id
string
title
string
abstract
string
text
string
label
string
primary_category
string
published
string
2512.15710v1
Artism: AI-Driven Dual-Engine System for Art Generation and Critique
This paper proposes a dual-engine AI architectural method designed to address the complex problem of exploring potential trajectories in the evolution of art. We present two interconnected components: AIDA (an artificial artist social network) and the Ismism Machine, a system for critical analysis. The core innovation ...
Artism: AI-Driven Dual-Engine System for Art Generation and Critique. This paper proposes a dual-engine AI architectural method designed to address the complex problem of exploring potential trajectories in the evolution of art. We present two interconnected components: AIDA (an artificial artist social network) and th...
ai_general
cs.AI
2025-12-17T18:58:42+00:00
2512.00772v1
SHRAG: AFrameworkfor Combining Human-Inspired Search with RAG
Retrieval-Augmented Generation (RAG) is gaining recognition as one of the key technological axes for next generation information retrieval, owing to its ability to mitigate the hallucination phenomenon in Large Language Models (LLMs)and effectively incorporate up-to-date information. However, specialized expertise is n...
SHRAG: AFrameworkfor Combining Human-Inspired Search with RAG. Retrieval-Augmented Generation (RAG) is gaining recognition as one of the key technological axes for next generation information retrieval, owing to its ability to mitigate the hallucination phenomenon in Large Language Models (LLMs)and effectively incorpor...
ir_retrieval
cs.IR
2025-11-30T08:06:47+00:00
2510.03984v1
Beyond Static Evaluation: Rethinking the Assessment of Personalized Agent Adaptability in Information Retrieval
Personalized AI agents are becoming central to modern information retrieval, yet most evaluation methodologies remain static, relying on fixed benchmarks and one-off metrics that fail to reflect how users' needs evolve over time. These limitations hinder our ability to assess whether agents can meaningfully adapt to in...
Beyond Static Evaluation: Rethinking the Assessment of Personalized Agent Adaptability in Information Retrieval. Personalized AI agents are becoming central to modern information retrieval, yet most evaluation methodologies remain static, relying on fixed benchmarks and one-off metrics that fail to reflect how users' n...
ir_retrieval
cs.IR
2025-10-05T00:35:37+00:00
2512.21002v2
Distilling the Essence: Efficient Reasoning Distillation via Sequence Truncation
Distilling the capabilities from a large reasoning model (LRM) to a smaller student model often involves training on substantial amounts of reasoning data. However, knowledge distillation (KD) over lengthy sequences with prompt (P), chain-of-thought (CoT), and answer (A) sections makes the process computationally expen...
Distilling the Essence: Efficient Reasoning Distillation via Sequence Truncation. Distilling the capabilities from a large reasoning model (LRM) to a smaller student model often involves training on substantial amounts of reasoning data. However, knowledge distillation (KD) over lengthy sequences with prompt (P), chain...
cl_nlp
cs.CL
2025-12-24T06:57:35+00:00
2512.20194v1
Generative Latent Coding for Ultra-Low Bitrate Image Compression
Most existing image compression approaches perform transform coding in the pixel space to reduce its spatial redundancy. However, they encounter difficulties in achieving both high-realism and high-fidelity at low bitrate, as the pixel-space distortion may not align with human perception. To address this issue, we intr...
Generative Latent Coding for Ultra-Low Bitrate Image Compression. Most existing image compression approaches perform transform coding in the pixel space to reduce its spatial redundancy. However, they encounter difficulties in achieving both high-realism and high-fidelity at low bitrate, as the pixel-space distortion m...
cv_vision
cs.CV
2025-12-23T09:35:40+00:00
2511.05452v2
Self-adaptive weighting and sampling for physics-informed neural networks
Physics-informed deep learning has emerged as a promising framework for solving partial differential equations (PDEs). Nevertheless, training these models on complex problems remains challenging, often leading to limited accuracy and efficiency. In this work, we introduce a hybrid adaptive sampling and weighting method...
Self-adaptive weighting and sampling for physics-informed neural networks. Physics-informed deep learning has emerged as a promising framework for solving partial differential equations (PDEs). Nevertheless, training these models on complex problems remains challenging, often leading to limited accuracy and efficiency....
stat_ml
stat.ML
2025-11-07T17:48:11+00:00
2512.24040v1
ROAD: Reflective Optimization via Automated Debugging for Zero-Shot Agent Alignment
Automatic Prompt Optimization (APO) has emerged as a critical technique for enhancing Large Language Model (LLM) performance, yet current state-of-the-art methods typically rely on large, labeled gold-standard development sets to compute fitness scores for evolutionary or Reinforcement Learning (RL) approaches. In real...
ROAD: Reflective Optimization via Automated Debugging for Zero-Shot Agent Alignment. Automatic Prompt Optimization (APO) has emerged as a critical technique for enhancing Large Language Model (LLM) performance, yet current state-of-the-art methods typically rely on large, labeled gold-standard development sets to compu...
ai_general
cs.AI
2025-12-30T07:31:34+00:00
2512.12735v1
Limits To (Machine) Learning
Machine learning (ML) methods are highly flexible, but their ability to approximate the true data-generating process is fundamentally constrained by finite samples. We characterize a universal lower bound, the Limits-to-Learning Gap (LLG), quantifying the unavoidable discrepancy between a model's empirical fit and the ...
Limits To (Machine) Learning. Machine learning (ML) methods are highly flexible, but their ability to approximate the true data-generating process is fundamentally constrained by finite samples. We characterize a universal lower bound, the Limits-to-Learning Gap (LLG), quantifying the unavoidable discrepancy between a ...
stat_ml
stat.ML
2025-12-14T15:18:45+00:00
2512.07306v1
Exact Synthetic Populations for Scalable Societal and Market Modeling
We introduce a constraint-programming framework for generating synthetic populations that reproduce target statistics with high precision while enforcing full individual consistency. Unlike data-driven approaches that infer distributions from samples, our method directly encodes aggregated statistics and structural rel...
Exact Synthetic Populations for Scalable Societal and Market Modeling. We introduce a constraint-programming framework for generating synthetic populations that reproduce target statistics with high precision while enforcing full individual consistency. Unlike data-driven approaches that infer distributions from sample...
stat_ml
stat.ML
2025-12-08T08:48:21+00:00
2512.03607v1
DeepRule: An Integrated Framework for Automated Business Rule Generation via Deep Predictive Modeling and Hybrid Search Optimization
This paper proposes DeepRule, an integrated framework for automated business rule generation in retail assortment and pricing optimization. Addressing the systematic misalignment between existing theoretical models and real-world economic complexities, we identify three critical gaps: (1) data modality mismatch where u...
DeepRule: An Integrated Framework for Automated Business Rule Generation via Deep Predictive Modeling and Hybrid Search Optimization. This paper proposes DeepRule, an integrated framework for automated business rule generation in retail assortment and pricing optimization. Addressing the systematic misalignment between...
ai_general
cs.AI
2025-12-03T09:40:33+00:00
2512.20901v1
Benchmarking and Enhancing VLM for Compressed Image Understanding
With the rapid development of Vision-Language Models (VLMs) and the growing demand for their applications, efficient compression of the image inputs has become increasingly important. Existing VLMs predominantly digest and understand high-bitrate compressed images, while their ability to interpret low-bitrate compresse...
Benchmarking and Enhancing VLM for Compressed Image Understanding. With the rapid development of Vision-Language Models (VLMs) and the growing demand for their applications, efficient compression of the image inputs has become increasingly important. Existing VLMs predominantly digest and understand high-bitrate compre...
cv_vision
cs.CV
2025-12-24T02:59:01+00:00
2512.10545v1
XDoGE: Multilingual Data Reweighting to Enhance Language Inclusivity in LLMs
Current large language models (LLMs) are trained on massive amounts of text data, primarily from a few dominant languages. Studies suggest that this over-reliance on high-resource languages, such as English, hampers LLM performance in mid- and low-resource languages. To mitigate this problem, we propose to (i) optimize...
XDoGE: Multilingual Data Reweighting to Enhance Language Inclusivity in LLMs. Current large language models (LLMs) are trained on massive amounts of text data, primarily from a few dominant languages. Studies suggest that this over-reliance on high-resource languages, such as English, hampers LLM performance in mid- an...
cl_nlp
cs.CL
2025-12-11T11:22:53+00:00
2512.05967v2
Enhancing Retrieval-Augmented Generation with Entity Linking for Educational Platforms
In the era of Large Language Models (LLMs), Retrieval-Augmented Generation (RAG) architectures are gaining significant attention for their ability to ground language generation in reliable knowledge sources. Despite their effectiveness, RAG systems based solely on semantic similarity often fail to ensure factual accura...
