id
string
title
string
abstract
string
text
string
label
string
primary_category
string
published
string
2512.23487v1
ML Compass: Navigating Capability, Cost, and Compliance Trade-offs in AI Model Deployment
We study how organizations should select among competing AI models when user utility, deployment costs, and compliance requirements jointly matter. Widely used capability leaderboards do not translate directly into deployment decisions, creating a capability -- deployment gap; to bridge it, we take a systems-level view...
ML Compass: Navigating Capability, Cost, and Compliance Trade-offs in AI Model Deployment. We study how organizations should select among competing AI models when user utility, deployment costs, and compliance requirements jointly matter. Widely used capability leaderboards do not translate directly into deployment dec...
lg_ml
cs.LG
2025-12-29T14:19:48+00:00
2512.17527v1
SafeBench-Seq: A Homology-Clustered, CPU-Only Baseline for Protein Hazard Screening with Physicochemical/Composition Features and Cluster-Aware Confidence Intervals
Foundation models for protein design raise concrete biosecurity risks, yet the community lacks a simple, reproducible baseline for sequence-level hazard screening that is explicitly evaluated under homology control and runs on commodity CPUs. We introduce SafeBench-Seq, a metadata-only, reproducible benchmark and basel...
SafeBench-Seq: A Homology-Clustered, CPU-Only Baseline for Protein Hazard Screening with Physicochemical/Composition Features and Cluster-Aware Confidence Intervals. Foundation models for protein design raise concrete biosecurity risks, yet the community lacks a simple, reproducible baseline for sequence-level hazard s...
lg_ml
cs.LG
2025-12-19T12:51:31+00:00
2512.13120v1
Towards Practical Large-scale Dynamical Heterogeneous Graph Embedding: Cold-start Resilient Recommendation
Deploying dynamic heterogeneous graph embeddings in production faces key challenges of scalability, data freshness, and cold-start. This paper introduces a practical, two-stage solution that balances deep graph representation with low-latency incremental updates. Our framework combines HetSGFormer, a scalable graph tra...
Towards Practical Large-scale Dynamical Heterogeneous Graph Embedding: Cold-start Resilient Recommendation. Deploying dynamic heterogeneous graph embeddings in production faces key challenges of scalability, data freshness, and cold-start. This paper introduces a practical, two-stage solution that balances deep graph r...
ir_retrieval
cs.IR
2025-12-15T09:19:23+00:00
2511.22065v1
Support Vector Machine Classifier with Rescaled Huberized Pinball Loss
Support vector machines are widely used in machine learning classification tasks, but traditional SVM models suffer from sensitivity to outliers and instability in resampling, which limits their performance in practical applications. To address these issues, this paper proposes a novel rescaled Huberized pinball loss f...
Support Vector Machine Classifier with Rescaled Huberized Pinball Loss. Support vector machines are widely used in machine learning classification tasks, but traditional SVM models suffer from sensitivity to outliers and instability in resampling, which limits their performance in practical applications. To address the...
stat_ml
stat.ML
2025-11-27T03:31:07+00:00
2601.00869v1
Cultural Encoding in Large Language Models: The Existence Gap in AI-Mediated Brand Discovery
As artificial intelligence systems increasingly mediate consumer information discovery, brands face algorithmic invisibility. This study investigates Cultural Encoding in Large Language Models (LLMs) -- systematic differences in brand recommendations arising from training data composition. Analyzing 1,909 pure-English ...
Cultural Encoding in Large Language Models: The Existence Gap in AI-Mediated Brand Discovery. As artificial intelligence systems increasingly mediate consumer information discovery, brands face algorithmic invisibility. This study investigates Cultural Encoding in Large Language Models (LLMs) -- systematic differences ...
ai_general
cs.AI
2025-12-30T13:50:14+00:00
2510.15722v1
The 3rd Place Solution of CCIR CUP 2025: A Framework for Retrieval-Augmented Generation in Multi-Turn Legal Conversation
Retrieval-Augmented Generation has made significant progress in the field of natural language processing. By combining the advantages of information retrieval and large language models, RAG can generate relevant and contextually appropriate responses based on items retrieved from reliable sources. This technology has d...
The 3rd Place Solution of CCIR CUP 2025: A Framework for Retrieval-Augmented Generation in Multi-Turn Legal Conversation. Retrieval-Augmented Generation has made significant progress in the field of natural language processing. By combining the advantages of information retrieval and large language models, RAG can gene...
ir_retrieval
cs.IR
2025-10-17T15:12:15+00:00
2510.24402v1
Metadata-Driven Retrieval-Augmented Generation for Financial Question Answering
Retrieval-Augmented Generation (RAG) struggles on long, structured financial filings where relevant evidence is sparse and cross-referenced. This paper presents a systematic investigation of advanced metadata-driven Retrieval-Augmented Generation (RAG) techniques, proposing and evaluating a novel, multi-stage RAG archi...
Metadata-Driven Retrieval-Augmented Generation for Financial Question Answering. Retrieval-Augmented Generation (RAG) struggles on long, structured financial filings where relevant evidence is sparse and cross-referenced. This paper presents a systematic investigation of advanced metadata-driven Retrieval-Augmented Gen...
ir_retrieval
cs.IR
2025-10-28T13:16:36+00:00
2512.07277v1
Efficient ASR for Low-Resource Languages: Leveraging Cross-Lingual Unlabeled Data
Automatic speech recognition for low-resource languages remains fundamentally constrained by the scarcity of labeled data and computational resources required by state-of-the-art models. We present a systematic investigation into cross-lingual continuous pretraining for low-resource languages, using Perso-Arabic langua...
Efficient ASR for Low-Resource Languages: Leveraging Cross-Lingual Unlabeled Data. Automatic speech recognition for low-resource languages remains fundamentally constrained by the scarcity of labeled data and computational resources required by state-of-the-art models. We present a systematic investigation into cross-l...
cl_nlp
cs.CL
2025-12-08T08:16:34+00:00
2601.11572v1
Discrete Semantic States and Hamiltonian Dynamics in LLM Embedding Spaces
We investigate the structure of Large Language Model (LLM) embedding spaces using mathematical concepts, particularly linear algebra and the Hamiltonian formalism, drawing inspiration from analogies with quantum mechanical systems. Motivated by the observation that LLM embeddings exhibit distinct states, suggesting dis...
Discrete Semantic States and Hamiltonian Dynamics in LLM Embedding Spaces. We investigate the structure of Large Language Model (LLM) embedding spaces using mathematical concepts, particularly linear algebra and the Hamiltonian formalism, drawing inspiration from analogies with quantum mechanical systems. Motivated by ...
lg_ml
cs.LG
2025-12-29T15:01:43+00:00
2512.22386v2
OxygenREC: An Instruction-Following Generative Framework for E-commerce Recommendation
Traditional recommendation systems suffer from inconsistency in multi-stage optimization objectives. Generative Recommendation (GR) mitigates them through an end-to-end framework; however, existing methods still rely on matching mechanisms based on inductive patterns. Although responsive, they lack the ability to uncov...
OxygenREC: An Instruction-Following Generative Framework for E-commerce Recommendation. Traditional recommendation systems suffer from inconsistency in multi-stage optimization objectives. Generative Recommendation (GR) mitigates them through an end-to-end framework; however, existing methods still rely on matching mec...
ir_retrieval
cs.IR
2025-12-26T21:13:59+00:00
2509.26378v1
MR$^2$-Bench: Going Beyond Matching to Reasoning in Multimodal Retrieval
Multimodal retrieval is becoming a crucial component of modern AI applications, yet its evaluation lags behind the demands of more realistic and challenging scenarios. Existing benchmarks primarily probe surface-level semantic correspondence (e.g., object-text matching) while failing to assess the deeper reasoning requ...
MR$^2$-Bench: Going Beyond Matching to Reasoning in Multimodal Retrieval. Multimodal retrieval is becoming a crucial component of modern AI applications, yet its evaluation lags behind the demands of more realistic and challenging scenarios. Existing benchmarks primarily probe surface-level semantic correspondence (e.g...
ir_retrieval
cs.IR
2025-09-30T15:09:14+00:00
2512.05824v1
Multimodal Oncology Agent for IDH1 Mutation Prediction in Low-Grade Glioma
Low-grade gliomas frequently present IDH1 mutations that define clinically distinct subgroups with specific prognostic and therapeutic implications. This work introduces a Multimodal Oncology Agent (MOA) integrating a histology tool based on the TITAN foundation model for IDH1 mutation prediction in low-grade glioma, c...
Multimodal Oncology Agent for IDH1 Mutation Prediction in Low-Grade Glioma. Low-grade gliomas frequently present IDH1 mutations that define clinically distinct subgroups with specific prognostic and therapeutic implications. This work introduces a Multimodal Oncology Agent (MOA) integrating a histology tool based on th...
ai_general
cs.AI
2025-12-05T15:43:02+00:00
2512.22984v1
Reverse Personalization
Recent text-to-image diffusion models have demonstrated remarkable generation of realistic facial images conditioned on textual prompts and human identities, enabling creating personalized facial imagery. However, existing prompt-based methods for removing or modifying identity-specific features rely either on the subj...
