search in high-dimensional spaces is a foundational component of many modern retrieval and recommendation systems. Currently, almost all algorithms follow an $ε$-Recall-Bounded principle when comparing performance... | ir_retrieval | cs.IR | 2025-11-21T03:20:54+00:00 |
2512.18809v2 | FedVideoMAE: Efficient Privacy-Preserving Federated Video Moderation | Short-form video moderation increasingly needs learning pipelines that protect user privacy without paying the full bandwidth and latency cost of cloud-centralized inference. We present FedVideoMAE, an on-device federated framework for video violence detection that combines self-supervised VideoMAE representations, LoR... | FedVideoMAE: Efficient Privacy-Preserving Federated Video Moderation. Short-form video moderation increasingly needs learning pipelines that protect user privacy without paying the full bandwidth and latency cost of cloud-centralized inference. We present FedVideoMAE, an on-device federated framework for video violence... | cv_vision | cs.CV | 2025-12-21T17:01:44+00:00 |
2512.24438v1 | Exploring Compositionality in Vision Transformers using Wavelet Representations | While insights into the workings of the transformer model have largely emerged by analysing their behaviour on language tasks, this work investigates the representations learnt by the Vision Transformer (ViT) encoder through the lens of compositionality. We introduce a framework, analogous to prior work on measuring co... | Exploring Compositionality in Vision Transformers using Wavelet Representations. While insights into the workings of the transformer model have largely emerged by analysing their behaviour on language tasks, this work investigates the representations learnt by the Vision Transformer (ViT) encoder through the lens of co... | cv_vision | cs.CV | 2025-12-30T19:43:40+00:00 |
2512.17891v1 | Keypoint Counting Classifiers: Turning Vision Transformers into Self-Explainable Models Without Training | Current approaches for designing self-explainable models (SEMs) require complicated training procedures and specific architectures which makes them impractical. With the advance of general purpose foundation models based on Vision Transformers (ViTs), this impracticability becomes even more problematic. Therefore, new ... | Keypoint Counting Classifiers: Turning Vision Transformers into Self-Explainable Models Without Training. Current approaches for designing self-explainable models (SEMs) require complicated training procedures and specific architectures which makes them impractical. With the advance of general purpose foundation models... | cv_vision | cs.CV | 2025-12-19T18:47:04+00:00 |
2512.15601v1 | You Never Know a Person, You Only Know Their Defenses: Detecting Levels of Psychological Defense Mechanisms in Supportive Conversations | Psychological defenses are strategies, often automatic, that people use to manage distress. Rigid or overuse of defenses is negatively linked to mental health and shapes what speakers disclose and how they accept or resist help. However, defenses are complex and difficult to reliably measure, particularly in clinical d... | You Never Know a Person, You Only Know Their Defenses: Detecting Levels of Psychological Defense Mechanisms in Supportive Conversations. Psychological defenses are strategies, often automatic, that people use to manage distress. Rigid or overuse of defenses is negatively linked to mental health and shapes what speakers... | cl_nlp | cs.CL | 2025-12-17T17:11:05+00:00 |
2512.18314v1 | MatSpray: Fusing 2D Material World Knowledge on 3D Geometry | Manual modeling of material parameters and 3D geometry is a time consuming yet essential task in the gaming and film industries. While recent advances in 3D reconstruction have enabled accurate approximations of scene geometry and appearance, these methods often fall short in relighting scenarios due to the lack of pre... | MatSpray: Fusing 2D Material World Knowledge on 3D Geometry. Manual modeling of material parameters and 3D geometry is a time consuming yet essential task in the gaming and film industries. While recent advances in 3D reconstruction have enabled accurate approximations of scene geometry and appearance, these methods of... | cv_vision | cs.CV | 2025-12-20T10:58:45+00:00 |
2511.16599v1 | Time dependent loss reweighting for flow matching and diffusion models is theoretically justified | This brief note clarifies that, in Generator Matching (which subsumes a large family of flow matching and diffusion models over continuous, manifold, and discrete spaces), both the Bregman divergence loss and the linear parameterization of the generator can depend on both the current state $X_t$ and the time $t$, and w... | Time dependent loss reweighting for flow matching and diffusion models is theoretically justified. This brief note clarifies that, in Generator Matching (which subsumes a large family of flow matching and diffusion models over continuous, manifold, and discrete spaces), both the Bregman divergence loss and the linear p... | stat_ml | stat.ML | 2025-11-20T17:55:21+00:00 |
2512.20556v1 | Multi-Grained Text-Guided Image Fusion for Multi-Exposure and Multi-Focus Scenarios | Image fusion aims to synthesize a single high-quality image from a pair of inputs captured under challenging conditions, such as differing exposure levels or focal depths. A core challenge lies in effectively handling disparities in dynamic range and focus depth between the inputs. With the advent of vision-language mo... | Multi-Grained Text-Guided Image Fusion for Multi-Exposure and Multi-Focus Scenarios. Image fusion aims to synthesize a single high-quality image from a pair of inputs captured under challenging conditions, such as differing exposure levels or focal depths. A core challenge lies in effectively handling disparities in dy... | cv_vision | cs.CV | 2025-12-23T17:55:35+00:00 |
2512.15228v1 | Accelerating High-Throughput Catalyst Screening by Direct Generation of Equilibrium Adsorption Structures | The adsorption energy serves as a crucial descriptor for the large-scale screening of catalysts. Nevertheless, the limited distribution of training data for the extensively utilised machine learning interatomic potential (MLIP), predominantly sourced from near-equilibrium structures, results in unreliable adsorption st... | Accelerating High-Throughput Catalyst Screening by Direct Generation of Equilibrium Adsorption Structures. The adsorption energy serves as a crucial descriptor for the large-scale screening of catalysts. Nevertheless, the limited distribution of training data for the extensively utilised machine learning interatomic po... | lg_ml | cs.LG | 2025-12-17T09:26:58+00:00 |