Enhancing Retrieval-Augmented Generation with Entity Linking for Educational Platforms. In the era of Large Language Models (LLMs), Retrieval-Augmented Generation (RAG) architectures are gaining significant attention for their ability to ground language generation in reliable knowledge sources. Despite their effectiven...
ir_retrieval
cs.IR
2025-12-05T18:59:18+00:00
2510.25514v1
Convergence of off-policy TD(0) with linear function approximation for reversible Markov chains
We study the convergence of off-policy TD(0) with linear function approximation when used to approximate the expected discounted reward in a Markov chain. It is well known that the combination of off-policy learning and function approximation can lead to divergence of the algorithm. Existing results for this setting mo...
Convergence of off-policy TD(0) with linear function approximation for reversible Markov chains. We study the convergence of off-policy TD(0) with linear function approximation when used to approximate the expected discounted reward in a Markov chain. It is well known that the combination of off-policy learning and fun...
stat_ml
stat.ML
2025-10-29T13:38:24+00:00
2512.17630v1
Confidence-Credibility Aware Weighted Ensembles of Small LLMs Outperform Large LLMs in Emotion Detection
This paper introduces a confidence-weighted, credibility-aware ensemble framework for text-based emotion detection, inspired by Condorcet's Jury Theorem (CJT). Unlike conventional ensembles that often rely on homogeneous architectures, our approach combines architecturally diverse small transformer-based large language...
Confidence-Credibility Aware Weighted Ensembles of Small LLMs Outperform Large LLMs in Emotion Detection. This paper introduces a confidence-weighted, credibility-aware ensemble framework for text-based emotion detection, inspired by Condorcet's Jury Theorem (CJT). Unlike conventional ensembles that often rely on homog...
cl_nlp
cs.CL
2025-12-19T14:33:14+00:00
2511.06937v1
Fine-Tuning Diffusion-Based Recommender Systems via Reinforcement Learning with Reward Function Optimization
Diffusion models recently emerged as a powerful paradigm for recommender systems, offering state-of-the-art performance by modeling the generative process of user-item interactions. However, training such models from scratch is both computationally expensive and yields diminishing returns once convergence is reached. T...
Fine-Tuning Diffusion-Based Recommender Systems via Reinforcement Learning with Reward Function Optimization. Diffusion models recently emerged as a powerful paradigm for recommender systems, offering state-of-the-art performance by modeling the generative process of user-item interactions. However, training such model...
ir_retrieval
cs.IR
2025-11-10T10:38:16+00:00
2509.15127v1
Learning Rate Should Scale Inversely with High-Order Data Moments in High-Dimensional Online Independent Component Analysis
We investigate the impact of high-order moments on the learning dynamics of an online Independent Component Analysis (ICA) algorithm under a high-dimensional data model composed of a weighted sum of two non-Gaussian random variables. This model allows precise control of the input moment structure via a weighting parame...
Learning Rate Should Scale Inversely with High-Order Data Moments in High-Dimensional Online Independent Component Analysis. We investigate the impact of high-order moments on the learning dynamics of an online Independent Component Analysis (ICA) algorithm under a high-dimensional data model composed of a weighted sum...
stat_ml
stat.ML
2025-09-18T16:34:59+00:00
2512.22416v2
Hallucination Detection and Evaluation of Large Language Model
Hallucinations in Large Language Models (LLMs) pose a significant challenge, generating misleading or unverifiable content that undermines trust and reliability. Existing evaluation methods, such as KnowHalu, employ multi-stage verification but suffer from high computational costs. To address this, we integrate the Hug...
Hallucination Detection and Evaluation of Large Language Model. Hallucinations in Large Language Models (LLMs) pose a significant challenge, generating misleading or unverifiable content that undermines trust and reliability. Existing evaluation methods, such as KnowHalu, employ multi-stage verification but suffer from...
cl_nlp
cs.CL
2025-12-27T00:17:03+00:00
2510.10870v2
Transfer Learning with Distance Covariance for Random Forest: Error Bounds and an EHR Application
We propose a method for transfer learning in nonparametric regression using a random forest (RF) with distance covariance-based feature weights, assuming the unknown source and target regression functions are sparsely different. Our method obtains residuals from a source domain-trained Centered RF (CRF) in the target d...
Transfer Learning with Distance Covariance for Random Forest: Error Bounds and an EHR Application. We propose a method for transfer learning in nonparametric regression using a random forest (RF) with distance covariance-based feature weights, assuming the unknown source and target regression functions are sparsely dif...
stat_ml
stat.ML
2025-10-13T00:31:56+00:00
2512.20578v2
Can LLMs Predict Their Own Failures? Self-Awareness via Internal Circuits
Large language models (LLMs) generate fluent and complex outputs but often fail to recognize their own mistakes and hallucinations. Existing approaches typically rely on external judges, multi-sample consistency, or text-based self-critique, which incur additional compute or correlate weakly with true correctness. We a...
Can LLMs Predict Their Own Failures? Self-Awareness via Internal Circuits. Large language models (LLMs) generate fluent and complex outputs but often fail to recognize their own mistakes and hallucinations. Existing approaches typically rely on external judges, multi-sample consistency, or text-based self-critique, whi...
cl_nlp
cs.CL
2025-12-23T18:21:32+00:00
2512.24834v1
GenZ: Foundational models as latent variable generators within traditional statistical models
We present GenZ, a hybrid model that bridges foundational models and statistical modeling through interpretable semantic features. While large language models possess broad domain knowledge, they often fail to capture dataset-specific patterns critical for prediction tasks. Our approach addresses this by discovering se...
GenZ: Foundational models as latent variable generators within traditional statistical models. We present GenZ, a hybrid model that bridges foundational models and statistical modeling through interpretable semantic features. While large language models possess broad domain knowledge, they often fail to capture dataset...
ai_general
cs.AI
2025-12-31T12:56:01+00:00
2601.05258v1
From Events to Trending: A Multi-Stage Hotspots Detection Method Based on Generative Query Indexing
LLM-based conversational systems have become a popular gateway for information access, yet most existing chatbots struggle to handle news-related trending queries effectively. To improve user experience, an effective trending query detection method is urgently needed to enable differentiated processing of such target t...
From Events to Trending: A Multi-Stage Hotspots Detection Method Based on Generative Query Indexing. LLM-based conversational systems have become a popular gateway for information access, yet most existing chatbots struggle to handle news-related trending queries effectively. To improve user experience, an effective tr...
ir_retrieval
cs.IR
2025-10-24T08:49:38+00:00
2512.22336v1
Agent2World: Learning to Generate Symbolic World Models via Adaptive Multi-Agent Feedback
Symbolic world models (e.g., PDDL domains or executable simulators) are central to model-based planning, but training LLMs to generate such world models is limited by the lack of large-scale verifiable supervision. Current approaches rely primarily on static validation methods that fail to catch behavior-level errors a...
Agent2World: Learning to Generate Symbolic World Models via Adaptive Multi-Agent Feedback. Symbolic world models (e.g., PDDL domains or executable simulators) are central to model-based planning, but training LLMs to generate such world models is limited by the lack of large-scale verifiable supervision. Current approa...
ai_general
cs.AI
2025-12-26T18:54:14+00:00
2510.25622v3
Taming the Long Tail: Denoising Collaborative Information for Robust Semantic ID Generation
Item IDs form the backbone of industrial recommender systems, but suffer from representation instability and poor long-tail generalization in large, dynamic item corpora. Semantic IDs (SIDs) mitigate these issues by enabling knowledge sharing through quantization of item content features. Existing methods attempt to en...
Taming the Long Tail: Denoising Collaborative Information for Robust Semantic ID Generation. Item IDs form the backbone of industrial recommender systems, but suffer from representation instability and poor long-tail generalization in large, dynamic item corpora. Semantic IDs (SIDs) mitigate these issues by enabling kn...
ir_retrieval
cs.IR
2025-10-29T15:27:23+00:00
2512.21749v1
Approximation Capabilities of Feedforward Neural Networks with GELU Activations
We derive an approximation error bound that holds simultaneously for a function and all its derivatives up to any prescribed order. The bounds apply to elementary functions, including multivariate polynomials, the exponential function, and the reciprocal function, and are obtained using feedforward neural networks with...
Approximation Capabilities of Feedforward Neural Networks with GELU Activations. We derive an approximation error bound that holds simultaneously for a function and all its derivatives up to any prescribed order. The bounds apply to elementary functions, including multivariate polynomials, the exponential function, and...
lg_ml
cs.LG
2025-12-25T17:56:44+00:00
2509.06002v1
A Survey of Real-World Recommender Systems: Challenges, Constraints, and Industrial Perspectives
Recommender systems have generated tremendous value for both users and businesses, drawing significant attention from academia and industry alike. However, due to practical constraints, academic research remains largely confined to offline dataset optimizations, lacking access to real user data and large-scale recommen...