Reverse Personalization. Recent text-to-image diffusion models have demonstrated remarkable generation of realistic facial images conditioned on textual prompts and human identities, enabling creating personalized facial imagery. However, existing prompt-based methods for removing or modifying identity-specific feature...
cv_vision
cs.CV
2025-12-28T16:06:55+00:00
2512.06945v2
Symmetric Aggregation of Conformity Scores for Efficient Uncertainty Sets
Access to multiple predictive models trained for the same task, whether in regression or classification, is increasingly common in many applications. Aggregating their predictive uncertainties to produce reliable and efficient uncertainty quantification is therefore a critical but still underexplored challenge, especia...
Symmetric Aggregation of Conformity Scores for Efficient Uncertainty Sets. Access to multiple predictive models trained for the same task, whether in regression or classification, is increasingly common in many applications. Aggregating their predictive uncertainties to produce reliable and efficient uncertainty quanti...
stat_ml
stat.ML
2025-12-07T17:54:07+00:00
2510.12077v1
Compressibility Measures Complexity: Minimum Description Length Meets Singular Learning Theory
We study neural network compressibility by using singular learning theory to extend the minimum description length (MDL) principle to singular models like neural networks. Through extensive experiments on the Pythia suite with quantization, factorization, and other compression techniques, we find that complexity estima...
Compressibility Measures Complexity: Minimum Description Length Meets Singular Learning Theory. We study neural network compressibility by using singular learning theory to extend the minimum description length (MDL) principle to singular models like neural networks. Through extensive experiments on the Pythia suite wi...
stat_ml
stat.ML
2025-10-14T02:38:02+00:00
2512.22999v1
JADAI: Jointly Amortizing Adaptive Design and Bayesian Inference
We consider problems of parameter estimation where design variables can be actively optimized to maximize information gain. To this end, we introduce JADAI, a framework that jointly amortizes Bayesian adaptive design and inference by training a policy, a history network, and an inference network end-to-end. The network...
JADAI: Jointly Amortizing Adaptive Design and Bayesian Inference. We consider problems of parameter estimation where design variables can be actively optimized to maximize information gain. To this end, we introduce JADAI, a framework that jointly amortizes Bayesian adaptive design and inference by training a policy, a...
stat_ml
stat.ML
2025-12-28T16:54:43+00:00
2512.22065v2
StreamAvatar: Streaming Diffusion Models for Real-Time Interactive Human Avatars
Real-time, streaming interactive avatars represent a critical yet challenging goal in digital human research. Although diffusion-based human avatar generation methods achieve remarkable success, their non-causal architecture and high computational costs make them unsuitable for streaming. Moreover, existing interactive...
StreamAvatar: Streaming Diffusion Models for Real-Time Interactive Human Avatars. Real-time, streaming interactive avatars represent a critical yet challenging goal in digital human research. Although diffusion-based human avatar generation methods achieve remarkable success, their non-causal architecture and high comp...
cv_vision
cs.CV
2025-12-26T15:41:24+00:00
2512.19982v1
WSD-MIL: Window Scale Decay Multiple Instance Learning for Whole Slide Image Classification
In recent years, the integration of pre-trained foundational models with multiple instance learning (MIL) has improved diagnostic accuracy in computational pathology. However, existing MIL methods focus on optimizing feature extractors and aggregation strategies while overlooking the complex semantic relationships amon...
WSD-MIL: Window Scale Decay Multiple Instance Learning for Whole Slide Image Classification. In recent years, the integration of pre-trained foundational models with multiple instance learning (MIL) has improved diagnostic accuracy in computational pathology. However, existing MIL methods focus on optimizing feature ex...
cv_vision
cs.CV
2025-12-23T02:10:24+00:00
2512.08411v2
Prismatic World Model: Learning Compositional Dynamics for Planning in Hybrid Systems
Model-based planning in robotic domains is challenged by the hybrid nature of physical dynamics, where continuous motion is punctuated by discrete events such as contacts and impacts. Conventional latent world models typically employ monolithic neural networks that enforce global continuity, which over-smooths distinct...
Prismatic World Model: Learning Compositional Dynamics for Planning in Hybrid Systems. Model-based planning in robotic domains is challenged by the hybrid nature of physical dynamics, where continuous motion is punctuated by discrete events such as contacts and impacts. Conventional latent world models typically employ...
ai_general
cs.AI
2025-12-09T09:40:34+00:00
2510.18777v1
A Frequentist Statistical Introduction to Variational Inference, Autoencoders, and Diffusion Models
While Variational Inference (VI) is central to modern generative models like Variational Autoencoders (VAEs) and Denoising Diffusion Models (DDMs), its pedagogical treatment is split across disciplines. In statistics, VI is typically framed as a Bayesian method for posterior approximation. In machine learning, however,...
A Frequentist Statistical Introduction to Variational Inference, Autoencoders, and Diffusion Models. While Variational Inference (VI) is central to modern generative models like Variational Autoencoders (VAEs) and Denoising Diffusion Models (DDMs), its pedagogical treatment is split across disciplines. In statistics, V...
stat_ml
stat.ML
2025-10-21T16:25:19+00:00
2509.21371v1
ReGeS: Reciprocal Retrieval-Generation Synergy for Conversational Recommender Systems
Connecting conversation with external domain knowledge is vital for conversational recommender systems (CRS) to correctly understand user preferences. However, existing solutions either require domain-specific engineering, which limits flexibility, or rely solely on large language models, which increases the risk of ha...
ReGeS: Reciprocal Retrieval-Generation Synergy for Conversational Recommender Systems. Connecting conversation with external domain knowledge is vital for conversational recommender systems (CRS) to correctly understand user preferences. However, existing solutions either require domain-specific engineering, which limi...
ir_retrieval
cs.IR
2025-09-22T17:47:57+00:00
2512.04829v2
Model-Based and Sample-Efficient AI-Assisted Math Discovery in Sphere Packing
Sphere packing, Hilbert's eighteenth problem, asks for the densest arrangement of congruent spheres in n-dimensional Euclidean space. Although relevant to areas such as cryptography, crystallography, and medical imaging, the problem remains unresolved: beyond a few special dimensions, neither optimal packings nor tight...
Model-Based and Sample-Efficient AI-Assisted Math Discovery in Sphere Packing. Sphere packing, Hilbert's eighteenth problem, asks for the densest arrangement of congruent spheres in n-dimensional Euclidean space. Although relevant to areas such as cryptography, crystallography, and medical imaging, the problem remains ...
ai_general
cs.AI
2025-12-04T14:11:52+00:00
2510.14880v1
Fantastic (small) Retrievers and How to Train Them: mxbai-edge-colbert-v0 Tech Report
In this work, we introduce mxbai-edge-colbert-v0 models, at two different parameter counts: 17M and 32M. As part of our research, we conduct numerous experiments to improve retrieval and late-interaction models, which we intend to distill into smaller models as proof-of-concepts. Our ultimate aim is to support retrieva...
Fantastic (small) Retrievers and How to Train Them: mxbai-edge-colbert-v0 Tech Report. In this work, we introduce mxbai-edge-colbert-v0 models, at two different parameter counts: 17M and 32M. As part of our research, we conduct numerous experiments to improve retrieval and late-interaction models, which we intend to di...
ir_retrieval
cs.IR
2025-10-16T17:00:35+00:00
2512.19943v1
SE360: Semantic Edit in 360$^\circ$ Panoramas via Hierarchical Data Construction
While instruction-based image editing is emerging, extending it to 360$^\circ$ panoramas introduces additional challenges. Existing methods often produce implausible results in both equirectangular projections (ERP) and perspective views. To address these limitations, we propose SE360, a novel framework for multi-condi...
SE360: Semantic Edit in 360$^\circ$ Panoramas via Hierarchical Data Construction. While instruction-based image editing is emerging, extending it to 360$^\circ$ panoramas introduces additional challenges. Existing methods often produce implausible results in both equirectangular projections (ERP) and perspective views....
cv_vision
cs.CV
2025-12-23T00:24:46+00:00
2512.24793v1
Self-Supervised Neural Architecture Search for Multimodal Deep Neural Networks
Neural architecture search (NAS), which automates the architectural design process of deep neural networks (DNN), has attracted increasing attention. Multimodal DNNs that necessitate feature fusion from multiple modalities benefit from NAS due to their structural complexity; however, constructing an architecture for mu...
