Communications 2025

934 papers
Communications 2025 published 932 papers, with a strong focus on the intersection of machine learning and the biological sciences. Key research directions included recovering biological signals from sparse transcriptomics, developing protein language models for sequence representation, and utilizing Euclidean equivariant representations for molecular geometry. Clinical applications were also a major theme, specifically regarding the grounding of large language models in authoritative evidence to enhance reliability and the longitudinal modeling of brain development and disease trajectories. Methodological trends highlighted the management of stochastic instability in large language model reasoning and advancements in deep learning for live cell imaging restoration. The year's contributions spanned diverse applications, such as temporal hypergraph convolutional networks for EEG emotion recognition, video anomaly detection using multi-scale feature fusion, and studies linking sensory perception to boredom in individuals with ADHD. Prolific authors for this edition included Minh–Triet Tran, Quanying Liu, and Pablo Rivas. Other research addressed image aesthetic quality assessment through generative prompts and real estate market prediction using entropy clustering and sentiment-based frameworks.

Top Research Directions

Research directions most represented by Communications 2025 papers.

Overcoming Data Sparsity in Single Cell Sequencing 5
The Generalizability Gap Between Healthy and Clinical Brains 3
Contextual Embeddings from Protein Language Models 3
Remodeling Protein Active Sites Through Targeted Mutagenesis 3
Grounding Language Models with External Knowledge Retrieval 3
The Longitudinal Validation Gap in Developmental Science 3
Computing Minimum Energy Paths in Molecular Systems 2
Hierarchical Bayesian Parameter Estimation through Partial Pooling 2
Mitigating Off-Target Risks in Precision Genome Editing 2
The Mechanics of Digital Self Discrepancy Anxiety 2
Cross-Species Generalization of Biological Deep Learning Models 2
The Information Cost of Discretizing Continuous Data 2
Real-Time Processing for High-Throughput Scientific Sensors 2
Alternative Binders and Industrial Waste Upcycling in Concrete 2
Bridging the Spatial Scale Mismatch in Climate Projections 2

Top Authors

AuthorPapers
Quanying Liu7
Pablo Rivas6
Thanh Duc Ngo4
Azmine Toushik Wasi3
Chunhua Yang3
Dongnan Liu3
Jionglong Su3
Kang Liu3
Leilei Sun3
Mingli Song3
Tongyu Zhu3
Xueqin Wang3
Buzhou Tang2
Catie Chang2
Charlotte Deane2
Christopher Li Hsian Chen2
Chuang Liu2
Daniel Dajun Zeng2
Daniel Haeufle2
Daoqiang Zhang2

Impactful Papers

Papers from Communications 2025 that shaped research clusters.

Influential

Papers

Title
Fast and Accurate Visuomotor Imitation Learning via 2D Consistency Flow Matching Policy
Fast and Efficient Layer-Aware Container Vulnerability Patching in Edge Computing
Fast randomized least‐squares solvers can be just as accurate and stable as classical direct solvers
Feature Crossing Attention Network with Field-Augmented Relational Tensors for CTR Prediction
FedAST: Semi-supervised Federated Learning via Adaptively Switched Training
FedDTG:Federated Data-Free Knowledge Distillation via Three-Player Generative Adversarial Networks
Federated Learning with Frequency Estimation for Smart Meter Systems
Federated task-adaptive learning for personalized selection of human IVF-derived embryos
FedGF: Layer-Wise Federated Learning with Group Fairness Guarantees
FedPSED: Federated Public Safety Emergency Detection Based on Adaptive Aggregation Strategy
Few-shot learning for novel object detection in autonomous driving
Financing Strategies of New Energy Vehicle Supply Chain with Capital Constraints of Vehicle Manufacturers
Fine-Grained Feature Fusion for Self-supervised Multimodal Sentiment Analysis
Fine-Grained Spatial-Temporal Framework for Engagement Prediction
Finite groups with 2-maximal subgroups of Sylow subgroups are ICCE -subgroups
Firework Swarm Learning (FSL): A Universal Machine Learning Framework
First-order Euler method is effective for computation of associative-memory network of Kuramoto oscillators
FISHER: An Efficient Sim2sim Training Framework Dedicated in Multi-AUV Target Tracking via Learning from Demonstrations
FlashMLA-ETAP: Efficient Transpose Attention Pipeline for Accelerating MLA Inference on NVIDIA H20 GPUs
Flexible integration of spatial and expression information for precise spot embedding via ZINB-based graph-enhanced autoencoder
Flight Trajectory Prediction Based on TCN-Transformer Method
FOCAL-UAT: Flexible One-Class Acoustic Representation Learning for Underwater Acoustic Target Recognition
Follow the Crowd but Keep Myself: Understanding the Role of Crowd Wisdom and Individual User Taste in Rating Prediction
Forecasting Energy Availability in Local Energy Communities via LSTM Federated Learning
Framework for Efficient Automated Alarm Analysis in Intelligent Connected Vehicles
Free Field Realisation of Boundary Vertex Algebras for Abelian Gauge Theories in Three Dimensions
Frequency-Guided Wavelet Transformer for Effective Moiré Pattern Removal
Frequency Spatial Hybrid Joint Source-Channel Coding with Channel Adaptive Attention Mechanism
From Latent to Engine Manifolds: Analyzing ImageBind’s Multimodal Embedding Space
Functional MRI signatures of autonomic physiology in aging
FusionASAG: An LLM-Enhanced Automatic Short Answer Grading Model for Subjective Questions in Online Education
Fusion of Domain Dependent Sensitive Semantics and Large Language Model for Research Text Security Classification Category
GallNet: A Hybrid Architecture for Gallbladder Cancer Detection
GEMs-LLM: Integrating Large Language Models with Goal-Aware Exploration for RL-Based Portfolio Optimization
Generalized Frobenius Manifolds with Non-flat Unity and Integrable Hierarchies
Generating Rationales Based on Human Explanations for Constrained Optimization
Generating synthetic task-based brain fingerprints for population neuroscience using deep learning
Generative Text Prompts for Image Aesthetic Quality Assessment
Gene signatures derived from transcriptomic-causal networks stratify colorectal cancer patients for effective targeted therapy
Genome-wide RNAi screening in C. elegans reveals OXPHOS and pyrimidine synthesis pathways as PKA regulators
Germline genetic scores associated with cancer gene expression and immune responses across multiple cancer types
Gesture encoding in human left precentral gyrus neuronal ensembles
GGMNet: Gradient Imagery-Guided Multi-scale Feature Fusion for Remote Sensing Change Detection
Global coastal water clarity has increased due to human intervention
Global non-uniformity in biophysical surface temperature responses to cropland expansion over non-forest vegetation
Global-View 3D Reconstruction of Complex Steel Components with Cobot-Driven Line Structured Light Sensor
GMFVAD: Using Grained Multi-modal Feature to Improve Video Anomaly Detection
GMT: Novel View Synthesis of Road Regions in Sparse-View Gaussian Splatting that Uses Motion Trajectory Priors
Grading University Students with LLMs: Performance and Acceptance of a Canvas-Based Automation
Granular-ball Computing Based Fireworks Algorithm for Global Optimization of Multi-modal Functions

« prev Page 9/19 next »