A Survey of Real-World Recommender Systems: Challenges, Constraints, and Industrial Perspectives. Recommender systems have generated tremendous value for both users and businesses, drawing significant attention from academia and industry alike. However, due to practical constraints, academic research remains largely co...
ir_retrieval
cs.IR
2025-09-07T10:29:41+00:00
2512.22334v3
SciEvalKit: An Open-source Evaluation Toolkit for Scientific General Intelligence
We introduce SciEvalKit, a unified benchmarking toolkit designed to evaluate AI models for science across a broad range of scientific disciplines and task capabilities. Unlike general-purpose evaluation platforms, SciEvalKit focuses on the core competencies of scientific intelligence, including Scientific Multimodal Pe...
SciEvalKit: An Open-source Evaluation Toolkit for Scientific General Intelligence. We introduce SciEvalKit, a unified benchmarking toolkit designed to evaluate AI models for science across a broad range of scientific disciplines and task capabilities. Unlike general-purpose evaluation platforms, SciEvalKit focuses on t...
ai_general
cs.AI
2025-12-26T17:36:02+00:00
2512.25066v2
From Inpainting to Editing: Unlocking Robust Mask-Free Visual Dubbing via Generative Bootstrapping
Audio-driven visual dubbing aims to synchronize a video's lip movements with new speech but is fundamentally challenged by the lack of ideal training data: paired videos differing only in lip motion. Existing methods circumvent this via mask-based inpainting. However, masking inevitably destroys spatiotemporal context,...
From Inpainting to Editing: Unlocking Robust Mask-Free Visual Dubbing via Generative Bootstrapping. Audio-driven visual dubbing aims to synchronize a video's lip movements with new speech but is fundamentally challenged by the lack of ideal training data: paired videos differing only in lip motion. Existing methods cir...
cv_vision
cs.CV
2025-12-31T18:58:30+00:00
2510.12416v7
Geopolitics, Geoeconomics, and Sovereign Risk: Different Shocks, Different Channels
Geopolitical and geoeconomic shocks reprice sovereign credit risk through different transmission channels. Using a daily panel of 42 advanced and emerging economies over 2018--2025, we show that geopolitical shocks raise sovereign CDS spreads primarily through direct sovereign repricing, while the Global Financial Cycl...
Geopolitics, Geoeconomics, and Sovereign Risk: Different Shocks, Different Channels. Geopolitical and geoeconomic shocks reprice sovereign credit risk through different transmission channels. Using a daily panel of 42 advanced and emerging economies over 2018--2025, we show that geopolitical shocks raise sovereign CDS ...
stat_ml
stat.ML
2025-10-14T11:51:36+00:00
2512.15089v2
Beyond Fast and Slow: Cognitive-Inspired Elastic Reasoning for Large Language Models
Large language models (LLMs) have demonstrated impressive performance across various language tasks. However, existing LLM reasoning strategies mainly rely on the LLM itself with fast or slow mode (like o1 thinking) and thus struggle to balance reasoning efficiency and accuracy across queries of varying difficulties. I...
Beyond Fast and Slow: Cognitive-Inspired Elastic Reasoning for Large Language Models. Large language models (LLMs) have demonstrated impressive performance across various language tasks. However, existing LLM reasoning strategies mainly rely on the LLM itself with fast or slow mode (like o1 thinking) and thus struggle ...
ai_general
cs.AI
2025-12-17T05:11:58+00:00
2510.06658v2
Can We Hide Machines in the Crowd? Quantifying Equivalence in LLM-in-the-loop Annotation Tasks
Many evaluations of large language models (LLMs) in text annotation focus primarily on the correctness of the output, typically comparing model-generated labels to human-annotated ``ground truth'' using standard performance metrics. In contrast, our study moves beyond effectiveness alone. We aim to explore how labeling...
Can We Hide Machines in the Crowd? Quantifying Equivalence in LLM-in-the-loop Annotation Tasks. Many evaluations of large language models (LLMs) in text annotation focus primarily on the correctness of the output, typically comparing model-generated labels to human-annotated ``ground truth'' using standard performance ...
ir_retrieval
cs.IR
2025-10-08T05:17:33+00:00
2511.13201v2
Cog-RAG: Cognitive-Inspired Dual-Hypergraph with Theme Alignment Retrieval-Augmented Generation
Retrieval-Augmented Generation (RAG) enhances the response quality and domain-specific performance of large language models (LLMs) by incorporating external knowledge to combat hallucinations. In recent research, graph structures have been integrated into RAG to enhance the capture of semantic relations between entitie...
Cog-RAG: Cognitive-Inspired Dual-Hypergraph with Theme Alignment Retrieval-Augmented Generation. Retrieval-Augmented Generation (RAG) enhances the response quality and domain-specific performance of large language models (LLMs) by incorporating external knowledge to combat hallucinations. In recent research, graph stru...
ir_retrieval
cs.IR
2025-11-17T10:10:33+00:00
2510.09513v1
Interpretable Generative and Discriminative Learning for Multimodal and Incomplete Clinical Data
Real-world clinical problems are often characterized by multimodal data, usually associated with incomplete views and limited sample sizes in their cohorts, posing significant limitations for machine learning algorithms. In this work, we propose a Bayesian approach designed to efficiently handle these challenges while ...
Interpretable Generative and Discriminative Learning for Multimodal and Incomplete Clinical Data. Real-world clinical problems are often characterized by multimodal data, usually associated with incomplete views and limited sample sizes in their cohorts, posing significant limitations for machine learning algorithms. I...
stat_ml
stat.ML
2025-10-10T16:20:24+00:00
2512.05033v2
Arbitrage: Efficient Reasoning via Advantage-Aware Speculation
Modern Large Language Models achieve impressive reasoning capabilities with long Chain of Thoughts, but they incur substantial computational cost during inference, and this motivates techniques to improve the performance-cost ratio. Among these techniques, Speculative Decoding accelerates inference by employing a fast ...
Arbitrage: Efficient Reasoning via Advantage-Aware Speculation. Modern Large Language Models achieve impressive reasoning capabilities with long Chain of Thoughts, but they incur substantial computational cost during inference, and this motivates techniques to improve the performance-cost ratio. Among these techniques,...
cl_nlp
cs.CL
2025-12-04T17:50:53+00:00
2512.19323v1
Alternative positional encoding functions for neural transformers
A key module in neural transformer-based deep architectures is positional encoding. This module enables a suitable way to encode positional information as input for transformer neural layers. This success has been rooted in the use of sinusoidal functions of various frequencies, in order to capture recurrent patterns o...
Alternative positional encoding functions for neural transformers. A key module in neural transformer-based deep architectures is positional encoding. This module enables a suitable way to encode positional information as input for transformer neural layers. This success has been rooted in the use of sinusoidal functio...
lg_ml
cs.LG
2025-12-22T12:17:47+00:00
2512.05681v1
Retrieving Semantically Similar Decisions under Noisy Institutional Labels: Robust Comparison of Embedding Methods
Retrieving case law is a time-consuming task predominantly carried out by querying databases. We provide a comparison of two models in three different settings for Czech Constitutional Court decisions: (i) a large general-purpose embedder (OpenAI), (ii) a domain-specific BERT-trained from scratch on ~30,000 decisions u...
Retrieving Semantically Similar Decisions under Noisy Institutional Labels: Robust Comparison of Embedding Methods. Retrieving case law is a time-consuming task predominantly carried out by querying databases. We provide a comparison of two models in three different settings for Czech Constitutional Court decisions: (i...
cl_nlp
cs.CL
2025-12-05T12:54:26+00:00
2601.00854v1
Motion-Compensated Latent Semantic Canvases for Visual Situational Awareness on Edge
We propose Motion-Compensated Latent Semantic Canvases (MCLSC) for visual situational awareness on resource-constrained edge devices. The core idea is to maintain persistent semantic metadata in two latent canvases - a slowly accumulating static layer and a rapidly updating dynamic layer - defined in a baseline coordin...
Motion-Compensated Latent Semantic Canvases for Visual Situational Awareness on Edge. We propose Motion-Compensated Latent Semantic Canvases (MCLSC) for visual situational awareness on resource-constrained edge devices. The core idea is to maintain persistent semantic metadata in two latent canvases - a slowly accumula...
cv_vision
cs.CV
2025-12-29T20:25:02+00:00
2512.17267v1
AutoMetrics: Approximate Human Judgements with Automatically Generated Evaluators
Evaluating user-facing AI applications remains a central challenge, especially in open-ended domains such as travel planning, clinical note generation, or dialogue. The gold standard is user feedback (e.g., thumbs up/down) or behavioral signals (e.g., retention), but these are often scarce in prototypes and research pr...