Self-Supervised Neural Architecture Search for Multimodal Deep Neural Networks. Neural architecture search (NAS), which automates the architectural design process of deep neural networks (DNN), has attracted increasing attention. Multimodal DNNs that necessitate feature fusion from multiple modalities benefit from NAS ...
lg_ml
cs.LG
2025-12-31T11:30:28+00:00
2512.24948v1
ProDM: Synthetic Reality-driven Property-aware Progressive Diffusion Model for Coronary Calcium Motion Correction in Non-gated Chest CT
Coronary artery calcium (CAC) scoring from chest CT is a well-established tool to stratify and refine clinical cardiovascular disease risk estimation. CAC quantification relies on the accurate delineation of calcified lesions, but is oftentimes affected by artifacts introduced by cardiac and respiratory motion. ECG-gat...
ProDM: Synthetic Reality-driven Property-aware Progressive Diffusion Model for Coronary Calcium Motion Correction in Non-gated Chest CT. Coronary artery calcium (CAC) scoring from chest CT is a well-established tool to stratify and refine clinical cardiovascular disease risk estimation. CAC quantification relies on the...
cv_vision
cs.CV
2025-12-31T16:29:05+00:00
2512.10793v2
LabelFusion: Fusing Large Language Models with Transformer Encoders for Robust Financial News Classification
Financial news plays a central role in shaping investor sentiment and short-term dynamics in commodity markets. Many downstream financial applications, such as commodity price prediction or sentiment modeling, therefore rely on the ability to automatically identify news articles relevant to specific assets. However, ob...
LabelFusion: Fusing Large Language Models with Transformer Encoders for Robust Financial News Classification. Financial news plays a central role in shaping investor sentiment and short-term dynamics in commodity markets. Many downstream financial applications, such as commodity price prediction or sentiment modeling, ...
cl_nlp
cs.CL
2025-12-11T16:39:07+00:00
2512.19287v1
Vibe Reasoning: Eliciting Frontier AI Mathematical Capabilities -- A Case Study on IMO 2025 Problem 6
We introduce Vibe Reasoning, a human-AI collaborative paradigm for solving complex mathematical problems. Our key insight is that frontier AI models already possess the knowledge required to solve challenging problems -- they simply do not know how, what, or when to apply it. Vibe Reasoning transforms AI's latent poten...
Vibe Reasoning: Eliciting Frontier AI Mathematical Capabilities -- A Case Study on IMO 2025 Problem 6. We introduce Vibe Reasoning, a human-AI collaborative paradigm for solving complex mathematical problems. Our key insight is that frontier AI models already possess the knowledge required to solve challenging problems...
ai_general
cs.AI
2025-12-22T11:30:19+00:00
2512.16233v1
DAG Learning from Zero-Inflated Count Data Using Continuous Optimization
We address network structure learning from zero-inflated count data by casting each node as a zero-inflated generalized linear model and optimizing a smooth, score-based objective under a directed acyclic graph constraint. Our Zero-Inflated Continuous Optimization (ZICO) approach uses node-wise likelihoods with canonic...
DAG Learning from Zero-Inflated Count Data Using Continuous Optimization. We address network structure learning from zero-inflated count data by casting each node as a zero-inflated generalized linear model and optimizing a smooth, score-based objective under a directed acyclic graph constraint. Our Zero-Inflated Conti...
stat_ml
stat.ML
2025-12-18T06:26:43+00:00
2512.23748v2
A Review of Diffusion-based Simulation-Based Inference: Foundations and Applications in Non-Ideal Data Scenarios
For complex simulation problems, inferring parameters often precludes the use of classical likelihood-based techniques due to intractable likelihoods. Simulation-based inference (SBI) methods offer a likelihood-free approach to directly learn posterior distributions $p(\bftheta \mid \xobs)$ from simulator outputs. Rece...
A Review of Diffusion-based Simulation-Based Inference: Foundations and Applications in Non-Ideal Data Scenarios. For complex simulation problems, inferring parameters often precludes the use of classical likelihood-based techniques due to intractable likelihoods. Simulation-based inference (SBI) methods offer a likeli...
lg_ml
cs.LG
2025-12-26T18:18:25+00:00
2602.22225v1
SmartChunk Retrieval: Query-Aware Chunk Compression with Planning for Efficient Document RAG
Retrieval-augmented generation (RAG) has strong potential for producing accurate and factual outputs by combining language models (LMs) with evidence retrieved from large text corpora. However, current pipelines are limited by static chunking and flat retrieval: documents are split into short, predetermined, fixed-size...
SmartChunk Retrieval: Query-Aware Chunk Compression with Planning for Efficient Document RAG. Retrieval-augmented generation (RAG) has strong potential for producing accurate and factual outputs by combining language models (LMs) with evidence retrieved from large text corpora. However, current pipelines are limited by...
ir_retrieval
cs.IR
2025-12-17T01:21:44+00:00
2512.06960v1
Learning Conditional Independence Differential Graphs From Time-Dependent Data
Estimation of differences in conditional independence graphs (CIGs) of two time series Gaussian graphical models (TSGGMs) is investigated where the two TSGGMs are known to have similar structure. The TSGGM structure is encoded in the inverse power spectral density (IPSD) of the time series. In several existing works, o...
Learning Conditional Independence Differential Graphs From Time-Dependent Data. Estimation of differences in conditional independence graphs (CIGs) of two time series Gaussian graphical models (TSGGMs) is investigated where the two TSGGMs are known to have similar structure. The TSGGM structure is encoded in the invers...
stat_ml
stat.ML
2025-12-07T18:45:04+00:00
2512.09804v2
OnCoCo 1.0: A Public Dataset for Fine-Grained Message Classification in Online Counseling Conversations
This paper presents OnCoCo 1.0, a new public dataset for fine-grained message classification in online counseling. It is based on a new, integrative system of categories, designed to improve the automated analysis of psychosocial online counseling conversations. Existing category systems, predominantly based on Motivat...
OnCoCo 1.0: A Public Dataset for Fine-Grained Message Classification in Online Counseling Conversations. This paper presents OnCoCo 1.0, a new public dataset for fine-grained message classification in online counseling. It is based on a new, integrative system of categories, designed to improve the automated analysis o...
cl_nlp
cs.CL
2025-12-10T16:18:20+00:00
2509.25755v2
HiFIRec Towards High-Frequency yet Low-Intention Behaviors for Multi-Behavior Recommendation
Multi behavior recommendation leverages multiple types of user-item interactions to address data sparsity and cold-start issues,providing personalized services in domains such as healthcare and ecommerce.Most existing methods utilize graph neural networks to model user intention in a unified manner,which inadequately c...
HiFIRec Towards High-Frequency yet Low-Intention Behaviors for Multi-Behavior Recommendation. Multi behavior recommendation leverages multiple types of user-item interactions to address data sparsity and cold-start issues,providing personalized services in domains such as healthcare and ecommerce.Most existing methods ...
ir_retrieval
cs.IR
2025-09-30T04:20:45+00:00
2512.18571v1
ESearch-R1: Learning Cost-Aware MLLM Agents for Interactive Embodied Search via Reinforcement Learning
Multimodal Large Language Models (MLLMs) have empowered embodied agents with remarkable capabilities in planning and reasoning. However, when facing ambiguous natural language instructions (e.g., "fetch the tool" in a cluttered room), current agents often fail to balance the high cost of physical exploration against th...
ESearch-R1: Learning Cost-Aware MLLM Agents for Interactive Embodied Search via Reinforcement Learning. Multimodal Large Language Models (MLLMs) have empowered embodied agents with remarkable capabilities in planning and reasoning. However, when facing ambiguous natural language instructions (e.g., "fetch the tool" in ...
ai_general
cs.AI
2025-12-21T02:45:08+00:00
2512.18174v1
Conscious Data Contribution via Community-Driven Chain-of-Thought Distillation
The current era of AI development places a heavy emphasis on training large models on increasingly scaled-up datasets. This paradigm has catalyzed entirely new product categories, such as LLM chatbots, while also raising concerns about data privacy and consumer choice. In this paper, we consider questions of data porta...
Conscious Data Contribution via Community-Driven Chain-of-Thought Distillation. The current era of AI development places a heavy emphasis on training large models on increasingly scaled-up datasets. This paradigm has catalyzed entirely new product categories, such as LLM chatbots, while also raising concerns about data...
lg_ml
cs.LG
2025-12-20T02:17:18+00:00
2512.24278v1
One-shot synthesis of rare gastrointestinal lesions improves diagnostic accuracy and clinical training
Rare gastrointestinal lesions are infrequently encountered in routine endoscopy, restricting the data available for developing reliable artificial intelligence (AI) models and training novice clinicians. Here we present EndoRare, a one-shot, retraining-free generative framework that synthesizes diverse, high-fidelity l...