2512.20749v1 | Stabilizing Multimodal Autoencoders: A Theoretical and Empirical Analysis of Fusion Strategies | In recent years, the development of multimodal autoencoders has gained significant attention due to their potential to handle multimodal complex data types and improve model performance. Understanding the stability and robustness of these models is crucial for optimizing their training, architecture, and real-world app... | Stabilizing Multimodal Autoencoders: A Theoretical and Empirical Analysis of Fusion Strategies. In recent years, the development of multimodal autoencoders has gained significant attention due to their potential to handle multimodal complex data types and improve model performance. Understanding the stability and robus... | lg_ml | cs.LG | 2025-12-23T20:12:01+00:00 |
2512.14733v1 | Where to Explore: A Reach and Cost-Aware Approach for Unbiased Data Collection in Recommender Systems | Exploration is essential to improve long-term recommendation quality, but it often degrades short-term business performance, especially in remote-first TV environments where users engage passively, expect instant relevance, and offer few chances for correction. This paper introduces an approach for delivering content-l... | Where to Explore: A Reach and Cost-Aware Approach for Unbiased Data Collection in Recommender Systems. Exploration is essential to improve long-term recommendation quality, but it often degrades short-term business performance, especially in remote-first TV environments where users engage passively, expect instant rele... | ir_retrieval | cs.IR | 2025-12-11T04:03:44+00:00 |
2510.11100v2 | HoMer: Addressing Heterogeneities by Modeling Sequential and Set-wise Contexts for CTR Prediction | Click-through rate (CTR) prediction, which models behavior sequence and non-sequential features (e.g., user/item profiles or cross features) to infer user interest, underpins industrial recommender systems. However, most methods face three forms of heterogeneity that degrade predictive performance: (i) Feature Heteroge... | HoMer: Addressing Heterogeneities by Modeling Sequential and Set-wise Contexts for CTR Prediction. Click-through rate (CTR) prediction, which models behavior sequence and non-sequential features (e.g., user/item profiles or cross features) to infer user interest, underpins industrial recommender systems. However, most ... | ir_retrieval | cs.IR | 2025-10-13T07:47:03+00:00 |
2512.22664v1 | Unleashing Foundation Vision Models: Adaptive Transfer for Diverse Data-Limited Scientific Domains | In the big data era, the computer vision field benefits from large-scale datasets such as LAION-2B, LAION-400M, and ImageNet-21K, Kinetics, on which popular models like the ViT and ConvNeXt series have been pre-trained, acquiring substantial knowledge. However, numerous downstream tasks in specialized and data-limited ... | Unleashing Foundation Vision Models: Adaptive Transfer for Diverse Data-Limited Scientific Domains. In the big data era, the computer vision field benefits from large-scale datasets such as LAION-2B, LAION-400M, and ImageNet-21K, Kinetics, on which popular models like the ViT and ConvNeXt series have been pre-trained, ... | cv_vision | cs.CV | 2025-12-27T17:32:59+00:00 |
2509.07289v2 | Kernel VICReg for Self-Supervised Learning in Reproducing Kernel Hilbert Space | Self-supervised learning (SSL) has emerged as a powerful paradigm for representation learning by optimizing geometric objectives, such as invariance to augmentations, variance preservation, and feature decorrelation, without requiring labels. However, most existing methods operate in Euclidean space, limiting their abi... | Kernel VICReg for Self-Supervised Learning in Reproducing Kernel Hilbert Space. Self-supervised learning (SSL) has emerged as a powerful paradigm for representation learning by optimizing geometric objectives, such as invariance to augmentations, variance preservation, and feature decorrelation, without requiring label... | stat_ml | stat.ML | 2025-09-08T23:49:21+00:00 |
2512.07355v1 | A Geometric Unification of Concept Learning with Concept Cones | Two traditions of interpretability have evolved side by side but seldom spoken to each other: Concept Bottleneck Models (CBMs), which prescribe what a concept should be, and Sparse Autoencoders (SAEs), which discover what concepts emerge. While CBMs use supervision to align activations with human-labeled concepts, SAEs... | A Geometric Unification of Concept Learning with Concept Cones. Two traditions of interpretability have evolved side by side but seldom spoken to each other: Concept Bottleneck Models (CBMs), which prescribe what a concept should be, and Sparse Autoencoders (SAEs), which discover what concepts emerge. While CBMs use su... | ai_general | cs.AI | 2025-12-08T09:51:46+00:00 |
2512.19125v1 | SAP: Syntactic Attention Pruning for Transformer-based Language Models | This paper introduces Syntactic Attention Pruning (SAP), a novel method for effectively pruning attention heads in Transformer models. Unlike conventional approaches that rely solely on mathematical analysis of model weights and activations, SAP incorporates both the syntactic structure and attention patterns of senten... | SAP: Syntactic Attention Pruning for Transformer-based Language Models. This paper introduces Syntactic Attention Pruning (SAP), a novel method for effectively pruning attention heads in Transformer models. Unlike conventional approaches that rely solely on mathematical analysis of model weights and activations, SAP in... | cl_nlp | cs.CL | 2025-12-22T08:05:01+00:00 |
2512.24792v1 | Projection-based Adversarial Attack using Physics-in-the-Loop Optimization for Monocular Depth Estimation | Deep neural networks (DNNs) remain vulnerable to adversarial attacks that cause misclassification when specific perturbations are added to input images. This vulnerability also threatens the reliability of DNN-based monocular depth estimation (MDE) models, making robustness enhancement a critical need in practical appl... | Projection-based Adversarial Attack using Physics-in-the-Loop Optimization for Monocular Depth Estimation. Deep neural networks (DNNs) remain vulnerable to adversarial attacks that cause misclassification when specific perturbations are added to input images. This vulnerability also threatens the reliability of DNN-bas... | cv_vision | cs.CV | 2025-12-31T11:30:03+00:00 |
2512.23739v1 | Break Out the Silverware -- Semantic Understanding of Stored Household Items | ``Bring me a plate.'' For domestic service robots, this simple command reveals a complex challenge: inferring where everyday items are stored, often out of sight in drawers, cabinets, or closets. Despite advances in vision and manipulation, robots still lack the commonsense reasoning needed to complete this task. We in... | Break Out the Silverware -- Semantic Understanding of Stored Household Items. ``Bring me a plate.'' For domestic service robots, this simple command reveals a complex challenge: inferring where everyday items are stored, often out of sight in drawers, cabinets, or closets. Despite advances in vision and manipulation, r... | cl_nlp | cs.CL | 2025-12-25T15:21:49+00:00 |