AutoMetrics: Approximate Human Judgements with Automatically Generated Evaluators. Evaluating user-facing AI applications remains a central challenge, especially in open-ended domains such as travel planning, clinical note generation, or dialogue. The gold standard is user feedback (e.g., thumbs up/down) or behavioral ...
cl_nlp
cs.CL
2025-12-19T06:32:46+00:00
2512.24404v1
Lifting Vision: Ground to Aerial Localization with Reasoning Guided Planning
Multimodal intelligence development recently show strong progress in visual understanding and high level reasoning. Though, most reasoning system still reply on textual information as the main medium for inference. This limit their effectiveness in spatial tasks such as visual navigation and geo-localization. This work...
Lifting Vision: Ground to Aerial Localization with Reasoning Guided Planning. Multimodal intelligence development recently show strong progress in visual understanding and high level reasoning. Though, most reasoning system still reply on textual information as the main medium for inference. This limit their effectiven...
lg_ml
cs.LG
2025-12-30T18:36:39+00:00
2512.13564v2
Memory in the Age of AI Agents
Memory has emerged, and will continue to remain, a core capability of foundation model-based agents. As research on agent memory rapidly expands and attracts unprecedented attention, the field has also become increasingly fragmented. Existing works that fall under the umbrella of agent memory often differ substantially...
Memory in the Age of AI Agents. Memory has emerged, and will continue to remain, a core capability of foundation model-based agents. As research on agent memory rapidly expands and attracts unprecedented attention, the field has also become increasingly fragmented. Existing works that fall under the umbrella of agent m...
cl_nlp
cs.CL
2025-12-15T17:22:34+00:00
2512.08629v1
See-Control: A Multimodal Agent Framework for Smartphone Interaction with a Robotic Arm
Recent advances in Multimodal Large Language Models (MLLMs) have enabled their use as intelligent agents for smartphone operation. However, existing methods depend on the Android Debug Bridge (ADB) for data transmission and action execution, limiting their applicability to Android devices. In this work, we introduce th...
See-Control: A Multimodal Agent Framework for Smartphone Interaction with a Robotic Arm. Recent advances in Multimodal Large Language Models (MLLMs) have enabled their use as intelligent agents for smartphone operation. However, existing methods depend on the Android Debug Bridge (ADB) for data transmission and action ...
ai_general
cs.AI
2025-12-09T14:14:37+00:00
2509.07531v1
FLeW: Facet-Level and Adaptive Weighted Representation Learning of Scientific Documents
Scientific document representation learning provides powerful embeddings for various tasks, while current methods face challenges across three approaches. 1) Contrastive training with citation-structural signals underutilizes citation information and still generates single-vector representations. 2) Fine-grained repres...
FLeW: Facet-Level and Adaptive Weighted Representation Learning of Scientific Documents. Scientific document representation learning provides powerful embeddings for various tasks, while current methods face challenges across three approaches. 1) Contrastive training with citation-structural signals underutilizes citat...
ir_retrieval
cs.IR
2025-09-09T09:08:44+00:00
2511.02706v1
Optimizing Kernel Discrepancies via Subset Selection
Kernel discrepancies are a powerful tool for analyzing worst-case errors in quasi-Monte Carlo (QMC) methods. Building on recent advances in optimizing such discrepancy measures, we extend the subset selection problem to the setting of kernel discrepancies, selecting an m-element subset from a large population of size $...
Optimizing Kernel Discrepancies via Subset Selection. Kernel discrepancies are a powerful tool for analyzing worst-case errors in quasi-Monte Carlo (QMC) methods. Building on recent advances in optimizing such discrepancy measures, we extend the subset selection problem to the setting of kernel discrepancies, selecting...
stat_ml
stat.ML
2025-11-04T16:25:08+00:00
2509.00258v1
Assessing One-Dimensional Cluster Stability by Extreme-Point Trimming
We develop a probabilistic method for assessing the tail behavior and geometric stability of one-dimensional n i.i.d. samples by tracking how their span contracts when the most extreme points are trimmed. Central to our approach is the diameter-shrinkage ratio, that quantifies the relative reduction in data range as ex...
Assessing One-Dimensional Cluster Stability by Extreme-Point Trimming. We develop a probabilistic method for assessing the tail behavior and geometric stability of one-dimensional n i.i.d. samples by tracking how their span contracts when the most extreme points are trimmed. Central to our approach is the diameter-shri...
stat_ml
stat.ML
2025-08-29T21:52:15+00:00
2512.00610v1
Statistical-computational gap in multiple Gaussian graph alignment
We investigate the existence of a statistical-computational gap in multiple Gaussian graph alignment. We first generalize a previously established informational threshold from Vassaux and Massoulié (2025) to regimes where the number of observed graphs $p$ may also grow with the number of nodes $n$: when $p \leq O(n/\lo...
Statistical-computational gap in multiple Gaussian graph alignment. We investigate the existence of a statistical-computational gap in multiple Gaussian graph alignment. We first generalize a previously established informational threshold from Vassaux and Massoulié (2025) to regimes where the number of observed graphs ...
stat_ml
stat.ML
2025-11-29T19:52:12+00:00
2512.22193v1
Tiny-YOLOSAM: Fast Hybrid Image Segmentation
The Segment Anything Model (SAM) enables promptable, high-quality segmentation but is often too computationally expensive for latency-critical settings. TinySAM is a lightweight, distilled SAM variant that preserves strong zero-shot mask quality, yet its "segment-everything" mode still requires hundreds of prompts and ...
Tiny-YOLOSAM: Fast Hybrid Image Segmentation. The Segment Anything Model (SAM) enables promptable, high-quality segmentation but is often too computationally expensive for latency-critical settings. TinySAM is a lightweight, distilled SAM variant that preserves strong zero-shot mask quality, yet its "segment-everything...
cv_vision
cs.CV
2025-12-20T12:28:39+00:00
2509.21584v4
IndiSeek learns information-guided disentangled representations
Learning disentangled representations is a fundamental task in multi-modal learning. In modern applications such as single-cell multi-omics, both shared and modality-specific features are critical for characterizing cell states and supporting downstream analyses. Ideally, modality-specific features should be independen...
IndiSeek learns information-guided disentangled representations. Learning disentangled representations is a fundamental task in multi-modal learning. In modern applications such as single-cell multi-omics, both shared and modality-specific features are critical for characterizing cell states and supporting downstream a...
stat_ml
stat.ML
2025-09-25T20:58:34+00:00
2512.19957v1
Zero-Shot Segmentation through Prototype-Guidance for Multi-Label Plant Species Identification
This paper presents an approach developed to address the PlantClef 2025 challenge, which consists of a fine-grained multi-label species identification, over high-resolution images. Our solution focused on employing class prototypes obtained from the training dataset as a proxy guidance for training a segmentation Visio...
Zero-Shot Segmentation through Prototype-Guidance for Multi-Label Plant Species Identification. This paper presents an approach developed to address the PlantClef 2025 challenge, which consists of a fine-grained multi-label species identification, over high-resolution images. Our solution focused on employing class pro...
ai_general
cs.AI
2025-12-23T01:06:55+00:00
2512.21948v2
Interpretable Machine Learning-Derived Spectral Indices for Vegetation Monitoring
Spectral indices such as NDVI have driven vegetation monitoring for decades, yet their design remains largely manual and ad hoc. Their usefulness stems not only from their empirical performance, but also from algebraic forms that remain compact and biologically interpretable. However, the space of possible algebraic ex...
Interpretable Machine Learning-Derived Spectral Indices for Vegetation Monitoring. Spectral indices such as NDVI have driven vegetation monitoring for decades, yet their design remains largely manual and ad hoc. Their usefulness stems not only from their empirical performance, but also from algebraic forms that remain ...
cv_vision
cs.CV
2025-12-26T09:48:41+00:00
2512.20812v1
Semantic Deception: When Reasoning Models Can't Compute an Addition
Large language models (LLMs) are increasingly used in situations where human values are at stake, such as decision-making tasks that involve reasoning when performed by humans. We investigate the so-called reasoning capabilities of LLMs over novel symbolic representations by introducing an experimental framework that t...
Semantic Deception: When Reasoning Models Can't Compute an Addition. Large language models (LLMs) are increasingly used in situations where human values are at stake, such as decision-making tasks that involve reasoning when performed by humans. We investigate the so-called reasoning capabilities of LLMs over novel sym...
cl_nlp
cs.CL
2025-12-23T22:22:18+00:00
2510.25753v2
How Data Mixing Shapes In-Context Learning: Asymptotic Equivalence for Transformers with MLPs
Pretrained Transformers demonstrate remarkable in-context learning (ICL) capabilities, enabling them to adapt to new tasks from demonstrations without parameter updates. However, theoretical studies often rely on simplified architectures (e.g., omitting MLPs), plain data models (e.g., linear regression with isotropic i...