One-shot synthesis of rare gastrointestinal lesions improves diagnostic accuracy and clinical training. Rare gastrointestinal lesions are infrequently encountered in routine endoscopy, restricting the data available for developing reliable artificial intelligence (AI) models and training novice clinicians. Here we pres...
cv_vision
cs.CV
2025-12-30T15:07:09+00:00
2512.23328v3
CubeBench: Diagnosing Interactive, Long-Horizon Spatial Reasoning Under Partial Observations
Large Language Model (LLM) agents, while proficient in the digital realm, face a significant gap in physical-world deployment due to the challenge of forming and maintaining a robust spatial mental model. We identify three core cognitive challenges hindering this transition: spatial reasoning, long-horizon state tracki...
CubeBench: Diagnosing Interactive, Long-Horizon Spatial Reasoning Under Partial Observations. Large Language Model (LLM) agents, while proficient in the digital realm, face a significant gap in physical-world deployment due to the challenge of forming and maintaining a robust spatial mental model. We identify three cor...
ai_general
cs.AI
2025-12-29T09:25:56+00:00
2510.23015v1
Coupled Flow Matching
We introduce Coupled Flow Matching (CPFM), a framework that integrates controllable dimensionality reduction and high-fidelity reconstruction. CPFM learns coupled continuous flows for both the high-dimensional data x and the low-dimensional embedding y, which enables sampling p(y|x) via a latent-space flow and p(x|y) v...
Coupled Flow Matching. We introduce Coupled Flow Matching (CPFM), a framework that integrates controllable dimensionality reduction and high-fidelity reconstruction. CPFM learns coupled continuous flows for both the high-dimensional data x and the low-dimensional embedding y, which enables sampling p(y|x) via a latent-...
stat_ml
stat.ML
2025-10-27T05:22:35+00:00
2512.22010v1
LongFly: Long-Horizon UAV Vision-and-Language Navigation with Spatiotemporal Context Integration
Unmanned aerial vehicles (UAVs) are crucial tools for post-disaster search and rescue, facing challenges such as high information density, rapid changes in viewpoint, and dynamic structures, especially in long-horizon navigation. However, current UAV vision-and-language navigation(VLN) methods struggle to model long-ho...
LongFly: Long-Horizon UAV Vision-and-Language Navigation with Spatiotemporal Context Integration. Unmanned aerial vehicles (UAVs) are crucial tools for post-disaster search and rescue, facing challenges such as high information density, rapid changes in viewpoint, and dynamic structures, especially in long-horizon navi...
cv_vision
cs.CV
2025-12-26T12:09:40+00:00
2510.07624v3
From Data to Rewards: a Bilevel Optimization Perspective on Maximum Likelihood Estimation
Generative models form the backbone of modern machine learning, underpinning state-of-the-art systems in text, vision, and multimodal applications. While Maximum Likelihood Estimation has traditionally served as the dominant training paradigm, recent work have highlighted its limitations, particularly in generalization...
From Data to Rewards: a Bilevel Optimization Perspective on Maximum Likelihood Estimation. Generative models form the backbone of modern machine learning, underpinning state-of-the-art systems in text, vision, and multimodal applications. While Maximum Likelihood Estimation has traditionally served as the dominant trai...
stat_ml
stat.ML
2025-10-08T23:45:37+00:00
2512.22382v1
Completed Hyperparameter Transfer across Modules, Width, Depth, Batch and Duration
Hyperparameter tuning can dramatically impact training stability and final performance of large-scale models. Recent works on neural network parameterisations, such as $μ$P, have enabled transfer of optimal global hyperparameters across model sizes. These works propose an empirical practice of search for optimal global...
Completed Hyperparameter Transfer across Modules, Width, Depth, Batch and Duration. Hyperparameter tuning can dramatically impact training stability and final performance of large-scale models. Recent works on neural network parameterisations, such as $μ$P, have enabled transfer of optimal global hyperparameters across...
lg_ml
cs.LG
2025-12-26T20:56:04+00:00
2509.18047v1
Functional effects models: Accounting for preference heterogeneity in panel data with machine learning
In this paper, we present a general specification for Functional Effects Models, which use Machine Learning (ML) methodologies to learn individual-specific preference parameters from socio-demographic characteristics, therefore accounting for inter-individual heterogeneity in panel choice data. We identify three specif...
Functional effects models: Accounting for preference heterogeneity in panel data with machine learning. In this paper, we present a general specification for Functional Effects Models, which use Machine Learning (ML) methodologies to learn individual-specific preference parameters from socio-demographic characteristics...
stat_ml
stat.ML
2025-09-22T17:22:18+00:00
2509.05541v1
Cryo-EM as a Stochastic Inverse Problem
Cryo-electron microscopy (Cryo-EM) enables high-resolution imaging of biomolecules, but structural heterogeneity remains a major challenge in 3D reconstruction. Traditional methods assume a discrete set of conformations, limiting their ability to recover continuous structural variability. In this work, we formulate cry...
Cryo-EM as a Stochastic Inverse Problem. Cryo-electron microscopy (Cryo-EM) enables high-resolution imaging of biomolecules, but structural heterogeneity remains a major challenge in 3D reconstruction. Traditional methods assume a discrete set of conformations, limiting their ability to recover continuous structural va...
stat_ml
stat.ML
2025-09-05T23:35:04+00:00
2512.21508v1
Fixed-Budget Parameter-Efficient Training with Frozen Encoders Improves Multimodal Chest X-Ray Classification
Multimodal chest X-Ray analysis often fine-tunes large vision-language models, which is computationally costly. We study parameter-efficient training (PET) strategies, including frozen encoders, BitFit, LoRA, and adapters for multi-label classification on the Indiana University Chest X-Ray dataset (3,851 image-report p...
Fixed-Budget Parameter-Efficient Training with Frozen Encoders Improves Multimodal Chest X-Ray Classification. Multimodal chest X-Ray analysis often fine-tunes large vision-language models, which is computationally costly. We study parameter-efficient training (PET) strategies, including frozen encoders, BitFit, LoRA, ...
cv_vision
cs.CV
2025-12-25T05:02:19+00:00
2512.19440v1
Binary Kernel Logistic Regression: a sparsity-inducing formulation and a convergent decomposition training algorithm
Kernel logistic regression (KLR) is a widely used supervised learning method for binary and multi-class classification, which provides estimates of the conditional probabilities of class membership for the data points. Unlike other kernel methods such as Support Vector Machines (SVMs), KLRs are generally not sparse. Pr...
Binary Kernel Logistic Regression: a sparsity-inducing formulation and a convergent decomposition training algorithm. Kernel logistic regression (KLR) is a widely used supervised learning method for binary and multi-class classification, which provides estimates of the conditional probabilities of class membership for ...
lg_ml
cs.LG
2025-12-22T14:40:30+00:00
2601.09728v1
Eliminating Agentic Workflow for Introduction Generation with Parametric Stage Tokens
In recent years, using predefined agentic workflows to guide large language models (LLMs) for literature classification and review has become a research focus. However, writing research introductions is more challenging. It requires rigorous logic, coherent structure, and abstract summarization. Existing workflows ofte...
Eliminating Agentic Workflow for Introduction Generation with Parametric Stage Tokens. In recent years, using predefined agentic workflows to guide large language models (LLMs) for literature classification and review has become a research focus. However, writing research introductions is more challenging. It requires ...
cl_nlp
cs.CL
2025-12-28T12:51:36+00:00
2512.10004v2
Exploring LLMs for Scientific Information Extraction Using The SciEx Framework
Large language models (LLMs) are increasingly touted as powerful tools for automating scientific information extraction. However, existing methods and tools often struggle with the realities of scientific literature: long-context documents, multi-modal content, and reconciling varied and inconsistent fine-grained infor...
Exploring LLMs for Scientific Information Extraction Using The SciEx Framework. Large language models (LLMs) are increasingly touted as powerful tools for automating scientific information extraction. However, existing methods and tools often struggle with the realities of scientific literature: long-context documents,...
ai_general
cs.AI
2025-12-10T19:00:20+00:00
2512.23324v1
On Conformant Planning and Model-Checking of $\exists^*\forall^*$ Hyperproperties
We study the connection of two problems within the planning and verification community: Conformant planning and model-checking of hyperproperties. Conformant planning is the task of finding a sequential plan that achieves a given objective independent of non-deterministic action effects during the plan's execution. Hyp...
On Conformant Planning and Model-Checking of $\exists^*\forall^*$ Hyperproperties. We study the connection of two problems within the planning and verification community: Conformant planning and model-checking of hyperproperties. Conformant planning is the task of finding a sequential plan that achieves a given objecti...
ai_general
cs.AI
2025-12-29T09:20:29+00:00
2512.22324v1
DeMoGen: Towards Decompositional Human Motion Generation with Energy-Based Diffusion Models
Human motions are compositional: complex behaviors can be described as combinations of simpler primitives. However, existing approaches primarily focus on forward modeling, e.g., learning holistic mappings from text to motion or composing a complex motion from a set of motion concepts. In this paper, we consider the in...