2512.20275v1 | Graph-Symbolic Policy Enforcement and Control (G-SPEC): A Neuro-Symbolic Framework for Safe Agentic AI in 5G Autonomous Networks | As networks evolve toward 5G Standalone and 6G, operators face orchestration challenges that exceed the limits of static automation and Deep Reinforcement Learning. Although Large Language Model (LLM) agents offer a path toward intent-based networking, they introduce stochastic risks, including topology hallucinations ... | Graph-Symbolic Policy Enforcement and Control (G-SPEC): A Neuro-Symbolic Framework for Safe Agentic AI in 5G Autonomous Networks. As networks evolve toward 5G Standalone and 6G, operators face orchestration challenges that exceed the limits of static automation and Deep Reinforcement Learning. Although Large Language M... | ai_general | cs.AI | 2025-12-23T11:27:17+00:00 |
2512.17970v1 | CodeGEMM: A Codebook-Centric Approach to Efficient GEMM in Quantized LLMs | Weight-only quantization is widely used to mitigate the memory-bound nature of LLM inference. Codebook-based methods extend this trend by achieving strong accuracy in the extremely low-bit regime (e.g., 2-bit). However, current kernels rely on dequantization, which repeatedly fetches centroids and reconstructs weights,... | CodeGEMM: A Codebook-Centric Approach to Efficient GEMM in Quantized LLMs. Weight-only quantization is widely used to mitigate the memory-bound nature of LLM inference. Codebook-based methods extend this trend by achieving strong accuracy in the extremely low-bit regime (e.g., 2-bit). However, current kernels rely on d... | lg_ml | cs.LG | 2025-12-19T06:16:32+00:00 |
2512.14157v1 | Incentivizing Tool-augmented Thinking with Images for Medical Image Analysis | Recent reasoning based medical MLLMs have made progress in generating step by step textual reasoning chains. However, they still struggle with complex tasks that necessitate dynamic and iterative focusing on fine-grained visual regions to achieve precise grounding and diagnosis. We introduce Ophiuchus, a versatile, too... | Incentivizing Tool-augmented Thinking with Images for Medical Image Analysis. Recent reasoning based medical MLLMs have made progress in generating step by step textual reasoning chains. However, they still struggle with complex tasks that necessitate dynamic and iterative focusing on fine-grained visual regions to ach... | ai_general | cs.AI | 2025-12-16T07:37:23+00:00 |
2602.23369v1 | Reason to Contrast: A Cascaded Multimodal Retrieval Framework | Traditional multimodal retrieval systems rely primarily on bi-encoder architectures, where performance is closely tied to embedding dimensionality. Recent work, Think-Then-Embed (TTE), shows that incorporating multimodal reasoning to elicit additional informative tokens before embedding can further improve retrieval. I... | Reason to Contrast: A Cascaded Multimodal Retrieval Framework. Traditional multimodal retrieval systems rely primarily on bi-encoder architectures, where performance is closely tied to embedding dimensionality. Recent work, Think-Then-Embed (TTE), shows that incorporating multimodal reasoning to elicit additional infor... | ir_retrieval | cs.IR | 2025-12-21T04:52:48+00:00 |
2512.17640v1 | Generative Human-Object Interaction Detection via Differentiable Cognitive Steering of Multi-modal LLMs | Human-object interaction (HOI) detection aims to localize human-object pairs and the interactions between them. Existing methods operate under a closed-world assumption, treating the task as a classification problem over a small, predefined verb set, which struggles to generalize to the long-tail of unseen or ambiguous... | Generative Human-Object Interaction Detection via Differentiable Cognitive Steering of Multi-modal LLMs. Human-object interaction (HOI) detection aims to localize human-object pairs and the interactions between them. Existing methods operate under a closed-world assumption, treating the task as a classification problem... | cv_vision | cs.CV | 2025-12-19T14:41:50+00:00 |
2512.12297v1 | F5-TTS-RO: Extending F5-TTS to Romanian TTS via Lightweight Input Adaptation | This work introduces a lightweight input-level adapter for the F5-TTS model that enables Romanian Language support. To preserve the existing capabilities of the model (voice cloning, English and Chinese support), we keep the original weights frozen, append a sub-network to the model and train it as an extension for the... | F5-TTS-RO: Extending F5-TTS to Romanian TTS via Lightweight Input Adaptation. This work introduces a lightweight input-level adapter for the F5-TTS model that enables Romanian Language support. To preserve the existing capabilities of the model (voice cloning, English and Chinese support), we keep the original weights ... | cl_nlp | cs.CL | 2025-12-13T11:41:54+00:00 |
2512.16530v1 | Plain language adaptations of biomedical text using LLMs: Comparision of evaluation metrics | This study investigated the application of Large Language Models (LLMs) for simplifying biomedical texts to enhance health literacy. Using a public dataset, which included plain language adaptations of biomedical abstracts, we developed and evaluated several approaches, specifically a baseline approach using a prompt t... | Plain language adaptations of biomedical text using LLMs: Comparision of evaluation metrics. This study investigated the application of Large Language Models (LLMs) for simplifying biomedical texts to enhance health literacy. Using a public dataset, which included plain language adaptations of biomedical abstracts, we ... | cl_nlp | cs.CL | 2025-12-18T13:37:58+00:00 |
2509.25507v1 | One-shot Conditional Sampling: MMD meets Nearest Neighbors | How can we generate samples from a conditional distribution that we never fully observe? This question arises across a broad range of applications in both modern machine learning and classical statistics, including image post-processing in computer vision, approximate posterior sampling in simulation-based inference, a... | One-shot Conditional Sampling: MMD meets Nearest Neighbors. How can we generate samples from a conditional distribution that we never fully observe? This question arises across a broad range of applications in both modern machine learning and classical statistics, including image post-processing in computer vision, app... | stat_ml | stat.ML | 2025-09-29T21:04:50+00:00 |
2512.08449v1 | From Accuracy to Impact: The Impact-Driven AI Framework (IDAIF) for Aligning Engineering Architecture with Theory of Change | This paper introduces the Impact-Driven AI Framework (IDAIF), a novel architectural methodology that integrates Theory of Change (ToC) principles with modern artificial intelligence system design. As AI systems increasingly influence high-stakes domains including healthcare, finance, and public policy, the alignment pr... | From Accuracy to Impact: The Impact-Driven AI Framework (IDAIF) for Aligning Engineering Architecture with Theory of Change. This paper introduces the Impact-Driven AI Framework (IDAIF), a novel architectural methodology that integrates Theory of Change (ToC) principles with modern artificial intelligence system design... | ai_general | cs.AI | 2025-12-09T10:21:02+00:00 |
