3
Towards an Intention Abstraction Layer for Autonomous Industrial Systems
This paper proposes an intention abstraction layer to help coordinate autonomous industrial subsystems and prevent goal conflicts in shared environments.
3
All Polarized but Still Different: a Multi-factorial Metric to Discriminate between Polarization Behaviors on Social Media
Researchers introduced GRAIL, a metric designed to provide a granular analysis of individual polarization behaviors on social media platforms.
3
Benchmarking Multimodal Large Language Models for Scientific Visualization Literacy
Researchers assessed six multimodal large language models on their ability to interpret and understand scientific visualizations.
3
BrainPilot: Automating Brain Discovery with Agentic Research
The authors introduce an agentic AI system designed to automate complex research workflows within the field of neuroscience.
3
Collaborative Spatial Learning with Multi-LLM Agents in Networked Social Experiments
This study explores how multi-LLM agent networks manage the balance between exploration and exploitation when solving spatial problems in different communication structures.
3
Analytic Abduction: Causal Decomposition and Governed Commitment for Human--AI Coordination
A new framework for analytic abduction is proposed to improve human-AI coordination by governing how models commit to causal explanations.
3
InCarEmo: A Multimodal Dataset for In-Cabin Emotion Recognition and Driver State Monitoring
Researchers released a multimodal dataset combining visual and conversational data to improve emotion recognition and state monitoring for automotive safety systems.
3
Reachability-Aware Pretraining for Efficient Target-Oriented Path Exploration in Temporal Knowledge Graph Reasoning
This paper proposes a reachability-aware pretraining method to improve the efficiency of reinforcement learning-based path exploration in temporal knowledge graph reasoning.
3
CFM-Bench: A Unified Multi-Domain, Multi-Task Benchmark for Channel Foundation Models
A new unified benchmark has been introduced to standardize the evaluation of foundation models applied to wireless communication tasks.
3
VLT: A Vision-Language-Time Series Multimodal Foundation Model for Industrial Intelligence
The VLT foundation model integrates vision, language, and time-series data to improve industrial equipment monitoring and prognostic health management.
3
Measuring How Students Rely on Generative AI in Academic Writing: Development and Multi-Source Validation of the Generative AI Reliance Types Scale (GenAI-RTS)
A new psychometric scale has been developed to measure how undergraduate students rely on generative AI tools for academic writing tasks.
3
Are Performance-Optimization Benchmarks Reliably Measuring Coding Agents?
A new study highlights that current performance-optimization benchmarks for coding agents may be unreliable due to issues with runtime instability and inconsistent scoring metrics.
3
Self-Evolving Human-Centered Framework for Explainable Depression Symptom Annotation
A new framework aims to improve the quality and explainability of depression symptom annotations by aligning them with clinical diagnostic standards.
3
Toward Robust In-Context Segmentation via Concept Guidance
This research proposes a concept-guided approach to improve the robustness and stability of in-context image segmentation models.
3
VOiLA: Vectorized Online Planning with Learned Diffusion Models for POMDP Agents
Researchers introduced a vectorized online planning approach using diffusion models to improve robot decision-making under uncertainty.
3
Grow-Prune-Freeze Networks: Adaptive & Continual Learning Technique for Olfactory Navigation
The Grow-Prune-Freeze network architecture provides a method for agents to perform continual learning in dynamic environments, specifically applied to olfactory navigation tasks.
3
Neuro-Symbolic Strong-AI Robots with Closed Knowledge Assumption: Learning and Deductions
The paper explores a neuro-symbolic approach to AGI, focusing on how robots can learn and perform deductions using a closed knowledge assumption.
3
LBA: Textual Hard-Label Adversarial Attack under Low Query Budgets
A new method for generating adversarial text examples has been proposed to improve efficiency under low query budget constraints.
3
REST: Receding Horizon Explorative Steiner Tree for Zero-Shot Object-Goal Navigation
A new navigation method called REST is proposed to improve zero-shot object-goal navigation by optimizing subgoal selection in unknown environments.
3
Quality-Aware Robust Multi-View Clustering for Heterogeneous Observation Noise
A new method for multi-view clustering has been proposed to handle heterogeneous observation noise in data, moving beyond binary clean-or-corrupted assumptions.
3
Eta Given Delta: Defining LLM Tool Efficiency With Marginal Tool Utility
A new quantitative metric has been developed to measure the efficiency and utility of tool calls within LLM agent trajectories.
3
"Skill Issues'': Data-Centric Optimization of Lakehouse Agents
The authors examine methods for optimizing the skill sets and environmental configurations of coding agents to improve their performance on data lakehouse infrastructure.
3
Towards a Unified Multidimensional Explainability Metric: Evaluating Trustworthiness in AI Models
This paper introduces a multidimensional framework designed to standardize the evaluation of explainability methods in machine learning models.
3
Gate-Zero Growth: A Geometric Framework for Function-Preserving Continual Learning
A new geometric framework called gate-zero growth is introduced to enable function-preserving continual learning by adding residual blocks without disrupting previously learned information.