DeMoGen: Towards Decompositional Human Motion Generation with Energy-Based Diffusion Models. Human motions are compositional: complex behaviors can be described as combinations of simpler primitives. However, existing approaches primarily focus on forward modeling, e.g., learning holistic mappings from text to motion o...
cv_vision
cs.CV
2025-12-26T15:06:09+00:00
2512.06266v1
Nanbeige4-3B Technical Report: Exploring the Frontier of Small Language Models
We present Nanbeige4-3B, a family of small-scale but high-performing language models. Pretrained on 23T high-quality tokens and finetuned on over 30 million diverse instructions, we extend the boundary of the scaling law for small language models. In pre-training, we design a Fine-Grained Warmup-Stable-Decay (FG-WSD) t...
Nanbeige4-3B Technical Report: Exploring the Frontier of Small Language Models. We present Nanbeige4-3B, a family of small-scale but high-performing language models. Pretrained on 23T high-quality tokens and finetuned on over 30 million diverse instructions, we extend the boundary of the scaling law for small language ...
cl_nlp
cs.CL
2025-12-06T03:36:27+00:00
2510.25718v1
Retrieval-Augmented Search for Large-Scale Map Collections with ColPali
Multimodal approaches have shown great promise for searching and navigating digital collections held by libraries, archives, and museums. In this paper, we introduce map-RAS: a retrieval-augmented search system for historic maps. In addition to introducing our framework, we detail our publicly-hosted demo for searching...
Retrieval-Augmented Search for Large-Scale Map Collections with ColPali. Multimodal approaches have shown great promise for searching and navigating digital collections held by libraries, archives, and museums. In this paper, we introduce map-RAS: a retrieval-augmented search system for historic maps. In addition to in...
ir_retrieval
cs.IR
2025-10-29T17:27:21+00:00
2601.02365v1
FUSE : Failure-aware Usage of Subagent Evidence for MultiModal Search and Recommendation
Multimodal creative assistants decompose user goals and route tasks to subagents for layout, styling, retrieval, and generation. Retrieval quality is pivotal, yet failures can arise at several stages: understanding user intent, choosing content types, finding candidates (recall), or ranking results. Meanwhile, sending ...
FUSE : Failure-aware Usage of Subagent Evidence for MultiModal Search and Recommendation. Multimodal creative assistants decompose user goals and route tasks to subagents for layout, styling, retrieval, and generation. Retrieval quality is pivotal, yet failures can arise at several stages: understanding user intent, ch...
ir_retrieval
cs.IR
2025-11-15T07:55:51+00:00
2512.23410v2
High-Dimensional Search, Low-Dimensional Solution: Decoupling Optimization from Representation
State-of-the-art models rely on massive widths despite exhibiting low Intrinsic Dimension (ID). We posit that this redundancy serves the non-convex optimization search rather than the final representation. We validate this hypothesis by decoupling the solution geometry via data-independent random projections, demonstra...
High-Dimensional Search, Low-Dimensional Solution: Decoupling Optimization from Representation. State-of-the-art models rely on massive widths despite exhibiting low Intrinsic Dimension (ID). We posit that this redundancy serves the non-convex optimization search rather than the final representation. We validate this h...
lg_ml
cs.LG
2025-12-29T12:13:15+00:00
2512.24615v1
Youtu-Agent: Scaling Agent Productivity with Automated Generation and Hybrid Policy Optimization
Existing Large Language Model (LLM) agent frameworks face two significant challenges: high configuration costs and static capabilities. Building a high-quality agent often requires extensive manual effort in tool integration and prompt engineering, while deployed agents struggle to adapt to dynamic environments without...
Youtu-Agent: Scaling Agent Productivity with Automated Generation and Hybrid Policy Optimization. Existing Large Language Model (LLM) agent frameworks face two significant challenges: high configuration costs and static capabilities. Building a high-quality agent often requires extensive manual effort in tool integrati...
ai_general
cs.AI
2025-12-31T04:17:36+00:00
2512.17374v3
Generative modeling of conditional probability distributions on the level-sets of collective variables
Given a probability distribution $μ$ in $\mathbb{R}^d$ represented by data, we study in this paper the generative modeling of the corresponding conditional probability distributions on the level-sets of a collective variable $ξ:\mathbb{R}^d \rightarrow \mathbb{R}^k$, where $1 \le k<d$. We propose a general and efficien...
Generative modeling of conditional probability distributions on the level-sets of collective variables. Given a probability distribution $μ$ in $\mathbb{R}^d$ represented by data, we study in this paper the generative modeling of the corresponding conditional probability distributions on the level-sets of a collective ...
stat_ml
stat.ML
2025-12-19T09:17:48+00:00
2512.06835v1
Decouple to Generalize: Context-First Self-Evolving Learning for Data-Scarce Vision-Language Reasoning
Recent vision-language models (VLMs) achieve remarkable reasoning through reinforcement learning (RL), which provides a feasible solution for realizing continuous self-evolving large vision-language models (LVLMs) in the era of experience. However, RL for VLMs requires abundant high-quality multimodal data, especially ...
Decouple to Generalize: Context-First Self-Evolving Learning for Data-Scarce Vision-Language Reasoning. Recent vision-language models (VLMs) achieve remarkable reasoning through reinforcement learning (RL), which provides a feasible solution for realizing continuous self-evolving large vision-language models (LVLMs) in...
ai_general
cs.AI
2025-12-07T13:17:31+00:00
2512.03727v1
Colored Markov Random Fields for Probabilistic Topological Modeling
Probabilistic Graphical Models (PGMs) encode conditional dependencies among random variables using a graph -nodes for variables, links for dependencies- and factorize the joint distribution into lower-dimensional components. This makes PGMs well-suited for analyzing complex systems and supporting decision-making. Recen...
Colored Markov Random Fields for Probabilistic Topological Modeling. Probabilistic Graphical Models (PGMs) encode conditional dependencies among random variables using a graph -nodes for variables, links for dependencies- and factorize the joint distribution into lower-dimensional components. This makes PGMs well-suite...
stat_ml
stat.ML
2025-12-03T12:16:07+00:00
2512.23671v2
Calibrated Multi-Level Quantile Forecasting
We develop an online method that guarantees calibration of quantile forecasts at multiple quantile levels simultaneously. In this work, a sequence of quantile forecasts is said to be calibrated provided that its $α$-level predictions are greater than or equal to the target value at an $α$ fraction of time steps, for ea...
Calibrated Multi-Level Quantile Forecasting. We develop an online method that guarantees calibration of quantile forecasts at multiple quantile levels simultaneously. In this work, a sequence of quantile forecasts is said to be calibrated provided that its $α$-level predictions are greater than or equal to the target v...
stat_ml
stat.ML
2025-12-29T18:25:36+00:00
2512.18225v1
GeoSense-AI: Fast Location Inference from Crisis Microblogs
This paper presents an applied AI pipeline for realtime geolocation from noisy microblog streams, unifying statistical hashtag segmentation, part-of-speech-driven proper-noun detection, dependency parsing around disaster lexicons, lightweight named-entity recognition, and gazetteer-grounded disambiguation to infer loca...
GeoSense-AI: Fast Location Inference from Crisis Microblogs. This paper presents an applied AI pipeline for realtime geolocation from noisy microblog streams, unifying statistical hashtag segmentation, part-of-speech-driven proper-noun detection, dependency parsing around disaster lexicons, lightweight named-entity rec...
cl_nlp
cs.CL
2025-12-20T05:46:57+00:00
2512.19228v1
Generation of Programmatic Rules for Document Forgery Detection Using Large Language Models
Document forgery poses a growing threat to legal, economic, and governmental processes, requiring increasingly sophisticated verification mechanisms. One approach involves the use of plausibility checks, rule-based procedures that assess the correctness and internal consistency of data, to detect anomalies or signs of ...
Generation of Programmatic Rules for Document Forgery Detection Using Large Language Models. Document forgery poses a growing threat to legal, economic, and governmental processes, requiring increasingly sophisticated verification mechanisms. One approach involves the use of plausibility checks, rule-based procedures t...
ai_general
cs.AI
2025-12-22T10:08:25+00:00
2512.03560v1
Reason-Plan-ReAct: A Reasoner-Planner Supervising a ReAct Executor for Complex Enterprise Tasks
Despite recent advances, autonomous agents often struggle to solve complex tasks in enterprise domains that require coordinating multiple tools and processing diverse data sources. This struggle is driven by two main limitations. First, single-agent architectures enforce a monolithic plan-execute loop, which directly c...
Reason-Plan-ReAct: A Reasoner-Planner Supervising a ReAct Executor for Complex Enterprise Tasks. Despite recent advances, autonomous agents often struggle to solve complex tasks in enterprise domains that require coordinating multiple tools and processing diverse data sources. This struggle is driven by two main limita...
ai_general
cs.AI
2025-12-03T08:28:40+00:00
2512.05013v1
Detecting Perspective Shifts in Multi-agent Systems
Generative models augmented with external tools and update mechanisms (or \textit{agents}) have demonstrated capabilities beyond intelligent prompting of base models. As agent use proliferates, dynamic multi-agent systems have naturally emerged. Recent work has investigated the theoretical and empirical properties of l...