How Data Mixing Shapes In-Context Learning: Asymptotic Equivalence for Transformers with MLPs. Pretrained Transformers demonstrate remarkable in-context learning (ICL) capabilities, enabling them to adapt to new tasks from demonstrations without parameter updates. However, theoretical studies often rely on simplified a...
stat_ml
stat.ML
2025-10-29T17:51:57+00:00
2512.08868v2
EcomBench: Towards Holistic Evaluation of Foundation Agents in E-commerce
Foundation agents have rapidly advanced in their ability to reason and interact with real environments, making the evaluation of their core capabilities increasingly important. While many benchmarks have been developed to assess agent performance, most concentrate on academic settings or artificially designed scenarios...
EcomBench: Towards Holistic Evaluation of Foundation Agents in E-commerce. Foundation agents have rapidly advanced in their ability to reason and interact with real environments, making the evaluation of their core capabilities increasingly important. While many benchmarks have been developed to assess agent performanc...
ai_general
cs.AI
2025-12-09T18:00:26+00:00
2510.13738v2
HyMiRec: A Hybrid Multi-interest Learning Framework for LLM-based Sequential Recommendation
Large language models (LLMs) have recently demonstrated strong potential for sequential recommendation. However, current LLM-based approaches face critical limitations in modeling users' long-term and diverse interests. First, due to inference latency and feature fetching bandwidth constraints, existing methods typical...
HyMiRec: A Hybrid Multi-interest Learning Framework for LLM-based Sequential Recommendation. Large language models (LLMs) have recently demonstrated strong potential for sequential recommendation. However, current LLM-based approaches face critical limitations in modeling users' long-term and diverse interests. First, ...
ir_retrieval
cs.IR
2025-10-15T16:45:59+00:00
2511.23310v2
OBLR-PO: A Theoretical Framework for Stable Reinforcement Learning
Existing reinforcement learning (RL)-based post-training methods for large language models have advanced rapidly, yet their design has largely been guided by heuristics rather than systematic theoretical principles. This gap limits our understanding of the properties of the gradient estimators and the associated optimi...
OBLR-PO: A Theoretical Framework for Stable Reinforcement Learning. Existing reinforcement learning (RL)-based post-training methods for large language models have advanced rapidly, yet their design has largely been guided by heuristics rather than systematic theoretical principles. This gap limits our understanding of...
stat_ml
stat.ML
2025-11-28T16:09:28+00:00
2511.11649v1
The Environmental Impact of Ensemble Techniques in Recommender Systems
Ensemble techniques in recommender systems have demonstrated accuracy improvements of 10-30%, yet their environmental impact remains unmeasured. While deep learning recommendation algorithms can generate up to 3,297 kg CO2 per paper, ensemble methods have not been sufficiently evaluated for energy consumption. This the...
The Environmental Impact of Ensemble Techniques in Recommender Systems. Ensemble techniques in recommender systems have demonstrated accuracy improvements of 10-30%, yet their environmental impact remains unmeasured. While deep learning recommendation algorithms can generate up to 3,297 kg CO2 per paper, ensemble metho...
ir_retrieval
cs.IR
2025-11-10T14:06:58+00:00
2512.24592v1
SliceLens: Fine-Grained and Grounded Error Slice Discovery for Multi-Instance Vision Tasks
Systematic failures of computer vision models on subsets with coherent visual patterns, known as error slices, pose a critical challenge for robust model evaluation. Existing slice discovery methods are primarily developed for image classification, limiting their applicability to multi-instance tasks such as detection,...
SliceLens: Fine-Grained and Grounded Error Slice Discovery for Multi-Instance Vision Tasks. Systematic failures of computer vision models on subsets with coherent visual patterns, known as error slices, pose a critical challenge for robust model evaluation. Existing slice discovery methods are primarily developed for i...
cv_vision
cs.CV
2025-12-31T03:28:41+00:00
2601.05264v1
Engineering the RAG Stack: A Comprehensive Review of the Architecture and Trust Frameworks for Retrieval-Augmented Generation Systems
This article provides a comprehensive systematic literature review of academic studies, industrial applications, and real-world deployments from 2018 to 2025, providing a practical guide and detailed overview of modern Retrieval-Augmented Generation (RAG) architectures. RAG offers a modular approach for integrating ext...
Engineering the RAG Stack: A Comprehensive Review of the Architecture and Trust Frameworks for Retrieval-Augmented Generation Systems. This article provides a comprehensive systematic literature review of academic studies, industrial applications, and real-world deployments from 2018 to 2025, providing a practical guid...
ir_retrieval
cs.IR
2025-11-07T16:26:29+00:00
2512.16739v2
AI-Driven Prediction of Cancer Pain Episodes: A Hybrid Decision Support Approach
Lung cancer patients frequently experience breakthrough pain episodes, with up to 91% requiring timely intervention. To enable proactive pain management, we propose a hybrid machine learning and large language model pipeline that predicts pain episodes within 48 and 72 hours of hospitalization using both structured and...
AI-Driven Prediction of Cancer Pain Episodes: A Hybrid Decision Support Approach. Lung cancer patients frequently experience breakthrough pain episodes, with up to 91% requiring timely intervention. To enable proactive pain management, we propose a hybrid machine learning and large language model pipeline that predicts...
ai_general
cs.AI
2025-12-18T16:37:29+00:00
2512.21080v3
LLM Personas as a Substitute for Field Experiments in Method Benchmarking
Field experiments (A/B tests) are often the most credible benchmark for methods (algorithms) in societal systems, but their cost and latency bottleneck rapid methodological progress. LLM-based persona simulation offers a cheap synthetic alternative, yet it is unclear whether replacing humans with personas preserves the...
LLM Personas as a Substitute for Field Experiments in Method Benchmarking. Field experiments (A/B tests) are often the most credible benchmark for methods (algorithms) in societal systems, but their cost and latency bottleneck rapid methodological progress. LLM-based persona simulation offers a cheap synthetic alternat...
ai_general
cs.AI
2025-12-24T09:56:00+00:00
2509.20134v1
Intelligent Algorithm Selection for Recommender Systems: Meta-Learning via in-depth algorithm feature engineering
The "No Free Lunch" theorem dictates that no single recommender algorithm is optimal for all users, creating a significant Algorithm Selection Problem. Standard meta-learning approaches aim to solve this by selecting an algorithm based on user features, but treat the fundamentally diverse algorithms themselves as equiv...
Intelligent Algorithm Selection for Recommender Systems: Meta-Learning via in-depth algorithm feature engineering. The "No Free Lunch" theorem dictates that no single recommender algorithm is optimal for all users, creating a significant Algorithm Selection Problem. Standard meta-learning approaches aim to solve this b...
ir_retrieval
cs.IR
2025-09-24T14:00:37+00:00
2601.08748v1
UR-Bench: A Benchmark for Multi-Hop Reasoning over Ultra-High-Resolution Images
Recent multimodal large language models (MLLMs) show strong capabilities in visual-language reasoning, yet their performance on ultra-high-resolution imagery remains largely unexplored. Existing visual question answering (VQA) benchmarks typically rely on medium-resolution data, offering limited visual complexity. To b...
UR-Bench: A Benchmark for Multi-Hop Reasoning over Ultra-High-Resolution Images. Recent multimodal large language models (MLLMs) show strong capabilities in visual-language reasoning, yet their performance on ultra-high-resolution imagery remains largely unexplored. Existing visual question answering (VQA) benchmarks t...
cv_vision
cs.CV
2025-12-31T02:22:50+00:00
2601.00038v1
Active learning for data-driven reduced models of parametric differential systems with Bayesian operator inference
This work develops an active learning framework to intelligently enrich data-driven reduced-order models (ROMs) of parametric dynamical systems, which can serve as the foundation of virtual assets in a digital twin. Data-driven ROMs are explainable, computationally efficient scientific machine learning models that aim ...
Active learning for data-driven reduced models of parametric differential systems with Bayesian operator inference. This work develops an active learning framework to intelligently enrich data-driven reduced-order models (ROMs) of parametric dynamical systems, which can serve as the foundation of virtual assets in a di...
stat_ml
stat.ML
2025-12-30T19:34:26+00:00
2512.24861v1
OFL-SAM2: Prompt SAM2 with Online Few-shot Learner for Efficient Medical Image Segmentation
The Segment Anything Model 2 (SAM2) has demonstrated remarkable promptable visual segmentation capabilities in video data, showing potential for extension to medical image segmentation (MIS) tasks involving 3D volumes and temporally correlated 2D image sequences. However, adapting SAM2 to MIS presents several challenge...