2512.19214v1 | HippMetric: A skeletal-representation-based framework for cross-sectional and longitudinal hippocampal substructural morphometry | Accurate characterization of hippocampal substructure is crucial for detecting subtle structural changes and identifying early neurodegenerative biomarkers. However, high inter-subject variability and complex folding pattern of human hippocampus hinder consistent cross-subject and longitudinal analysis. Most existing a... | HippMetric: A skeletal-representation-based framework for cross-sectional and longitudinal hippocampal substructural morphometry. Accurate characterization of hippocampal substructure is crucial for detecting subtle structural changes and identifying early neurodegenerative biomarkers. However, high inter-subject varia... | cv_vision | cs.CV | 2025-12-22T09:53:55+00:00 |
2512.22867v1 | MUSON: A Reasoning-oriented Multimodal Dataset for Socially Compliant Navigation in Urban Environments | Socially compliant navigation requires structured reasoning over dynamic pedestrians and physical constraints to ensure safe and interpretable decisions. However, existing social navigation datasets often lack explicit reasoning supervision and exhibit highly long-tailed action distributions, limiting models' ability t... | MUSON: A Reasoning-oriented Multimodal Dataset for Socially Compliant Navigation in Urban Environments. Socially compliant navigation requires structured reasoning over dynamic pedestrians and physical constraints to ensure safe and interpretable decisions. However, existing social navigation datasets often lack explic... | cv_vision | cs.CV | 2025-12-28T10:41:39+00:00 |
2510.24601v2 | Comparison of generalised additive models and neural networks in applications: A systematic review | Neural networks have become a popular tool in predictive modelling, more commonly associated with machine learning and artificial intelligence than with statistics. Generalised Additive Models (GAMs) are flexible non-linear statistical models that retain interpretability. Both are state-of-the-art in their own right, w... | Comparison of generalised additive models and neural networks in applications: A systematic review. Neural networks have become a popular tool in predictive modelling, more commonly associated with machine learning and artificial intelligence than with statistics. Generalised Additive Models (GAMs) are flexible non-lin... | stat_ml | stat.ML | 2025-10-28T16:28:42+00:00 |
2512.23142v1 | Domain-Shift Immunity in Deep Deformable Registration via Local Feature Representations | Deep learning has advanced deformable image registration, surpassing traditional optimization-based methods in both accuracy and efficiency. However, learning-based models are widely believed to be sensitive to domain shift, with robustness typically pursued through large and diverse training datasets, without explaini... | Domain-Shift Immunity in Deep Deformable Registration via Local Feature Representations. Deep learning has advanced deformable image registration, surpassing traditional optimization-based methods in both accuracy and efficiency. However, learning-based models are widely believed to be sensitive to domain shift, with r... | cv_vision | cs.CV | 2025-12-29T02:10:52+00:00 |
2511.00490v1 | Accuracy estimation of neural networks by extreme value theory | Neural networks are able to approximate any continuous function on a compact set. However, it is not obvious how to quantify the error of the neural network, i.e., the remaining bias between the function and the neural network. Here, we propose the application of extreme value theory to quantify large values of the err... | Accuracy estimation of neural networks by extreme value theory. Neural networks are able to approximate any continuous function on a compact set. However, it is not obvious how to quantify the error of the neural network, i.e., the remaining bias between the function and the neural network. Here, we propose the applica... | stat_ml | stat.ML | 2025-11-01T10:52:51+00:00 |
2512.07424v1 | OnePiece: The Great Route to Generative Recommendation -- A Case Study from Tencent Algorithm Competition | In past years, the OpenAI's Scaling-Laws shows the amazing intelligence with the next-token prediction paradigm in neural language modeling, which pointing out a free-lunch way to enhance the model performance by scaling the model parameters. In RecSys, the retrieval stage is also follows a 'next-token prediction' para... | OnePiece: The Great Route to Generative Recommendation -- A Case Study from Tencent Algorithm Competition. In past years, the OpenAI's Scaling-Laws shows the amazing intelligence with the next-token prediction paradigm in neural language modeling, which pointing out a free-lunch way to enhance the model performance by ... | ir_retrieval | cs.IR | 2025-12-08T10:56:56+00:00 |
2512.25067v1 | FineTec: Fine-Grained Action Recognition Under Temporal Corruption via Skeleton Decomposition and Sequence Completion | Recognizing fine-grained actions from temporally corrupted skeleton sequences remains a significant challenge, particularly in real-world scenarios where online pose estimation often yields substantial missing data. Existing methods often struggle to accurately recover temporal dynamics and fine-grained spatial structu... | FineTec: Fine-Grained Action Recognition Under Temporal Corruption via Skeleton Decomposition and Sequence Completion. Recognizing fine-grained actions from temporally corrupted skeleton sequences remains a significant challenge, particularly in real-world scenarios where online pose estimation often yields substantial... | cv_vision | cs.CV | 2025-12-31T18:59:12+00:00 |
2512.23676v1 | Web World Models | Language agents increasingly require persistent worlds in which they can act, remember, and learn. Existing approaches sit at two extremes: conventional web frameworks provide reliable but fixed contexts backed by databases, while fully generative world models aim for unlimited environments at the expense of controllab... | Web World Models. Language agents increasingly require persistent worlds in which they can act, remember, and learn. Existing approaches sit at two extremes: conventional web frameworks provide reliable but fixed contexts backed by databases, while fully generative world models aim for unlimited environments at the exp... | ai_general | cs.AI | 2025-12-29T18:31:45+00:00 |
2512.23020v2 | OpenGround: Active Cognition-based Reasoning for Open-World 3D Visual Grounding | 3D visual grounding aims to locate objects based on natural language descriptions in 3D scenes. Existing methods rely on a pre-defined Object Lookup Table (OLT) to query Visual Language Models (VLMs) for reasoning about object locations, which limits the applications in scenarios with undefined or unforeseen targets. T... | OpenGround: Active Cognition-based Reasoning for Open-World 3D Visual Grounding. 3D visual grounding aims to locate objects based on natural language descriptions in 3D scenes. Existing methods rely on a pre-defined Object Lookup Table (OLT) to query Visual Language Models (VLMs) for reasoning about object locations, w... | cv_vision | cs.CV | 2025-12-28T17:44:20+00:00 |