Detecting Perspective Shifts in Multi-agent Systems. Generative models augmented with external tools and update mechanisms (or \textit{agents}) have demonstrated capabilities beyond intelligent prompting of base models. As agent use proliferates, dynamic multi-agent systems have naturally emerged. Recent work has inves...
ai_general
cs.AI
2025-12-04T17:24:56+00:00
2512.19676v3
Efficient Vision Mamba for MRI Super-Resolution via Hybrid Selective Scanning
Background: High-resolution MRI is critical for diagnosis, but long acquisition times limit clinical use. Super-resolution (SR) can enhance resolution post-scan, yet existing deep learning methods face fidelity-efficiency trade-offs. Purpose: To develop a computationally efficient and accurate deep learning framework f...
Efficient Vision Mamba for MRI Super-Resolution via Hybrid Selective Scanning. Background: High-resolution MRI is critical for diagnosis, but long acquisition times limit clinical use. Super-resolution (SR) can enhance resolution post-scan, yet existing deep learning methods face fidelity-efficiency trade-offs. Purpose...
cv_vision
cs.CV
2025-12-22T18:53:13+00:00
2512.18746v1
MemEvolve: Meta-Evolution of Agent Memory Systems
Self-evolving memory systems are unprecedentedly reshaping the evolutionary paradigm of large language model (LLM)-based agents. Prior work has predominantly relied on manually engineered memory architectures to store trajectories, distill experience, and synthesize reusable tools, enabling agents to evolve on the fly ...
MemEvolve: Meta-Evolution of Agent Memory Systems. Self-evolving memory systems are unprecedentedly reshaping the evolutionary paradigm of large language model (LLM)-based agents. Prior work has predominantly relied on manually engineered memory architectures to store trajectories, distill experience, and synthesize re...
cl_nlp
cs.CL
2025-12-21T14:26:14+00:00
2512.14014v1
MobileWorldBench: Towards Semantic World Modeling For Mobile Agents
World models have shown great utility in improving the task performance of embodied agents. While prior work largely focuses on pixel-space world models, these approaches face practical limitations in GUI settings, where predicting complex visual elements in future states is often difficult. In this work, we explore an...
MobileWorldBench: Towards Semantic World Modeling For Mobile Agents. World models have shown great utility in improving the task performance of embodied agents. While prior work largely focuses on pixel-space world models, these approaches face practical limitations in GUI settings, where predicting complex visual elem...
ai_general
cs.AI
2025-12-16T02:16:42+00:00
2512.18450v1
Agent-Based Output Drift Detection for Breast Cancer Response Prediction in a Multisite Clinical Decision Support System
Modern clinical decision support systems can concurrently serve multiple, independent medical imaging institutions, but their predictive performance may degrade across sites due to variations in patient populations, imaging hardware, and acquisition protocols. Continuous surveillance of predictive model outputs offers ...
Agent-Based Output Drift Detection for Breast Cancer Response Prediction in a Multisite Clinical Decision Support System. Modern clinical decision support systems can concurrently serve multiple, independent medical imaging institutions, but their predictive performance may degrade across sites due to variations in pat...
ai_general
cs.AI
2025-12-20T17:49:14+00:00
2511.15822v1
Atlas Gaussian processes on restricted domains and point clouds
In real-world applications, data often reside in restricted domains with unknown boundaries, or as high-dimensional point clouds lying on a lower-dimensional, nontrivial, unknown manifold. Traditional Gaussian Processes (GPs) struggle to capture the underlying geometry in such settings. Some existing methods assume a f...
Atlas Gaussian processes on restricted domains and point clouds. In real-world applications, data often reside in restricted domains with unknown boundaries, or as high-dimensional point clouds lying on a lower-dimensional, nontrivial, unknown manifold. Traditional Gaussian Processes (GPs) struggle to capture the under...
stat_ml
stat.ML
2025-11-19T19:28:05+00:00
2512.21106v2
Semantic Refinement with LLMs for Graph Representations
Graph-structured data exhibit substantial heterogeneity in where their predictive signals originate: in some domains, node-level semantics dominate, while in others, structural patterns play a central role. This structure-semantics heterogeneity implies that no graph learning model with a fixed inductive bias can gener...
Semantic Refinement with LLMs for Graph Representations. Graph-structured data exhibit substantial heterogeneity in where their predictive signals originate: in some domains, node-level semantics dominate, while in others, structural patterns play a central role. This structure-semantics heterogeneity implies that no g...
cl_nlp
cs.CL
2025-12-24T11:10:28+00:00
2601.00880v1
Universal Conditional Logic: A Formal Language for Prompt Engineering
We present Universal Conditional Logic (UCL), a mathematical framework for prompt optimization that transforms prompt engineering from heuristic practice into systematic optimization. Through systematic evaluation (N=305, 11 models, 4 iterations), we demonstrate significant token reduction (29.8%, t(10)=6.36, p < 0.001...
Universal Conditional Logic: A Formal Language for Prompt Engineering. We present Universal Conditional Logic (UCL), a mathematical framework for prompt optimization that transforms prompt engineering from heuristic practice into systematic optimization. Through systematic evaluation (N=305, 11 models, 4 iterations), w...
ai_general
cs.AI
2025-12-31T05:27:00+00:00
2510.04602v3
Wasserstein Gradient Flows for Scalable and Regularized Barycenter Computation
Wasserstein barycenters provide a principled approach for aggregating probability measures, while preserving the geometry of their ambient space. Existing discrete methods are not scalable as they assume access to the complete set of samples from the input measures. Meanwhile, neural network approaches do scale well, b...
Wasserstein Gradient Flows for Scalable and Regularized Barycenter Computation. Wasserstein barycenters provide a principled approach for aggregating probability measures, while preserving the geometry of their ambient space. Existing discrete methods are not scalable as they assume access to the complete set of sample...
stat_ml
stat.ML
2025-10-06T09:07:12+00:00
2512.21058v2
Beyond Pixel Simulation: Pathology Image Generation via Diagnostic Semantic Tokens and Prototype Control
In computational pathology, understanding and generation have evolved along disparate paths: advanced understanding models already exhibit diagnostic-level competence, whereas generative models largely simulate pixels. Progress remains hindered by three coupled factors: the scarcity of large, high-quality image-text co...
Beyond Pixel Simulation: Pathology Image Generation via Diagnostic Semantic Tokens and Prototype Control. In computational pathology, understanding and generation have evolved along disparate paths: advanced understanding models already exhibit diagnostic-level competence, whereas generative models largely simulate pix...
cv_vision
cs.CV
2025-12-24T08:52:08+00:00
2512.21000v1
CoSeNet: A Novel Approach for Optimal Segmentation of Correlation Matrices
In this paper, we propose a novel approach for the optimal identification of correlated segments in noisy correlation matrices. The proposed model is known as CoSeNet (Correlation Seg-mentation Network) and is based on a four-layer algorithmic architecture that includes several processing layers: input, formatting, re-...
CoSeNet: A Novel Approach for Optimal Segmentation of Correlation Matrices. In this paper, we propose a novel approach for the optimal identification of correlated segments in noisy correlation matrices. The proposed model is known as CoSeNet (Correlation Seg-mentation Network) and is based on a four-layer algorithmic ...
lg_ml
cs.LG
2025-12-24T06:55:43+00:00
2512.20991v1
FinAgent: An Agentic AI Framework Integrating Personal Finance and Nutrition Planning
The issue of limited household budgets and nutritional demands continues to be a challenge especially in the middle-income environment where food prices fluctuate. This paper introduces a price aware agentic AI system, which combines personal finance management with diet optimization. With household income and fixed ex...
FinAgent: An Agentic AI Framework Integrating Personal Finance and Nutrition Planning. The issue of limited household budgets and nutritional demands continues to be a challenge especially in the middle-income environment where food prices fluctuate. This paper introduces a price aware agentic AI system, which combines...
ai_general
cs.AI
2025-12-24T06:33:17+00:00
2512.21867v1
DPAR: Dynamic Patchification for Efficient Autoregressive Visual Generation
Decoder-only autoregressive image generation typically relies on fixed-length tokenization schemes whose token counts grow quadratically with resolution, substantially increasing the computational and memory demands of attention. We present DPAR, a novel decoder-only autoregressive model that dynamically aggregates ima...
DPAR: Dynamic Patchification for Efficient Autoregressive Visual Generation. Decoder-only autoregressive image generation typically relies on fixed-length tokenization schemes whose token counts grow quadratically with resolution, substantially increasing the computational and memory demands of attention. We present DP...
cv_vision
cs.CV
2025-12-26T05:03:47+00:00
2601.08838v1
Companion Agents: A Table-Information Mining Paradigm for Text-to-SQL
Large-scale Text-to-SQL benchmarks such as BIRD typically assume complete and accurate database annotations as well as readily available external knowledge, which fails to reflect common industrial settings where annotations are missing, incomplete, or erroneous. This mismatch substantially limits the real-world applic...
