AI.info

Latest Research — Page 85

Browse Latest Research on AI.info.

• Learning to Approximate Uniform Facility Location via Graph Neural Networks
• AREAs-Lab: An Interactive Environment for AI-driven Requirement Elicitation for AI Systems
• ReFORM: Reflected Flows for On-support Offline RL via Noise Manipulation
• UniJEPA: Enhancing Robot Policy via Unified Continuous and Discrete Representation Learning
• BrowseComp-$V^3$: A Visual, Vertical, and Verifiable Benchmark for Multimodal Browsing Agents
• PAC-Bayesian Reinforcement Learning Trains Generalizable Policies
• Persistent Memory in Multi-Agent LLM Inference: What It Costs, What It Buys, and When You Can Tell
• REALM: An MLLM-Agent Framework for Open World 3D Reasoning Segmentation and Editing on Gaussian Splatting
• VisMem: Latent Vision Memory Unlocks Potential of Vision-Language Models
• SharpNet: Enhancing MLPs to Represent Functions with Controlled Non-differentiability
• Global Convergence in Deep Networks via the Second Law of Thermodynamics
• Multimodal Model Diffing for Feature Discovery and Control
• Data-Efficient Adaptation of LLMs via Attention Head Reweighting
• BEND 2.0: Benchmarking the Capabilities Needed to Design DNA
• Adaptive Dual Reasoner: Large Reasoning Models Can Think Efficiently by Hybrid Reasoning
• The Imperfect Learner: Incorporating Developmental Trajectories in Memory-based Student Simulation
• Neural Theorem Proving for Verification Conditions: A Real-World Benchmark
• MTSQL-R1: Towards Long-Horizon Multi-Turn Text-to-SQL via Agentic Training
• SPoRC-VIST: A Benchmark for Evaluating Generative Natural Narrative in Vision-Language Models
• Any Resolution Any Geometry: From Multi-View To Multi-Patch
• When Health Assistants Agree Too Readily: A Benchmark of Time-Critical Self-Diagnosis
• Reward Models Inherit Value Biases from Pretraining
• EfficientFSL: Enhancing Few-Shot Classification via Query-Only Tuning in Vision Transformers
• Fixed Aggregation Features Can Rival GNNs
• Optimistic Online-to-Batch Conversions for Accelerated Convergence and Universality
• Informed Initialization for Bayesian Optimization and Active Learning
• C$^3$TG: Conflict-aware, Composite, and Collaborative Controlled Text Generation
• On the Hardness of Computing Counterfactual and Semifactual Explanations in XAI
• Data-efficient U-Net for Segmentation of Carbide Microstructures in SEM Images of Steel Alloys
• Do All Individual Layers Help? An Empirical Study of Task-Interfering Layers in Vision-Language Models
• WorldGrow: Generating Infinite 3D World
• AOV-GS: Active Open-Vocabulary Semantic Mapping with 3D Gaussian Splatting
• The Numerical Linear Algebra of Large Language Models
• TAUE: Training-free Noise Transplant and Cultivation Diffusion Model
• When Data is the Algorithm: A Systematic Study and Curation of Preference Optimization Datasets
• MAServeBench: Benchmarking Multi-Agent LLM Serving Systems with Hierarchical Collaboration Workloads
• FOCUS: First-Order Control Using Optimal Transport Supervision for Imitation from Observations
• Towards Holistic Modeling for Video Frame Interpolation with Auto-regressive Diffusion Transformers
• OlmoEarth: Stable Latent Image Modeling for Multimodal Earth Observation
• A Diffusion Model for Regular Time Series Generation from Irregular Data with Completion and Masking