OFL-SAM2: Prompt SAM2 with Online Few-shot Learner for Efficient Medical Image Segmentation. The Segment Anything Model 2 (SAM2) has demonstrated remarkable promptable visual segmentation capabilities in video data, showing potential for extension to medical image segmentation (MIS) tasks involving 3D volumes and tempo...
cv_vision
cs.CV
2025-12-31T13:41:16+00:00
2512.19317v1
SafeMed-R1: Adversarial Reinforcement Learning for Generalizable and Robust Medical Reasoning in Vision-Language Models
Vision--Language Models (VLMs) show significant promise for Medical Visual Question Answering (VQA), yet their deployment in clinical settings is hindered by severe vulnerability to adversarial attacks. Standard adversarial training, while effective for simpler tasks, often degrades both generalization performance and ...
SafeMed-R1: Adversarial Reinforcement Learning for Generalizable and Robust Medical Reasoning in Vision-Language Models. Vision--Language Models (VLMs) show significant promise for Medical Visual Question Answering (VQA), yet their deployment in clinical settings is hindered by severe vulnerability to adversarial attac...
ai_general
cs.AI
2025-12-22T12:07:33+00:00
2512.07515v4
TPA: Next Token Probability Attribution for Detecting Hallucinations in RAG
Detecting hallucinations in Retrieval-Augmented Generation remains a challenge. Prior approaches attribute hallucinations to a binary conflict between internal knowledge stored in FFNs and the retrieved context. However, this perspective is incomplete, failing to account for the impact of other components of the LLM, s...
TPA: Next Token Probability Attribution for Detecting Hallucinations in RAG. Detecting hallucinations in Retrieval-Augmented Generation remains a challenge. Prior approaches attribute hallucinations to a binary conflict between internal knowledge stored in FFNs and the retrieved context. However, this perspective is in...
cl_nlp
cs.CL
2025-12-08T12:50:41+00:00
2512.18073v2
FPBench: A Comprehensive Benchmark of Multimodal Large Language Models for Fingerprint Analysis
Multimodal LLMs (MLLMs) are capable of performing complex data analysis, visual question answering, generation, and reasoning tasks. However, their ability to analyze biometric data is relatively underexplored. In this work, we investigate the effectiveness of MLLMs in understanding fine structural and textural details...
FPBench: A Comprehensive Benchmark of Multimodal Large Language Models for Fingerprint Analysis. Multimodal LLMs (MLLMs) are capable of performing complex data analysis, visual question answering, generation, and reasoning tasks. However, their ability to analyze biometric data is relatively underexplored. In this work...
cv_vision
cs.CV
2025-12-19T21:23:57+00:00
2512.24410v1
Comparing Approaches to Automatic Summarization in Less-Resourced Languages
Automatic text summarization has achieved high performance in high-resourced languages like English, but comparatively less attention has been given to summarization in less-resourced languages. This work compares a variety of different approaches to summarization from zero-shot prompting of LLMs large and small to fin...
Comparing Approaches to Automatic Summarization in Less-Resourced Languages. Automatic text summarization has achieved high performance in high-resourced languages like English, but comparatively less attention has been given to summarization in less-resourced languages. This work compares a variety of different approa...
cl_nlp
cs.CL
2025-12-30T18:45:00+00:00
2512.10211v2
ID-PaS+ : Identity-Aware Predict-and-Search for General Mixed-Integer Linear Programs
Mixed-Integer Linear Programs (MIPs) are powerful and flexible tools for modeling a wide range of real-world combinatorial optimization problems. Predict-and-Search methods operate by using a predictive model to estimate promising variable assignments and then guiding a search procedure toward high-quality solutions. R...
ID-PaS+ : Identity-Aware Predict-and-Search for General Mixed-Integer Linear Programs. Mixed-Integer Linear Programs (MIPs) are powerful and flexible tools for modeling a wide range of real-world combinatorial optimization problems. Predict-and-Search methods operate by using a predictive model to estimate promising va...
ai_general
cs.AI
2025-12-11T01:58:28+00:00
2512.16041v1
Are We on the Right Way to Assessing LLM-as-a-Judge?
LLM-as-a-Judge has been widely adopted as an evaluation method and served as supervised rewards in model training. However, existing benchmarks for LLM-as-a-Judge are mainly relying on human-annotated ground truth, which introduces human bias that undermines the assessment of reliability and imposes scalability constra...
Are We on the Right Way to Assessing LLM-as-a-Judge?. LLM-as-a-Judge has been widely adopted as an evaluation method and served as supervised rewards in model training. However, existing benchmarks for LLM-as-a-Judge are mainly relying on human-annotated ground truth, which introduces human bias that undermines the ass...
cl_nlp
cs.CL
2025-12-17T23:49:55+00:00
2512.23130v2
PathoSyn: Imaging-Pathology MRI Synthesis via Disentangled Deviation Diffusion
We present PathoSyn, a unified generative framework for Magnetic Resonance Imaging (MRI) image synthesis that reformulates imaging-pathology as a disentangled additive deviation on a stable anatomical manifold. Current generative models typically operate in the global pixel domain or rely on binary masks, these paradig...
PathoSyn: Imaging-Pathology MRI Synthesis via Disentangled Deviation Diffusion. We present PathoSyn, a unified generative framework for Magnetic Resonance Imaging (MRI) image synthesis that reformulates imaging-pathology as a disentangled additive deviation on a stable anatomical manifold. Current generative models typ...
cv_vision
cs.CV
2025-12-29T01:13:50+00:00
2511.20503v2
Manifold Percolation: from generative model to Reinforce learning
Generative modeling is typically framed as learning mapping rules, but from an observer's perspective without access to these rules, the task becomes disentangling the geometric support from the probability distribution. We propose that continuum percolation is uniquely suited to this support analysis, as the sampling ...
Manifold Percolation: from generative model to Reinforce learning. Generative modeling is typically framed as learning mapping rules, but from an observer's perspective without access to these rules, the task becomes disentangling the geometric support from the probability distribution. We propose that continuum percol...
stat_ml
stat.ML
2025-11-25T17:12:42+00:00
2510.19999v1
Enhanced Cyclic Coordinate Descent Methods for Elastic Net Penalized Linear Models
We present a novel enhanced cyclic coordinate descent (ECCD) framework for solving generalized linear models with elastic net constraints that reduces training time in comparison to existing state-of-the-art methods. We redesign the CD method by performing a Taylor expansion around the current iterate to avoid nonlinea...
Enhanced Cyclic Coordinate Descent Methods for Elastic Net Penalized Linear Models. We present a novel enhanced cyclic coordinate descent (ECCD) framework for solving generalized linear models with elastic net constraints that reduces training time in comparison to existing state-of-the-art methods. We redesign the CD ...
stat_ml
stat.ML
2025-10-22T20:01:25+00:00
2512.19527v1
Deep Learning for Unrelated-Machines Scheduling: Handling Variable Dimensions
Deep learning has been effectively applied to many discrete optimization problems. However, learning-based scheduling on unrelated parallel machines remains particularly difficult to design. Not only do the numbers of jobs and machines vary, but each job-machine pair has a unique processing time, dynamically altering f...
Deep Learning for Unrelated-Machines Scheduling: Handling Variable Dimensions. Deep learning has been effectively applied to many discrete optimization problems. However, learning-based scheduling on unrelated parallel machines remains particularly difficult to design. Not only do the numbers of jobs and machines vary,...
lg_ml
cs.LG
2025-12-22T16:18:29+00:00
2512.19823v3
Learning to Refocus with Video Diffusion Models
Focus is a cornerstone of photography, yet autofocus systems often fail to capture the intended subject, and users frequently wish to adjust focus after capture. We introduce a novel method for realistic post-capture refocusing using video diffusion models. From a single defocused image, our approach generates a percep...
Learning to Refocus with Video Diffusion Models. Focus is a cornerstone of photography, yet autofocus systems often fail to capture the intended subject, and users frequently wish to adjust focus after capture. We introduce a novel method for realistic post-capture refocusing using video diffusion models. From a single...
cv_vision
cs.CV
2025-12-22T19:29:57+00:00
2512.19937v1
Interpolative Decoding: Exploring the Spectrum of Personality Traits in LLMs
Recent research has explored using very large language models (LLMs) as proxies for humans in tasks such as simulation, surveys, and studies. While LLMs do not possess a human psychology, they often can emulate human behaviors with sufficiently high fidelity to drive simulations to test human behavioral hypotheses, exh...
Interpolative Decoding: Exploring the Spectrum of Personality Traits in LLMs. Recent research has explored using very large language models (LLMs) as proxies for humans in tasks such as simulation, surveys, and studies. While LLMs do not possess a human psychology, they often can emulate human behaviors with sufficient...
ai_general
cs.AI
2025-12-23T00:00:17+00:00
2601.00873v1
Quantum Machine Learning Approaches for Coordinated Stealth Attack Detection in Distributed Generation Systems
Coordinated stealth attacks are a serious cybersecurity threat to distributed generation systems because they modify control and measurement signals while remaining close to normal behavior, making them difficult to detect using standard intrusion detection methods. This study investigates quantum machine learning appr...