Companion Agents: A Table-Information Mining Paradigm for Text-to-SQL. Large-scale Text-to-SQL benchmarks such as BIRD typically assume complete and accurate database annotations as well as readily available external knowledge, which fails to reflect common industrial settings where annotations are missing, incomplete,...
cl_nlp
cs.CL
2025-12-17T07:11:55+00:00
2511.02986v1
Scalable Single-Cell Gene Expression Generation with Latent Diffusion Models
Computational modeling of single-cell gene expression is crucial for understanding cellular processes, but generating realistic expression profiles remains a major challenge. This difficulty arises from the count nature of gene expression data and complex latent dependencies among genes. Existing generative models ofte...
Scalable Single-Cell Gene Expression Generation with Latent Diffusion Models. Computational modeling of single-cell gene expression is crucial for understanding cellular processes, but generating realistic expression profiles remains a major challenge. This difficulty arises from the count nature of gene expression dat...
stat_ml
stat.ML
2025-11-04T20:44:12+00:00
2512.08819v1
Do Depth-Grown Models Overcome the Curse of Depth? An In-Depth Analysis
Gradually growing the depth of Transformers during training can not only reduce training cost but also lead to improved reasoning performance, as shown by MIDAS (Saunshi et al., 2024). Thus far, however, a mechanistic understanding of these gains has been missing. In this work, we establish a connection to recent work ...
Do Depth-Grown Models Overcome the Curse of Depth? An In-Depth Analysis. Gradually growing the depth of Transformers during training can not only reduce training cost but also lead to improved reasoning performance, as shown by MIDAS (Saunshi et al., 2024). Thus far, however, a mechanistic understanding of these gains ...
cl_nlp
cs.CL
2025-12-09T17:12:04+00:00
2512.21683v1
Contrastive Graph Modeling for Cross-Domain Few-Shot Medical Image Segmentation
Cross-domain few-shot medical image segmentation (CD-FSMIS) offers a promising and data-efficient solution for medical applications where annotations are severely scarce and multimodal analysis is required. However, existing methods typically filter out domain-specific information to improve generalization, which inadv...
Contrastive Graph Modeling for Cross-Domain Few-Shot Medical Image Segmentation. Cross-domain few-shot medical image segmentation (CD-FSMIS) offers a promising and data-efficient solution for medical applications where annotations are severely scarce and multimodal analysis is required. However, existing methods typica...
cv_vision
cs.CV
2025-12-25T14:00:17+00:00
2510.25544v2
Error Bounds and Optimal Schedules for Masked Diffusions with Factorized Approximations
Recently proposed generative models for discrete data, such as Masked Diffusion Models (MDMs), exploit conditional independence approximations to reduce the computational cost of popular Auto-Regressive Models (ARMs), at the price of some bias in the sampling distribution. We study the resulting computation-vs-accuracy...
Error Bounds and Optimal Schedules for Masked Diffusions with Factorized Approximations. Recently proposed generative models for discrete data, such as Masked Diffusion Models (MDMs), exploit conditional independence approximations to reduce the computational cost of popular Auto-Regressive Models (ARMs), at the price ...
stat_ml
stat.ML
2025-10-29T14:11:03+00:00
2512.19992v1
S$^3$IT: A Benchmark for Spatially Situated Social Intelligence Test
The integration of embodied agents into human environments demands embodied social intelligence: reasoning over both social norms and physical constraints. However, existing evaluations fail to address this integration, as they are limited to either disembodied social reasoning (e.g., in text) or socially-agnostic phys...
S$^3$IT: A Benchmark for Spatially Situated Social Intelligence Test. The integration of embodied agents into human environments demands embodied social intelligence: reasoning over both social norms and physical constraints. However, existing evaluations fail to address this integration, as they are limited to either ...
ai_general
cs.AI
2025-12-23T02:36:56+00:00
2601.04211v1
Qwerty AI: Explainable Automated Age Rating and Content Safety Assessment for Russian-Language Screenplays
We present Qwerty AI, an end-to-end system for automated age-rating and content-safety assessment of Russian-language screenplays according to Federal Law No. 436-FZ. The system processes full-length scripts (up to 700 pages in under 2 minutes), segments them into narrative units, detects content violations across five...
Qwerty AI: Explainable Automated Age Rating and Content Safety Assessment for Russian-Language Screenplays. We present Qwerty AI, an end-to-end system for automated age-rating and content-safety assessment of Russian-language screenplays according to Federal Law No. 436-FZ. The system processes full-length scripts (up ...
cl_nlp
cs.CL
2025-12-10T17:41:35+00:00
2512.10273v1
Reverse Thinking Enhances Missing Information Detection in Large Language Models
Large Language Models (LLMs) have demonstrated remarkable capabilities in various reasoning tasks, yet they often struggle with problems involving missing information, exhibiting issues such as incomplete responses, factual errors, and hallucinations. While forward reasoning approaches like Chain-of-Thought (CoT) and T...
Reverse Thinking Enhances Missing Information Detection in Large Language Models. Large Language Models (LLMs) have demonstrated remarkable capabilities in various reasoning tasks, yet they often struggle with problems involving missing information, exhibiting issues such as incomplete responses, factual errors, and ha...
ai_general
cs.AI
2025-12-11T04:25:17+00:00
2512.16967v1
Physics-Informed Lightweight Machine Learning for Aviation Visibility Nowcasting Across Multiple Climatic Regimes
Short-term prediction (nowcasting) of low-visibility and precipitation events is critical for aviation safety and operational efficiency. Current operational approaches rely on computationally intensive numerical weather prediction guidance and human-issued TAF products, which often exhibit conservative biases and limi...
Physics-Informed Lightweight Machine Learning for Aviation Visibility Nowcasting Across Multiple Climatic Regimes. Short-term prediction (nowcasting) of low-visibility and precipitation events is critical for aviation safety and operational efficiency. Current operational approaches rely on computationally intensive nu...
lg_ml
cs.LG
2025-12-18T11:40:21+00:00
2512.20976v1
XGrid-Mapping: Explicit Implicit Hybrid Grid Submaps for Efficient Incremental Neural LiDAR Mapping
Large-scale incremental mapping is fundamental to the development of robust and reliable autonomous systems, as it underpins incremental environmental understanding with sequential inputs for navigation and decision-making. LiDAR is widely used for this purpose due to its accuracy and robustness. Recently, neural LiDAR...
XGrid-Mapping: Explicit Implicit Hybrid Grid Submaps for Efficient Incremental Neural LiDAR Mapping. Large-scale incremental mapping is fundamental to the development of robust and reliable autonomous systems, as it underpins incremental environmental understanding with sequential inputs for navigation and decision-mak...
cv_vision
cs.CV
2025-12-24T06:08:50+00:00
2512.20086v1
Spatio-Temporal Graphs Beyond Grids: Benchmark for Maritime Anomaly Detection
Spatio-temporal graph neural networks (ST-GNNs) have achieved notable success in structured domains such as road traffic and public transportation, where spatial entities can be naturally represented as fixed nodes. In contrast, many real-world systems including maritime traffic lack such fixed anchors, making the cons...
Spatio-Temporal Graphs Beyond Grids: Benchmark for Maritime Anomaly Detection. Spatio-temporal graph neural networks (ST-GNNs) have achieved notable success in structured domains such as road traffic and public transportation, where spatial entities can be naturally represented as fixed nodes. In contrast, many real-wo...
lg_ml
cs.LG
2025-12-23T06:28:12+00:00
2512.15082v1
The Semantic Architect: How FEAML Bridges Structured Data and LLMs for Multi-Label Tasks
Existing feature engineering methods based on large language models (LLMs) have not yet been applied to multi-label learning tasks. They lack the ability to model complex label dependencies and are not specifically adapted to the characteristics of multi-label tasks. To address the above issues, we propose Feature Engi...
The Semantic Architect: How FEAML Bridges Structured Data and LLMs for Multi-Label Tasks. Existing feature engineering methods based on large language models (LLMs) have not yet been applied to multi-label learning tasks. They lack the ability to model complex label dependencies and are not specifically adapted to the ...
lg_ml
cs.LG
2025-12-17T04:58:44+00:00
2512.17488v1
TwinSegNet: A Digital Twin-Enabled Federated Learning Framework for Brain Tumor Analysis
Brain tumor segmentation is critical in diagnosis and treatment planning for the disease. Yet, current deep learning methods rely on centralized data collection, which raises privacy concerns and limits generalization across diverse institutions. In this paper, we propose TwinSegNet, which is a privacy-preserving feder...