2512.20417v1 | Chain-of-Anomaly Thoughts with Large Vision-Language Models | Automated video surveillance with Large Vision-Language Models is limited by their inherent bias towards normality, often failing to detect crimes. While Chain-of-Thought reasoning strategies show significant potential for improving performance in language tasks, the lack of inductive anomaly biases in their reasoning ... | Chain-of-Anomaly Thoughts with Large Vision-Language Models. Automated video surveillance with Large Vision-Language Models is limited by their inherent bias towards normality, often failing to detect crimes. While Chain-of-Thought reasoning strategies show significant potential for improving performance in language ta... | cv_vision | cs.CV | 2025-12-23T15:01:05+00:00 |
2601.05257v1 | KP-Agent: Keyword Pruning in Sponsored Search Advertising via LLM-Powered Contextual Bandits | Sponsored search advertising (SSA) requires advertisers to constantly adjust keyword strategies. While bid adjustment and keyword generation are well-studied, keyword pruning-refining keyword sets to enhance campaign performance-remains under-explored. This paper addresses critical inefficiencies in current practices a... | KP-Agent: Keyword Pruning in Sponsored Search Advertising via LLM-Powered Contextual Bandits. Sponsored search advertising (SSA) requires advertisers to constantly adjust keyword strategies. While bid adjustment and keyword generation are well-studied, keyword pruning-refining keyword sets to enhance campaign performan... | ir_retrieval | cs.IR | 2025-10-20T16:13:48+00:00 |
2512.07075v1 | Do Large Language Models Truly Understand Cross-cultural Differences? | In recent years, large language models (LLMs) have demonstrated strong performance on multilingual tasks. Given its wide range of applications, cross-cultural understanding capability is a crucial competency. However, existing benchmarks for evaluating whether LLMs genuinely possess this capability suffer from three ke... | Do Large Language Models Truly Understand Cross-cultural Differences?. In recent years, large language models (LLMs) have demonstrated strong performance on multilingual tasks. Given its wide range of applications, cross-cultural understanding capability is a crucial competency. However, existing benchmarks for evaluat... | cl_nlp | cs.CL | 2025-12-08T01:21:58+00:00 |
2512.11614v2 | Bounding Hallucinations: Information-Theoretic Guarantees for RAG Systems via Merlin-Arthur Protocols | Retrieval-augmented generation (RAG) relies on retrieved context to guide large language models (LLM), yet treats retrieval as a weak heuristic rather than verifiable evidence -- leading to unsupported answers, hallucinations, and reliance on spurious context. We introduce a novel training framework that treats the RAG... | Bounding Hallucinations: Information-Theoretic Guarantees for RAG Systems via Merlin-Arthur Protocols. Retrieval-augmented generation (RAG) relies on retrieved context to guide large language models (LLM), yet treats retrieval as a weak heuristic rather than verifiable evidence -- leading to unsupported answers, halluc... | cl_nlp | cs.CL | 2025-12-12T14:50:38+00:00 |
2512.19734v1 | The Deleuzian Representation Hypothesis | We propose an alternative to sparse autoencoders (SAEs) as a simple and effective unsupervised method for extracting interpretable concepts from neural networks. The core idea is to cluster differences in activations, which we formally justify within a discriminant analysis framework. To enhance the diversity of extrac... | The Deleuzian Representation Hypothesis. We propose an alternative to sparse autoencoders (SAEs) as a simple and effective unsupervised method for extracting interpretable concepts from neural networks. The core idea is to cluster differences in activations, which we formally justify within a discriminant analysis fram... | lg_ml | cs.LG | 2025-12-17T11:51:25+00:00 |
2510.00073v1 | Identifying All ε-Best Arms in (Misspecified) Linear Bandits | Motivated by the need to efficiently identify multiple candidates in high trial-and-error cost tasks such as drug discovery, we propose a near-optimal algorithm to identify all ε-best arms (i.e., those at most ε worse than the optimum). Specifically, we introduce LinFACT, an algorithm designed to optimize the identific... | Identifying All ε-Best Arms in (Misspecified) Linear Bandits. Motivated by the need to efficiently identify multiple candidates in high trial-and-error cost tasks such as drug discovery, we propose a near-optimal algorithm to identify all ε-best arms (i.e., those at most ε worse than the optimum). Specifically, we intr... | stat_ml | stat.ML | 2025-09-29T22:26:52+00:00 |
2512.23765v1 | Entropy-Aware Speculative Decoding Toward Improved LLM Reasoning | Speculative decoding (SD) accelerates large language model (LLM) reasoning by using a small draft model to generate candidate tokens, which the target LLM either accepts directly or regenerates upon rejection. However, excessive alignment between the draft and target models constrains SD to the performance of the targe... | Entropy-Aware Speculative Decoding Toward Improved LLM Reasoning. Speculative decoding (SD) accelerates large language model (LLM) reasoning by using a small draft model to generate candidate tokens, which the target LLM either accepts directly or regenerates upon rejection. However, excessive alignment between the dra... | cl_nlp | cs.CL | 2025-12-29T00:45:19+00:00 |
2512.20104v1 | Effect of Activation Function and Model Optimizer on the Performance of Human Activity Recognition System Using Various Deep Learning Models | Human Activity Recognition (HAR) plays a vital role in healthcare, surveillance, and innovative environments, where reliable action recognition supports timely decision-making and automation. Although deep learning-based HAR systems are widely adopted, the impact of Activation Functions (AFs) and Model Optimizers (MOs)... | Effect of Activation Function and Model Optimizer on the Performance of Human Activity Recognition System Using Various Deep Learning Models. Human Activity Recognition (HAR) plays a vital role in healthcare, surveillance, and innovative environments, where reliable action recognition supports timely decision-making an... | cv_vision | cs.CV | 2025-12-23T07:01:45+00:00 |
2510.22670v1 | Tools are under-documented: Simple Document Expansion Boosts Tool Retrieval | Large Language Models (LLMs) have recently demonstrated strong capabilities in tool use, yet progress in tool retrieval remains hindered by incomplete and heterogeneous tool documentation. To address this challenge, we introduce Tool-DE, a new benchmark and framework that systematically enriches tool documentation with... | Tools are under-documented: Simple Document Expansion Boosts Tool Retrieval. Large Language Models (LLMs) have recently demonstrated strong capabilities in tool use, yet progress in tool retrieval remains hindered by incomplete and heterogeneous tool documentation. To address this challenge, we introduce Tool-DE, a new... | ir_retrieval | cs.IR | 2025-10-26T13:17:01+00:00 |