Quantum Machine Learning Approaches for Coordinated Stealth Attack Detection in Distributed Generation Systems. Coordinated stealth attacks are a serious cybersecurity threat to distributed generation systems because they modify control and measurement signals while remaining close to normal behavior, making them diffi...
lg_ml
cs.LG
2025-12-30T18:32:13+00:00
2512.05926v1
BalLOT: Balanced $k$-means clustering with optimal transport
We consider the fundamental problem of balanced $k$-means clustering. In particular, we introduce an optimal transport approach to alternating minimization called BalLOT, and we show that it delivers a fast and effective solution to this problem. We establish this with a variety of numerical experiments before proving ...
BalLOT: Balanced $k$-means clustering with optimal transport. We consider the fundamental problem of balanced $k$-means clustering. In particular, we introduce an optimal transport approach to alternating minimization called BalLOT, and we show that it delivers a fast and effective solution to this problem. We establis...
stat_ml
stat.ML
2025-12-05T18:04:35+00:00
2512.03514v1
M3DR: Towards Universal Multilingual Multimodal Document Retrieval
Multimodal document retrieval systems have shown strong progress in aligning visual and textual content for semantic search. However, most existing approaches remain heavily English-centric, limiting their effectiveness in multilingual contexts. In this work, we present M3DR (Multilingual Multimodal Document Retrieval)...
M3DR: Towards Universal Multilingual Multimodal Document Retrieval. Multimodal document retrieval systems have shown strong progress in aligning visual and textual content for semantic search. However, most existing approaches remain heavily English-centric, limiting their effectiveness in multilingual contexts. In thi...
ir_retrieval
cs.IR
2025-12-03T07:17:59+00:00
2512.23422v1
Entropy-Guided Token Dropout: Training Autoregressive Language Models with Limited Domain Data
As access to high-quality, domain-specific data grows increasingly scarce, multi-epoch training has become a practical strategy for adapting large language models (LLMs). However, autoregressive models often suffer from performance degradation under repeated data exposure, where overfitting leads to a marked decline in...
Entropy-Guided Token Dropout: Training Autoregressive Language Models with Limited Domain Data. As access to high-quality, domain-specific data grows increasingly scarce, multi-epoch training has become a practical strategy for adapting large language models (LLMs). However, autoregressive models often suffer from perf...
cl_nlp
cs.CL
2025-12-29T12:35:51+00:00
2512.13634v2
Universality of high-dimensional scaling limits of stochastic gradient descent
We consider statistical tasks in high dimensions whose loss depends on the data only through its projection into a fixed-dimensional subspace spanned by the parameter vectors and certain ground truth vectors. This includes classifying mixture distributions with cross-entropy loss with one and two-layer networks, and le...
Universality of high-dimensional scaling limits of stochastic gradient descent. We consider statistical tasks in high dimensions whose loss depends on the data only through its projection into a fixed-dimensional subspace spanned by the parameter vectors and certain ground truth vectors. This includes classifying mixtu...
stat_ml
stat.ML
2025-12-15T18:30:26+00:00
2512.06814v1
CAuSE: Decoding Multimodal Classifiers using Faithful Natural Language Explanation
Multimodal classifiers function as opaque black box models. While several techniques exist to interpret their predictions, very few of them are as intuitive and accessible as natural language explanations (NLEs). To build trust, such explanations must faithfully capture the classifier's internal decision making behavio...
CAuSE: Decoding Multimodal Classifiers using Faithful Natural Language Explanation. Multimodal classifiers function as opaque black box models. While several techniques exist to interpret their predictions, very few of them are as intuitive and accessible as natural language explanations (NLEs). To build trust, such ex...
cl_nlp
cs.CL
2025-12-07T12:15:21+00:00
2512.24251v1
Deep Reinforcement Learning for Solving the Fleet Size and Mix Vehicle Routing Problem
The Fleet Size and Mix Vehicle Routing Problem (FSMVRP) is a prominent variant of the Vehicle Routing Problem (VRP), extensively studied in operations research and computational science. FSMVRP requires simultaneous decisions on fleet composition and routing, making it highly applicable to real-world scenarios such as ...
Deep Reinforcement Learning for Solving the Fleet Size and Mix Vehicle Routing Problem. The Fleet Size and Mix Vehicle Routing Problem (FSMVRP) is a prominent variant of the Vehicle Routing Problem (VRP), extensively studied in operations research and computational science. FSMVRP requires simultaneous decisions on fle...
ai_general
cs.AI
2025-12-30T14:26:33+00:00
2509.12824v2
DiffHash: Text-Guided Targeted Attack via Diffusion Models against Deep Hashing Image Retrieval
Deep hashing models have been widely adopted to tackle the challenges of large-scale image retrieval. However, these approaches face serious security risks due to their vulnerability to adversarial examples. Despite the increasing exploration of targeted attacks on deep hashing models, existing approaches still suffer ...
DiffHash: Text-Guided Targeted Attack via Diffusion Models against Deep Hashing Image Retrieval. Deep hashing models have been widely adopted to tackle the challenges of large-scale image retrieval. However, these approaches face serious security risks due to their vulnerability to adversarial examples. Despite the inc...
ir_retrieval
cs.IR
2025-09-16T08:49:53+00:00
2511.04939v2
Search Is Not Retrieval: Decoupling Semantic Matching from Contextual Assembly in RAG
Retrieval systems are essential to contemporary AI pipelines, although most confuse two separate processes: finding relevant information and giving enough context for reasoning. We introduce the Search-Is-Not-Retrieve (SINR) framework, a dual-layer architecture that distinguishes between fine-grained search representat...
Search Is Not Retrieval: Decoupling Semantic Matching from Contextual Assembly in RAG. Retrieval systems are essential to contemporary AI pipelines, although most confuse two separate processes: finding relevant information and giving enough context for reasoning. We introduce the Search-Is-Not-Retrieve (SINR) framewor...
ir_retrieval
cs.IR
2025-11-07T02:51:20+00:00
2512.22466v1
AMBIT: Augmenting Mobility Baselines with Interpretable Trees
Origin-destination (OD) flow prediction remains a core task in GIS and urban analytics, yet practical deployments face two conflicting needs: high accuracy and clear interpretability. This paper develops AMBIT, a gray-box framework that augments physical mobility baselines with interpretable tree models. We begin with ...
AMBIT: Augmenting Mobility Baselines with Interpretable Trees. Origin-destination (OD) flow prediction remains a core task in GIS and urban analytics, yet practical deployments face two conflicting needs: high accuracy and clear interpretability. This paper develops AMBIT, a gray-box framework that augments physical mo...
lg_ml
cs.LG
2025-12-27T04:59:16+00:00
2512.23532v2
Iterative Inference-time Scaling with Adaptive Frequency Steering for Image Super-Resolution
Diffusion models have become a leading paradigm for image super-resolution (SR), but existing methods struggle to guarantee both the high-frequency perceptual quality and the low-frequency structural fidelity of generated images. Although inference-time scaling can theoretically improve this trade-off by allocating mor...
Iterative Inference-time Scaling with Adaptive Frequency Steering for Image Super-Resolution. Diffusion models have become a leading paradigm for image super-resolution (SR), but existing methods struggle to guarantee both the high-frequency perceptual quality and the low-frequency structural fidelity of generated imag...
cv_vision
cs.CV
2025-12-29T15:09:20+00:00
2512.12537v1
NagaNLP: Bootstrapping NLP for Low-Resource Nagamese Creole with Human-in-the-Loop Synthetic Data
The vast majority of the world's languages, particularly creoles like Nagamese, remain severely under-resourced in Natural Language Processing (NLP), creating a significant barrier to their representation in digital technology. This paper introduces NagaNLP, a comprehensive open-source toolkit for Nagamese, bootstrappe...
NagaNLP: Bootstrapping NLP for Low-Resource Nagamese Creole with Human-in-the-Loop Synthetic Data. The vast majority of the world's languages, particularly creoles like Nagamese, remain severely under-resourced in Natural Language Processing (NLP), creating a significant barrier to their representation in digital techn...
cl_nlp
cs.CL
2025-12-14T04:08:26+00:00
2512.24957v2
AMAP Agentic Planning Technical Report
We present STAgent, an agentic large language model tailored for spatio-temporal understanding, designed to solve complex tasks such as constrained point-of-interest discovery and itinerary planning. STAgent is a specialized model capable of interacting with ten distinct tools within spatio-temporal scenarios, enabling...