TwinSegNet: A Digital Twin-Enabled Federated Learning Framework for Brain Tumor Analysis. Brain tumor segmentation is critical in diagnosis and treatment planning for the disease. Yet, current deep learning methods rely on centralized data collection, which raises privacy concerns and limits generalization across diver...
cv_vision
cs.CV
2025-12-19T11:59:41+00:00
2512.07694v1
Automated Generation of Custom MedDRA Queries Using SafeTerm Medical Map
In pre-market drug safety review, grouping related adverse event terms into standardised MedDRA queries or the FDA Office of New Drugs Custom Medical Queries (OCMQs) is critical for signal detection. We present a novel quantitative artificial intelligence system that understands and processes medical terminology and au...
Automated Generation of Custom MedDRA Queries Using SafeTerm Medical Map. In pre-market drug safety review, grouping related adverse event terms into standardised MedDRA queries or the FDA Office of New Drugs Custom Medical Queries (OCMQs) is critical for signal detection. We present a novel quantitative artificial int...
cl_nlp
cs.CL
2025-12-08T16:33:03+00:00
2512.16227v1
An Information-Theoretic Framework for Robust Large Language Model Editing
Large Language Models (LLMs) have become indispensable tools in science, technology, and society, enabling transformative advances across diverse fields. However, errors or outdated information within these models can undermine their accuracy and restrict their safe deployment. Developing efficient strategies for updat...
An Information-Theoretic Framework for Robust Large Language Model Editing. Large Language Models (LLMs) have become indispensable tools in science, technology, and society, enabling transformative advances across diverse fields. However, errors or outdated information within these models can undermine their accuracy a...
cl_nlp
cs.CL
2025-12-18T06:21:17+00:00
2510.11792v1
On Thompson Sampling and Bilateral Uncertainty in Additive Bayesian Optimization
In Bayesian Optimization (BO), additive assumptions can mitigate the twin difficulties of modeling and searching a complex function in high dimension. However, common acquisition functions, like the Additive Lower Confidence Bound, ignore pairwise covariances between dimensions, which we'll call \textit{bilateral uncer...
On Thompson Sampling and Bilateral Uncertainty in Additive Bayesian Optimization. In Bayesian Optimization (BO), additive assumptions can mitigate the twin difficulties of modeling and searching a complex function in high dimension. However, common acquisition functions, like the Additive Lower Confidence Bound, ignore...
stat_ml
stat.ML
2025-10-13T18:00:06+00:00
2512.07000v2
Benchmarking Deep Neural Networks for Modern Recommendation Systems
This paper presents a requirement-oriented benchmark of seven deep neural architectures, CNN, RNN, GNN, Autoencoder, Transformer, Neural Collaborative Filtering, and Siamese Networks, across three real-world datasets: Retail E-commerce, Amazon Products, and Netflix Prize. To ensure a fair and comprehensive comparison a...
Benchmarking Deep Neural Networks for Modern Recommendation Systems. This paper presents a requirement-oriented benchmark of seven deep neural architectures, CNN, RNN, GNN, Autoencoder, Transformer, Neural Collaborative Filtering, and Siamese Networks, across three real-world datasets: Retail E-commerce, Amazon Product...
ir_retrieval
cs.IR
2025-12-07T21:06:24+00:00
2512.17385v1
UCoder: Unsupervised Code Generation by Internal Probing of Large Language Models
Large language models (LLMs) have demonstrated remarkable capabilities in code generation tasks. However, their effectiveness heavily relies on supervised training with extensive labeled (e.g., question-answering pairs) or unlabeled datasets (e.g., code snippets), which are often expensive and difficult to obtain at sc...
UCoder: Unsupervised Code Generation by Internal Probing of Large Language Models. Large language models (LLMs) have demonstrated remarkable capabilities in code generation tasks. However, their effectiveness heavily relies on supervised training with extensive labeled (e.g., question-answering pairs) or unlabeled data...
cl_nlp
cs.CL
2025-12-19T09:42:04+00:00
2512.05318v1
To Think or Not to Think: The Hidden Cost of Meta-Training with Excessive CoT Examples
Chain-of-thought (CoT) prompting combined with few-shot in-context learning (ICL) has unlocked significant reasoning capabilities in large language models (LLMs). However, ICL with CoT examples is ineffective on novel tasks when the pre-training knowledge is insufficient. We study this problem in a controlled setting u...
To Think or Not to Think: The Hidden Cost of Meta-Training with Excessive CoT Examples. Chain-of-thought (CoT) prompting combined with few-shot in-context learning (ICL) has unlocked significant reasoning capabilities in large language models (LLMs). However, ICL with CoT examples is ineffective on novel tasks when the...
cl_nlp
cs.CL
2025-12-04T23:28:23+00:00
2512.10882v4
Computational emotion analysis with multimodal LLMs: Current evidence on an emerging methodological opportunity
Research increasingly leverages audio-visual materials to analyze emotions in political communication. Multimodal large language models (mLLMs) promise to enable such analyses through in-context learning. However, we lack systematic evidence on whether current mLLMs can reliably measure emotions in real-world political...
Computational emotion analysis with multimodal LLMs: Current evidence on an emerging methodological opportunity. Research increasingly leverages audio-visual materials to analyze emotions in political communication. Multimodal large language models (mLLMs) promise to enable such analyses through in-context learning. Ho...
cl_nlp
cs.CL
2025-12-11T18:11:46+00:00
2512.21736v3
SyncAnyone: Implicit Disentanglement via Progressive Self-Correction for Lip-Syncing in the wild
High-quality AI-powered video dubbing demands precise audio-lip synchronization, high-fidelity visual generation, and faithful preservation of identity and background. Most existing methods rely on a mask-based training strategy, where the mouth region is masked in talking-head videos, and the model learns to synthesiz...
SyncAnyone: Implicit Disentanglement via Progressive Self-Correction for Lip-Syncing in the wild. High-quality AI-powered video dubbing demands precise audio-lip synchronization, high-fidelity visual generation, and faithful preservation of identity and background. Most existing methods rely on a mask-based training st...
cv_vision
cs.CV
2025-12-25T16:49:40+00:00
2511.04221v1
Coordination-Free Lane Partitioning for Convergent ANN Search
Production vector search systems often fan out each query across parallel lanes (threads, replicas, or shards) to meet latency service-level objectives (SLOs). In practice, these lanes rediscover the same candidates, so extra compute does not increase coverage. We present a coordination-free lane partitioner that turns...
Coordination-Free Lane Partitioning for Convergent ANN Search. Production vector search systems often fan out each query across parallel lanes (threads, replicas, or shards) to meet latency service-level objectives (SLOs). In practice, these lanes rediscover the same candidates, so extra compute does not increase cover...
ir_retrieval
cs.IR
2025-11-06T09:36:18+00:00
2512.09266v1
Robust and Sparse Estimation of Unbounded Density Ratio under Heavy Contamination
We examine the non-asymptotic properties of robust density ratio estimation (DRE) in contaminated settings. Weighted DRE is the most promising among existing methods, exhibiting doubly strong robustness from an asymptotic perspective. This study demonstrates that Weighted DRE achieves sparse consistency even under heav...
Robust and Sparse Estimation of Unbounded Density Ratio under Heavy Contamination. We examine the non-asymptotic properties of robust density ratio estimation (DRE) in contaminated settings. Weighted DRE is the most promising among existing methods, exhibiting doubly strong robustness from an asymptotic perspective. Th...
stat_ml
stat.ML
2025-12-10T02:43:21+00:00
2512.05734v1
KANFormer for Predicting Fill Probabilities via Survival Analysis in Limit Order Books
This paper introduces KANFormer, a novel deep-learning-based model for predicting the time-to-fill of limit orders by leveraging both market- and agent-level information. KANFormer combines a Dilated Causal Convolutional network with a Transformer encoder, enhanced by Kolmogorov-Arnold Networks (KANs), which improve no...
KANFormer for Predicting Fill Probabilities via Survival Analysis in Limit Order Books. This paper introduces KANFormer, a novel deep-learning-based model for predicting the time-to-fill of limit orders by leveraging both market- and agent-level information. KANFormer combines a Dilated Causal Convolutional network wit...
ai_general
cs.AI
2025-12-05T14:15:02+00:00
2512.15420v2
FlowBind: Efficient Any-to-Any Generation with Bidirectional Flows
Any-to-any generation seeks to translate between arbitrary subsets of modalities, enabling flexible cross-modal synthesis. Despite recent success, existing flow-based approaches are challenged by their inefficiency, as they require large-scale datasets often with restrictive pairing constraints, incur high computationa...
FlowBind: Efficient Any-to-Any Generation with Bidirectional Flows. Any-to-any generation seeks to translate between arbitrary subsets of modalities, enabling flexible cross-modal synthesis. Despite recent success, existing flow-based approaches are challenged by their inefficiency, as they require large-scale datasets...
lg_ml
cs.LG
2025-12-17T13:08:18+00:00