2512.21760v1 | A-QCF-Net: An Adaptive Quaternion Cross-Fusion Network for Multimodal Liver Tumor Segmentation from Unpaired Datasets | Multimodal medical imaging provides complementary information that is crucial for accurate delineation of pathology, but the development of deep learning models is limited by the scarcity of large datasets in which different modalities are paired and spatially aligned. This paper addresses this fundamental limitation b... | A-QCF-Net: An Adaptive Quaternion Cross-Fusion Network for Multimodal Liver Tumor Segmentation from Unpaired Datasets. Multimodal medical imaging provides complementary information that is crucial for accurate delineation of pathology, but the development of deep learning models is limited by the scarcity of large data... | cv_vision | cs.CV | 2025-12-25T18:42:21+00:00 |
2509.21940v2 | Sequential 1-bit Mean Estimation with Near-Optimal Sample Complexity | In this paper, we study the problem of distributed mean estimation with 1-bit communication constraints. We propose a mean estimator that is based on (randomized and sequentially-chosen) interval queries, whose 1-bit outcome indicates whether the given sample lies in the specified interval. Our estimator is $(ε, δ)$-PA... | Sequential 1-bit Mean Estimation with Near-Optimal Sample Complexity. In this paper, we study the problem of distributed mean estimation with 1-bit communication constraints. We propose a mean estimator that is based on (randomized and sequentially-chosen) interval queries, whose 1-bit outcome indicates whether the giv... | stat_ml | stat.ML | 2025-09-26T06:22:57+00:00 |
2512.17878v1 | Weighted Stochastic Differential Equation to Implement Wasserstein-Fisher-Rao Gradient Flow | Score-based diffusion models currently constitute the state of the art in continuous generative modeling. These methods are typically formulated via overdamped or underdamped Ornstein--Uhlenbeck-type stochastic differential equations, in which sampling is driven by a combination of deterministic drift and Brownian diff... | Weighted Stochastic Differential Equation to Implement Wasserstein-Fisher-Rao Gradient Flow. Score-based diffusion models currently constitute the state of the art in continuous generative modeling. These methods are typically formulated via overdamped or underdamped Ornstein--Uhlenbeck-type stochastic differential equ... | lg_ml | cs.LG | 2025-12-19T18:31:27+00:00 |
2512.22758v1 | A Micro-Macro Machine Learning Framework for Predicting Childhood Obesity Risk Using NHANES and Environmental Determinants | Childhood obesity remains a major public health challenge in the United States, strongly influenced by a combination of individual-level, household-level, and environmental-level risk factors. Traditional epidemiological studies typically analyze these levels independently, limiting insights into how structural environ... | A Micro-Macro Machine Learning Framework for Predicting Childhood Obesity Risk Using NHANES and Environmental Determinants. Childhood obesity remains a major public health challenge in the United States, strongly influenced by a combination of individual-level, household-level, and environmental-level risk factors. Tra... | lg_ml | cs.LG | 2025-12-28T03:20:04+00:00 |
2511.00444v1 | LIR: The First Workshop on Late Interaction and Multi Vector Retrieval @ ECIR 2026 | Late interaction retrieval methods, pioneered by ColBERT, have emerged as a powerful alternative to single-vector neural IR. By leveraging fine-grained, token-level representations, they have been demonstrated to deliver strong generalisation and robustness, particularly in out-of-domain settings. They have recently be... | LIR: The First Workshop on Late Interaction and Multi Vector Retrieval @ ECIR 2026. Late interaction retrieval methods, pioneered by ColBERT, have emerged as a powerful alternative to single-vector neural IR. By leveraging fine-grained, token-level representations, they have been demonstrated to deliver strong generali... | ir_retrieval | cs.IR | 2025-11-01T08:21:33+00:00 |
2512.05946v1 | Variational Quantum Rainbow Deep Q-Network for Optimizing Resource Allocation Problem | Resource allocation remains NP-hard due to combinatorial complexity. While deep reinforcement learning (DRL) methods, such as the Rainbow Deep Q-Network (DQN), improve scalability through prioritized replay and distributional heads, classical function approximators limit their representational power. We introduce Varia... | Variational Quantum Rainbow Deep Q-Network for Optimizing Resource Allocation Problem. Resource allocation remains NP-hard due to combinatorial complexity. While deep reinforcement learning (DRL) methods, such as the Rainbow Deep Q-Network (DQN), improve scalability through prioritized replay and distributional heads, ... | ai_general | cs.AI | 2025-12-05T18:43:18+00:00 |
2512.24901v1 | Spectral Graph Neural Networks for Cognitive Task Classification in fMRI Connectomes | Cognitive task classification using machine learning plays a central role in decoding brain states from neuroimaging data. By integrating machine learning with brain network analysis, complex connectivity patterns can be extracted from functional magnetic resonance imaging connectomes. This process transforms raw blood... | Spectral Graph Neural Networks for Cognitive Task Classification in fMRI Connectomes. Cognitive task classification using machine learning plays a central role in decoding brain states from neuroimaging data. By integrating machine learning with brain network analysis, complex connectivity patterns can be extracted fro... | lg_ml | cs.LG | 2025-12-31T14:54:06+00:00 |
2510.12013v1 | Statistical Guarantees for High-Dimensional Stochastic Gradient Descent | Stochastic Gradient Descent (SGD) and its Ruppert-Polyak averaged variant (ASGD) lie at the heart of modern large-scale learning, yet their theoretical properties in high-dimensional settings are rarely understood. In this paper, we provide rigorous statistical guarantees for constant learning-rate SGD and ASGD in high... | Statistical Guarantees for High-Dimensional Stochastic Gradient Descent. Stochastic Gradient Descent (SGD) and its Ruppert-Polyak averaged variant (ASGD) lie at the heart of modern large-scale learning, yet their theoretical properties in high-dimensional settings are rarely understood. In this paper, we provide rigoro... | stat_ml | stat.ML | 2025-10-13T23:27:56+00:00 |
2510.21598v1 | Fisher meets Feynman: score-based variational inference with a product of experts | We introduce a highly expressive yet distinctly tractable family for black-box variational inference (BBVI). Each member of this family is a weighted product of experts (PoE), and each weighted expert in the product is proportional to a multivariate $t$-distribution. These products of experts can model distributions wi... | Fisher meets Feynman: score-based variational inference with a product of experts. We introduce a highly expressive yet distinctly tractable family for black-box variational inference (BBVI). Each member of this family is a weighted product of experts (PoE), and each weighted expert in the product is proportional to a ... | stat_ml | stat.ML | 2025-10-24T16:02:18+00:00 |
