15
MobiDiff: Semantic-Aware Multi-Channel Discrete Diffusion for Human Mobility Data Generation
The MobiDiff framework uses discrete diffusion to generate realistic human mobility data while preserving privacy and modeling semantic events.
Top stories in Research
15
Spatio-Temporal Scheduling Prediction Under Backhaul Delay for Resilient Coordinated Beamforming
This paper addresses backhaul latency in 5G networks by using predictive scheduling to maintain the performance of coordinated beamforming.
15
PS4: Proxy-Supervised Joint Training for Real Target Speaker Extraction
The PS4 framework uses proxy supervision and a newly compiled dataset to train target speaker extraction models on real conversational audio without clean reference signals.
15
Goal-Driven Reasoning in DatalogMTL with Magic Sets
A new goal-driven reasoning method for DatalogMTL utilizes the magic sets technique to improve computational efficiency in temporal reasoning tasks.
14
Drift-Aware Temporal Graph Rewiring (DATGR) for Adaptive Semantic Modeling in Biomedical Text
The Drift-Aware Temporal Graph Rewiring framework dynamically updates concept relationships to help language models adapt to evolving terminology in biomedical literature.
14
Agentic AI and Retrieval-Augmented Models in Straight-Through Underwriting
This paper explores the integration of agentic AI and retrieval-augmented generation models to automate complex decision workflows in actuarial underwriting.
14
MeCo: One-Step MeanFlow-based Corrector for Multi-Channel Speech Separation
The paper introduces MeCo, a one-step generative corrector based on MeanFlow designed to improve human listening quality in multi-channel speech separation.
14
KG-SoftMAP: Soft Knowledge-Graph Priors for Bayesian Network Structure Learning from Sparse Discrete Data
Researchers have introduced KG-SoftMAP, a method that integrates weighted knowledge graphs as soft priors to improve Bayesian network structure learning from sparse datasets.
14
Phase-Preserving Trimodal Transformer for Tropical Forest Biomass Estimation Using Optical and PolInSAR Data
A new physics-informed deep learning architecture called the Trimodal Coherent Co-attention Transformer has been proposed to improve above-ground biomass estimation in tropical forests.
14
Predicting Male Fertility Using Machine Learning: A Semen Parameters Based Analysis with the VISEM Dataset
The study evaluates the effectiveness of various machine learning algorithms in classifying male fertility status using semen analysis data from the VISEM dataset.
14
Adversarial Social Epistemology for Assemblies of Humans and Large Language Models
This paper outlines a theoretical framework for analyzing how information is shared and potentially distorted in environments where humans interact closely with large language models.
14
A Multi-cluster Boundary Learning Method for Out-of-Scope Intent Detection via MiniLM Embedding
A new multi-cluster boundary learning method using MiniLM embeddings has been proposed to improve out-of-scope intent detection in human-machine interaction systems.
14
Aleena: Alignment Agent for Research Software Engineering Collaborations
The Aleena agent framework aims to maintain shared project understanding and design rationale across diverse communication channels in collaborative research software engineering.
14
ICDAR 2026 HIPE-OCRepair Competition on LLM-Assisted OCR Post-Correction for Historical Documents
This paper outlines the results of a competition evaluating the ability of large language models to correct optical character recognition errors in historical texts.
14
LDFE: Laplacian Decoupled Feature Enhancement Block for Dual-Stream CNN-based RGB-IR Object Detection
The proposed Laplacian Decoupled Feature Enhancement block improves object detection in challenging conditions by better fusing visible and infrared image features.
14
INTENT: An LSTM Framework for Vehicle Intention Prediction in Intersection Scenarios with Comprehensive Ablation Analysis
Researchers developed an LSTM-based framework designed to predict vehicle intentions at complex road intersections to improve autonomous driving safety.
14
Evaluating the Effect of Frame Rate in Sequence-Based Classification of Autism-Related Self-Stimulatory Hand Idiosyncrasies
This study evaluates how temporal sampling rates and neural network architectures affect the automated detection of autism-related self-stimulatory behaviors in video.
14
A Graph Neural Network Model for Real-Time Gesture Recognition Based on sEMG Signals
Researchers developed a graph neural network model that analyzes muscle activation patterns to achieve real-time hand gesture recognition for prosthetics and augmented reality.
14
Idiobionics: The Unification of Privacy and Intelligent Robotic Prostheses
The paper introduces a framework for balancing user privacy with the data requirements of intelligent, sensor-enabled robotic prostheses.
14
AI-integrated models for assessing agricultural resilience
A new paper introduces an AI-powered tool that combines economic and biophysical models to help policymakers analyze agricultural supply chain disruptions using natural language.
13
只剩7天！第三届蚂蚁InTech奖申报即将截止，图灵奖得主坐镇评审
The submission deadline for the third Ant InTech Award, which features a judging panel including Turing Award winners, is approaching in seven days.
13
The power of collaboration: How we can reduce traffic congestion
Google Research published work exploring algorithmic and collaborative approaches to mitigating traffic congestion.
13
LLM for the development of FCM
This paper explores the capability of local large language models, such as Qwen2.5-32B, to extract quantitative data from text to develop fuzzy cognitive maps.
13
COBART: Controlled, Optimized, Bidirectional and Auto-Regressive Transformer for Ad Headline Generation
A new fine-tuning method using prefix control tokens with BART allows for more customizable and click-optimized generation of online advertisement headlines.