AMAP Agentic Planning Technical Report. We present STAgent, an agentic large language model tailored for spatio-temporal understanding, designed to solve complex tasks such as constrained point-of-interest discovery and itinerary planning. STAgent is a specialized model capable of interacting with ten distinct tools wi...
ai_general
cs.AI
2025-12-31T16:39:09+00:00
2512.11544v1
AI-MASLD Metabolic Dysfunction and Information Steatosis of Large Language Models in Unstructured Clinical Narratives
This study aims to simulate real-world clinical scenarios to systematically evaluate the ability of Large Language Models (LLMs) to extract core medical information from patient chief complaints laden with noise and redundancy, and to verify whether they exhibit a functional decline analogous to Metabolic Dysfunction-A...
AI-MASLD Metabolic Dysfunction and Information Steatosis of Large Language Models in Unstructured Clinical Narratives. This study aims to simulate real-world clinical scenarios to systematically evaluate the ability of Large Language Models (LLMs) to extract core medical information from patient chief complaints laden ...
ai_general
cs.AI
2025-12-12T13:25:19+00:00
2512.22186v1
Learning Tennis Strategy Through Curriculum-Based Dueling Double Deep Q-Networks
Tennis strategy optimization is a challenging sequential decision-making problem involving hierarchical scoring, stochastic outcomes, long-horizon credit assignment, physical fatigue, and adaptation to opponent skill. I present a reinforcement learning framework that integrates a custom tennis simulation environment wi...
Learning Tennis Strategy Through Curriculum-Based Dueling Double Deep Q-Networks. Tennis strategy optimization is a challenging sequential decision-making problem involving hierarchical scoring, stochastic outcomes, long-horizon credit assignment, physical fatigue, and adaptation to opponent skill. I present a reinforc...
lg_ml
cs.LG
2025-12-20T04:22:07+00:00
2512.16383v2
Multivariate Uncertainty Quantification with Tomographic Quantile Forests
Quantifying predictive uncertainty is essential for safe and trustworthy real-world AI deployment. Yet, fully nonparametric estimation of conditional distributions remains challenging for multivariate targets. We propose Tomographic Quantile Forests (TQF), a nonparametric, uncertainty-aware, tree-based regression model...
Multivariate Uncertainty Quantification with Tomographic Quantile Forests. Quantifying predictive uncertainty is essential for safe and trustworthy real-world AI deployment. Yet, fully nonparametric estimation of conditional distributions remains challenging for multivariate targets. We propose Tomographic Quantile For...
lg_ml
cs.LG
2025-12-18T10:25:34+00:00
2512.20033v3
FlashLips: 100-FPS Mask-Free Latent Lip-Sync using Reconstruction Instead of Diffusion or GANs
We present FlashLips, a two-stage, mask-free lip-sync system that decouples lips control from rendering and achieves real-time performance, with our U-Net variant running at over 100 FPS on a single GPU, while matching the visual quality of larger state-of-the-art models. Stage 1 is a compact, one-step latent-space edi...
FlashLips: 100-FPS Mask-Free Latent Lip-Sync using Reconstruction Instead of Diffusion or GANs. We present FlashLips, a two-stage, mask-free lip-sync system that decouples lips control from rendering and achieves real-time performance, with our U-Net variant running at over 100 FPS on a single GPU, while matching the v...
cv_vision
cs.CV
2025-12-23T03:54:48+00:00
2511.10824v1
Neural Local Wasserstein Regression
We study the estimation problem of distribution-on-distribution regression, where both predictors and responses are probability measures. Existing approaches typically rely on a global optimal transport map or tangent-space linearization, which can be restrictive in approximation capacity and distort geometry in multiv...
Neural Local Wasserstein Regression. We study the estimation problem of distribution-on-distribution regression, where both predictors and responses are probability measures. Existing approaches typically rely on a global optimal transport map or tangent-space linearization, which can be restrictive in approximation ca...
stat_ml
stat.ML
2025-11-13T21:54:18+00:00
2512.15036v2
Spectral Representation-based Reinforcement Learning
In real-world applications with large state and action spaces, reinforcement learning (RL) typically employs function approximations to represent core components like the policies, value functions, and dynamics models. Although powerful approximations such as neural networks offer great expressiveness, they often prese...
Spectral Representation-based Reinforcement Learning. In real-world applications with large state and action spaces, reinforcement learning (RL) typically employs function approximations to represent core components like the policies, value functions, and dynamics models. Although powerful approximations such as neural...
lg_ml
cs.LG
2025-12-17T02:54:42+00:00
2512.20770v2
OccuFly: A 3D Vision Benchmark for Semantic Scene Completion from the Aerial Perspective
Semantic Scene Completion (SSC) is essential for 3D perception in mobile robotics, as it enables holistic scene understanding by jointly estimating dense volumetric occupancy and per-voxel semantics. Although SSC has been widely studied in terrestrial domains such as autonomous driving, aerial settings like autonomous ...
OccuFly: A 3D Vision Benchmark for Semantic Scene Completion from the Aerial Perspective. Semantic Scene Completion (SSC) is essential for 3D perception in mobile robotics, as it enables holistic scene understanding by jointly estimating dense volumetric occupancy and per-voxel semantics. Although SSC has been widely s...
cv_vision
cs.CV
2025-12-23T21:14:55+00:00
2512.23280v1
Chinese Morph Resolution in E-commerce Live Streaming Scenarios
E-commerce live streaming in China, particularly on platforms like Douyin, has become a major sales channel, but hosts often use morphs to evade scrutiny and engage in false advertising. This study introduces the Live Auditory Morph Resolution (LiveAMR) task to detect such violations. Unlike previous morph research foc...
Chinese Morph Resolution in E-commerce Live Streaming Scenarios. E-commerce live streaming in China, particularly on platforms like Douyin, has become a major sales channel, but hosts often use morphs to evade scrutiny and engage in false advertising. This study introduces the Live Auditory Morph Resolution (LiveAMR) t...
cl_nlp
cs.CL
2025-12-29T08:04:48+00:00
2512.14288v1
Leveraging LLMs for Collaborative Ontology Engineering in Parkinson Disease Monitoring and Alerting
This paper explores the integration of Large Language Models (LLMs) in the engineering of a Parkinson's Disease (PD) monitoring and alerting ontology through four key methodologies: One Shot (OS) prompt techniques, Chain of Thought (CoT) prompts, X-HCOME, and SimX-HCOME+. The primary objective is to determine whether L...
Leveraging LLMs for Collaborative Ontology Engineering in Parkinson Disease Monitoring and Alerting. This paper explores the integration of Large Language Models (LLMs) in the engineering of a Parkinson's Disease (PD) monitoring and alerting ontology through four key methodologies: One Shot (OS) prompt techniques, Chai...
ai_general
cs.AI
2025-12-16T10:58:26+00:00
2512.21829v1
Tilt Matching for Scalable Sampling and Fine-Tuning
We propose a simple, scalable algorithm for using stochastic interpolants to sample from unnormalized densities and for fine-tuning generative models. The approach, Tilt Matching, arises from a dynamical equation relating the flow matching velocity to one targeting the same distribution tilted by a reward, implicitly s...
Tilt Matching for Scalable Sampling and Fine-Tuning. We propose a simple, scalable algorithm for using stochastic interpolants to sample from unnormalized densities and for fine-tuning generative models. The approach, Tilt Matching, arises from a dynamical equation relating the flow matching velocity to one targeting t...
stat_ml
stat.ML
2025-12-26T02:12:10+00:00
2512.08404v1
Are generative AI text annotations systematically biased?
This paper investigates bias in GLLM annotations by conceptually replicating manual annotations of Boukes (2024). Using various GLLMs (Llama3.1:8b, Llama3.3:70b, GPT4o, Qwen2.5:72b) in combination with five different prompts for five concepts (political content, interactivity, rationality, incivility, and ideology). We...
Are generative AI text annotations systematically biased?. This paper investigates bias in GLLM annotations by conceptually replicating manual annotations of Boukes (2024). Using various GLLMs (Llama3.1:8b, Llama3.3:70b, GPT4o, Qwen2.5:72b) in combination with five different prompts for five concepts (political content...
cl_nlp
cs.CL
2025-12-09T09:36:43+00:00
2512.19410v1
Research Program: Theory of Learning in Dynamical Systems
Modern learning systems increasingly interact with data that evolve over time and depend on hidden internal state. We ask a basic question: when is such a dynamical system learnable from observations alone? This paper proposes a research program for understanding learnability in dynamical systems through the lens of ne...
Research Program: Theory of Learning in Dynamical Systems. Modern learning systems increasingly interact with data that evolve over time and depend on hidden internal state. We ask a basic question: when is such a dynamical system learnable from observations alone? This paper proposes a research program for understandi...
lg_ml
cs.LG
2025-12-22T14:05:31+00:00