2512.19516v1 | LacaDM: A Latent Causal Diffusion Model for Multiobjective Reinforcement Learning | Multiobjective reinforcement learning (MORL) poses significant challenges due to the inherent conflicts between objectives and the difficulty of adapting to dynamic environments. Traditional methods often struggle to generalize effectively, particularly in large and complex state-action spaces. To address these limitat... | LacaDM: A Latent Causal Diffusion Model for Multiobjective Reinforcement Learning. Multiobjective reinforcement learning (MORL) poses significant challenges due to the inherent conflicts between objectives and the difficulty of adapting to dynamic environments. Traditional methods often struggle to generalize effective... | lg_ml | cs.LG | 2025-12-22T16:08:03+00:00 |
2509.13189v1 | SURGIN: SURrogate-guided Generative INversion for subsurface multiphase flow with quantified uncertainty | We present a direct inverse modeling method named SURGIN, a SURrogate-guided Generative INversion framework tailed for subsurface multiphase flow data assimilation. Unlike existing inversion methods that require adaptation for each new observational configuration, SURGIN features a zero-shot conditional generation capa... | SURGIN: SURrogate-guided Generative INversion for subsurface multiphase flow with quantified uncertainty. We present a direct inverse modeling method named SURGIN, a SURrogate-guided Generative INversion framework tailed for subsurface multiphase flow data assimilation. Unlike existing inversion methods that require ad... | stat_ml | stat.ML | 2025-09-16T15:42:22+00:00 |
2601.00834v3 | Intrinsic-Metric Physics-Informed Neural Networks (IM-PINN) for Reaction-Diffusion Dynamics on Complex Riemannian Manifolds | Simulating nonlinear reaction-diffusion dynamics on complex, non-Euclidean manifolds remains a fundamental challenge in computational morphogenesis, constrained by high-fidelity mesh generation costs and symplectic drift in discrete time-stepping schemes. This study introduces the Intrinsic-Metric Physics-Informed Neur... | Intrinsic-Metric Physics-Informed Neural Networks (IM-PINN) for Reaction-Diffusion Dynamics on Complex Riemannian Manifolds. Simulating nonlinear reaction-diffusion dynamics on complex, non-Euclidean manifolds remains a fundamental challenge in computational morphogenesis, constrained by high-fidelity mesh generation c... | lg_ml | cs.LG | 2025-12-26T12:41:05+00:00 |
2601.02392v1 | Self-Supervised Masked Autoencoders with Dense-Unet for Coronary Calcium Removal in limited CT Data | Coronary calcification creates blooming artifacts in Computed Tomography Angiography (CTA), severely hampering the diagnosis of lumen stenosis. While Deep Convolutional Neural Networks (DCNNs) like Dense-Unet have shown promise in removing these artifacts via inpainting, they often require large labeled datasets which ... | Self-Supervised Masked Autoencoders with Dense-Unet for Coronary Calcium Removal in limited CT Data. Coronary calcification creates blooming artifacts in Computed Tomography Angiography (CTA), severely hampering the diagnosis of lumen stenosis. While Deep Convolutional Neural Networks (DCNNs) like Dense-Unet have shown... | cv_vision | cs.CV | 2025-12-25T12:36:37+00:00 |
2510.25402v2 | Towards Automated Quality Assurance of Patent Specifications: A Multi-Dimensional LLM Framework | Although AI drafting tools have gained prominence in patent writing, the systematic evaluation of AI-generated patent content quality represents a significant research gap. To address this gap, We propose to evaluate patents using regulatory compliance, technical coherence, and figure-reference consistency detection mo... | Towards Automated Quality Assurance of Patent Specifications: A Multi-Dimensional LLM Framework. Although AI drafting tools have gained prominence in patent writing, the systematic evaluation of AI-generated patent content quality represents a significant research gap. To address this gap, We propose to evaluate patent... | ir_retrieval | cs.IR | 2025-10-29T11:20:18+00:00 |
2512.17160v2 | Can Synthetic Images Serve as Effective and Efficient Class Prototypes? | Vision-Language Models (VLMs) have shown strong performance in zero-shot image classification tasks. However, existing methods, including Contrastive Language-Image Pre-training (CLIP), all rely on annotated text-to-image pairs for aligning visual and textual modalities. This dependency introduces substantial cost and ... | Can Synthetic Images Serve as Effective and Efficient Class Prototypes?. Vision-Language Models (VLMs) have shown strong performance in zero-shot image classification tasks. However, existing methods, including Contrastive Language-Image Pre-training (CLIP), all rely on annotated text-to-image pairs for aligning visual... | cv_vision | cs.CV | 2025-12-19T01:39:43+00:00 |
2509.03236v5 | OneSearch: A Preliminary Exploration of the Unified End-to-End Generative Framework for E-commerce Search | Traditional e-commerce search systems employ multi-stage cascading architectures (MCA) that progressively filter items through recall, pre-ranking, and ranking stages. While effective at balancing computational efficiency with business conversion, these systems suffer from fragmented computation and optimization object... | OneSearch: A Preliminary Exploration of the Unified End-to-End Generative Framework for E-commerce Search. Traditional e-commerce search systems employ multi-stage cascading architectures (MCA) that progressively filter items through recall, pre-ranking, and ranking stages. While effective at balancing computational ef... | ir_retrieval | cs.IR | 2025-09-03T11:50:04+00:00 |
2512.14036v1 | DTRec: Learning Dynamic Reasoning Trajectories for Sequential Recommendation | Inspired by advances in LLMs, reasoning-enhanced sequential recommendation performs multi-step deliberation before making final predictions, unlocking greater potential for capturing user preferences. However, current methods are constrained by static reasoning trajectories that are ill-suited for the diverse complexit... | DTRec: Learning Dynamic Reasoning Trajectories for Sequential Recommendation. Inspired by advances in LLMs, reasoning-enhanced sequential recommendation performs multi-step deliberation before making final predictions, unlocking greater potential for capturing user preferences. However, current methods are constrained ... | ir_retrieval | cs.IR | 2025-12-16T03:04:43+00:00 |
