📊 Статистика дайджестов

Всего дайджестов: 34123 Добавлено сегодня: 101

Последнее обновление: сегодня
📄 Neighbor GRPO: Contrastive ODE Policy Optimization Aligns Flow Models
2025-11-25
Авторы:

Dailan He, Guanlin Feng, Xingtong Ge, Yazhe Niu, Yi Zhang, Bingqi Ma, Guanglu Song, Yu Liu, Hongsheng Li

Саммари на русском не найдено
Доступные поля: ['id', 'arxiv_id', 'title', 'authors', 'abstract', 'summary_ru', 'categories', 'published_date', 'created_at']
Annotation:
Group Relative Policy Optimization (GRPO) has shown promise in aligning image and video generative models with human preferences. However, applying it to modern flow matching models is challenging because of its deterministic sampling paradigm. Current methods address this issue by converting Ordinary Differential Equations (ODEs) to Stochastic Differential Equations (SDEs), which introduce stochasticity. However, this SDE-based GRPO suffers from issues of inefficient credit assignment and incom...
ID: 2511.16955v1 cs.CV, cs.LG, eess.IV
arXiv PDF
📄 Saving Foundation Flow-Matching Priors for Inverse Problems
2025-11-22
Авторы:

Yuxiang Wan, Ryan Devera, Wenjie Zhang, Ju Sun

Саммари на русском не найдено
Доступные поля: ['id', 'arxiv_id', 'title', 'authors', 'abstract', 'summary_ru', 'categories', 'published_date', 'created_at']
Annotation:
Foundation flow-matching (FM) models promise a universal prior for solving inverse problems (IPs), yet today they trail behind domain-specific or even untrained priors. How can we unlock their potential? We introduce FMPlug, a plug-in framework that redefines how foundation FMs are used in IPs. FMPlug combines an instance-guided, time-dependent warm-start strategy with a sharp Gaussianity regularization, adding problem-specific guidance while preserving the Gaussian structures. This leads to a s...
ID: 2511.16520v1 cs.LG, eess.IV
arXiv PDF
📄 Reconstruction of three-dimensional shapes of normal and disease-related erythrocytes from partial observations using multi-fidelity neural networks
2025-11-21
Авторы:

Haizhou Wen, He Li, Zhen Li

Саммари на русском не найдено
Доступные поля: ['id', 'arxiv_id', 'title', 'authors', 'abstract', 'summary_ru', 'categories', 'published_date', 'created_at']
Annotation:
Reconstruction of 3D erythrocyte or red blood cell (RBC) morphology from partial observations, such as microscope images, is essential for understanding the physiology of RBC aging and the pathology of various RBC disorders. In this study, we propose a multi-fidelity neural network (MFNN) approach to fuse high-fidelity cross-sections of an RBC, with a morphologically similar low-fidelity reference 3D RBC shape to recover its full 3D surface. The MFNN predictor combines a convolutional neural net...
ID: 2511.14962v1 physics.comp-ph, cs.LG, eess.IV, physics.bio-ph, q-bio.QM
arXiv PDF
📄 Quality-Controlled Multimodal Emotion Recognition in Conversations with Identity-Based Transfer Learning and MAMBA Fusion
2025-11-21
Авторы:

Zanxu Wang, Homayoon Beigi

Саммари на русском не найдено
Доступные поля: ['id', 'arxiv_id', 'title', 'authors', 'abstract', 'summary_ru', 'categories', 'published_date', 'created_at']
Annotation:
This paper addresses data quality issues in multimodal emotion recognition in conversation (MERC) through systematic quality control and multi-stage transfer learning. We implement a quality control pipeline for MELD and IEMOCAP datasets that validates speaker identity, audio-text alignment, and face detection. We leverage transfer learning from speaker and face recognition, assuming that identity-discriminative embeddings capture not only stable acoustic and Facial traits but also person-specif...
ID: 2511.14969v1 eess.AS, cs.AI, cs.LG, eess.IV, eess.SP
arXiv PDF
📄 Data-driven Prediction of Species-Specific Plant Responses to Spectral-Shifting Films from Leaf Phenotypic and Photosynthetic Traits
2025-11-20
Авторы:

Jun Hyeun Kang, Jung Eek Son, Tae In Ahn

Саммари на русском не найдено
Доступные поля: ['id', 'arxiv_id', 'title', 'authors', 'abstract', 'summary_ru', 'categories', 'published_date', 'created_at']
Annotation:
The application of spectral-shifting films in greenhouses to shift green light to red light has shown variable growth responses across crop species. However, the yield enhancement of crops under altered light quality is related to the collective effects of the specific biophysical characteristics of each species. Considering only one attribute of a crop has limitations in understanding the relationship between sunlight quality adjustments and crop growth performance. Therefore, this study aims t...
ID: 2511.15173v1 q-bio.QM, cs.CV, cs.LG, eess.IV
arXiv PDF
📄 FERMI-ML: A Flexible and Resource-Efficient Memory-In-Situ SRAM Macro for TinyML acceleration
2025-11-19
Авторы:

Mukul Lokhande, Akash Sankhe, S. V. Jaya Chand, Santosh Kumar Vishvakarma

Саммари на русском не найдено
Доступные поля: ['id', 'arxiv_id', 'title', 'authors', 'abstract', 'summary_ru', 'categories', 'published_date', 'created_at']
Annotation:
The growing demand for low-power and area-efficient TinyML inference on AIoT devices necessitates memory architectures that minimise data movement while sustaining high computational efficiency. This paper presents FERMI-ML, a Flexible and Resource-Efficient Memory-In-Situ (MIS) SRAM macro designed for TinyML acceleration. The proposed 9T XNOR-based RX9T bit-cell integrates a 5T storage cell with a 4T XNOR compute unit, enabling variable-precision MAC and CAM operations within the same array. A ...
ID: 2511.12544v1 cs.AR, cs.ET, cs.LG, eess.IV
arXiv PDF
📄 Synergy vs. Noise: Performance-Guided Multimodal Fusion For Biochemical Recurrence-Free Survival in Prostate Cancer
2025-11-18
Авторы:

Seth Alain Chang, Muhammad Mueez Amjad, Noorul Wahab, Ethar Alzaid, Nasir Rajpoot, Adam Shephard

Саммари на русском не найдено
Доступные поля: ['id', 'arxiv_id', 'title', 'authors', 'abstract', 'summary_ru', 'categories', 'published_date', 'created_at']
Annotation:
Multimodal deep learning (MDL) has emerged as a transformative approach in computational pathology. By integrating complementary information from multiple data sources, MDL models have demonstrated superior predictive performance across diverse clinical tasks compared to unimodal models. However, the assumption that combining modalities inherently improves performance remains largely unexamined. We hypothesise that multimodal gains depend critically on the predictive quality of individual modali...
ID: 2511.11452v1 q-bio.QM, cs.CV, cs.LG, eess.IV
arXiv PDF
📄 Few-Shot Multimodal Medical Imaging: A Theoretical Framework
2025-11-07
Авторы:

Md Talha Mohsin, Ismail Abdulrashid

Саммари на русском не найдено
Доступные поля: ['id', 'arxiv_id', 'title', 'authors', 'abstract', 'summary_ru', 'categories', 'published_date', 'created_at']
Annotation:
Medical imaging relies heavily on large, labeled datasets. But, unfortunately, they are not always easily accessible in clinical settings. Additionally, many practitioners often face various structural obstacles like limited data availability, fragmented data systems, and unbalanced datasets. These barriers often lead to the increased diagnostic uncertainty, underrepresentation of certain conditions, reduced model robustness, and biased diagnostic decisions. In response to these challenges, appr...
ID: 2511.01140v1 stat.ML, cs.AI, cs.CV, cs.LG, eess.IV
arXiv PDF
📄 Accounting for Underspecification in Statistical Claims of Model Superiority
2025-11-06
Авторы:

Thomas Sanchez, Pedro M. Gordaliza, Meritxell Bach Cuadra

Саммари на русском не найдено
Доступные поля: ['id', 'arxiv_id', 'title', 'authors', 'abstract', 'summary_ru', 'categories', 'published_date', 'created_at']
Annotation:
Machine learning methods are increasingly applied in medical imaging, yet many reported improvements lack statistical robustness: recent works have highlighted that small but significant performance gains are highly likely to be false positives. However, these analyses do not take \emph{underspecification} into account -- the fact that models achieving similar validation scores may behave differently on unseen data due to random initialization or training dynamics. Here, we extend a recent stati...
ID: 2511.02453v1 cs.LG, eess.IV
arXiv PDF
📄 Dark-Field X-Ray Imaging Significantly Improves Deep-Learning based Detection of Synthetic Early-Stage Lung Tumors in Preclinical Models
2025-11-04
Авторы:

Joyoni Dey, Hunter C. Meyer, Murtuza S. Taqi

Саммари на русском не найдено
Доступные поля: ['id', 'arxiv_id', 'title', 'authors', 'abstract', 'summary_ru', 'categories', 'published_date', 'created_at']
Annotation:
Low-dose computed tomography (LDCT) is the current standard for lung cancer screening, yet its adoption and accessibility remain limited. Many regions lack LDCT infrastructure, and even among those screened, early-stage cancer detection often yield false positives, as shown in the National Lung Screening Trial (NLST) with a sensitivity of 93.8 percent and a false-positive rate of 26.6 percent. We aim to investigate whether X-ray dark-field imaging (DFI) radiograph, a technique sensitive to small...
ID: 2510.27679v1 physics.med-ph, cs.CV, cs.LG, eess.IV, physics.optics
arXiv PDF
Показано 11 - 20 из 56 записей