2512.21902v1 | Explainable Statute Prediction via Attention-based Model and LLM Prompting | In this paper, we explore the problem of automatic statute prediction where for a given case description, a subset of relevant statutes are to be predicted. Here, the term "statute" refers to a section, a sub-section, or an article of any specific Act. Addressing this problem would be useful in several applications suc... | Explainable Statute Prediction via Attention-based Model and LLM Prompting. In this paper, we explore the problem of automatic statute prediction where for a given case description, a subset of relevant statutes are to be predicted. Here, the term "statute" refers to a section, a sub-section, or an article of any speci... | cl_nlp | cs.CL | 2025-12-26T07:29:51+00:00 |
2510.01560v1 | AI Foundation Model for Time Series with Innovations Representation | This paper introduces an Artificial Intelligence (AI) foundation model for time series in engineering applications, where causal operations are required for real-time monitoring and control. Since engineering time series are governed by physical, rather than linguistic, laws, large-language-model-based AI foundation mo... | AI Foundation Model for Time Series with Innovations Representation. This paper introduces an Artificial Intelligence (AI) foundation model for time series in engineering applications, where causal operations are required for real-time monitoring and control. Since engineering time series are governed by physical, rath... | stat_ml | stat.ML | 2025-10-02T01:14:20+00:00 |
2512.18176v1 | Atlas is Your Perfect Context: One-Shot Customization for Generalizable Foundational Medical Image Segmentation | Accurate medical image segmentation is essential for clinical diagnosis and treatment planning. While recent interactive foundation models (e.g., nnInteractive) enhance generalization through large-scale multimodal pretraining, they still depend on precise prompts and often perform below expectations in contexts that a... | Atlas is Your Perfect Context: One-Shot Customization for Generalizable Foundational Medical Image Segmentation. Accurate medical image segmentation is essential for clinical diagnosis and treatment planning. While recent interactive foundation models (e.g., nnInteractive) enhance generalization through large-scale mul... | cv_vision | cs.CV | 2025-12-20T02:29:16+00:00 |
2512.24172v1 | Deep Global Clustering for Hyperspectral Image Segmentation: Concepts, Applications, and Open Challenges | Hyperspectral imaging (HSI) analysis faces computational bottlenecks due to massive data volumes that exceed available memory. While foundation models pre-trained on large remote sensing datasets show promise, their learned representations often fail to transfer to domain-specific applications like close-range agricult... | Deep Global Clustering for Hyperspectral Image Segmentation: Concepts, Applications, and Open Challenges. Hyperspectral imaging (HSI) analysis faces computational bottlenecks due to massive data volumes that exceed available memory. While foundation models pre-trained on large remote sensing datasets show promise, thei... | cv_vision | cs.CV | 2025-12-30T12:10:43+00:00 |
2512.11653v2 | Causal Inference in Energy Demand Prediction | Energy demand prediction is critical for grid operators, industrial energy consumers, and service providers. Energy demand is influenced by multiple factors, including weather conditions (e.g. temperature, humidity, wind speed, solar radiation), and calendar information (e.g. hour of day and month of year), which furth... | Causal Inference in Energy Demand Prediction. Energy demand prediction is critical for grid operators, industrial energy consumers, and service providers. Energy demand is influenced by multiple factors, including weather conditions (e.g. temperature, humidity, wind speed, solar radiation), and calendar information (e.... | ai_general | cs.AI | 2025-12-12T15:30:46+00:00 |
2512.22470v1 | DarkPatterns-LLM: A Multi-Layer Benchmark for Detecting Manipulative and Harmful AI Behavior | The proliferation of Large Language Models (LLMs) has intensified concerns about manipulative or deceptive behaviors that can undermine user autonomy, trust, and well-being. Existing safety benchmarks predominantly rely on coarse binary labels and fail to capture the nuanced psychological and social mechanisms constitu... | DarkPatterns-LLM: A Multi-Layer Benchmark for Detecting Manipulative and Harmful AI Behavior. The proliferation of Large Language Models (LLMs) has intensified concerns about manipulative or deceptive behaviors that can undermine user autonomy, trust, and well-being. Existing safety benchmarks predominantly rely on coa... | ai_general | cs.AI | 2025-12-27T05:05:46+00:00 |
2512.19746v1 | Robust Causal Directionality Inference in Quantum Inference under MNAR Observation and High-Dimensional Noise | In quantum mechanics, observation actively shapes the system, paralleling the statistical notion of Missing Not At Random (MNAR). This study introduces a unified framework for \textbf{robust causal directionality inference} in quantum engineering, determining whether relations are system$\to$observation, observation$\t... | Robust Causal Directionality Inference in Quantum Inference under MNAR Observation and High-Dimensional Noise. In quantum mechanics, observation actively shapes the system, paralleling the statistical notion of Missing Not At Random (MNAR). This study introduces a unified framework for \textbf{robust causal directional... | stat_ml | stat.ML | 2025-12-18T05:26:25+00:00 |
2512.23916v2 | Constraint Breeds Generalization: Temporal Dynamics as an Inductive Bias | Conventional deep learning prioritizes unconstrained optimization, yet biological systems operate under strict metabolic constraints. We propose that these physical constraints shape dynamics to function not as limitations, but as a temporal inductive bias that breeds generalization. Through a phase-space analysis of s... | Constraint Breeds Generalization: Temporal Dynamics as an Inductive Bias. Conventional deep learning prioritizes unconstrained optimization, yet biological systems operate under strict metabolic constraints. We propose that these physical constraints shape dynamics to function not as limitations, but as a temporal indu... | lg_ml | cs.LG | 2025-12-30T00:34:24+00:00 |
2512.23835v1 | Explaining News Bias Detection: A Comparative SHAP Analysis of Transformer Model Decision Mechanisms | Automated bias detection in news text is heavily used to support journalistic analysis and media accountability, yet little is known about how bias detection models arrive at their decisions or why they fail. In this work, we present a comparative interpretability study of two transformer-based bias detection models: a... | Explaining News Bias Detection: A Comparative SHAP Analysis of Transformer Model Decision Mechanisms. Automated bias detection in news text is heavily used to support journalistic analysis and media accountability, yet little is known about how bias detection models arrive at their decisions or why they fail. In this w... | cl_nlp | cs.CL | 2025-12-29T19:58:11+00:00 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.