AI.info

Latest Research — Page 101

Browse Latest Research on AI.info.

• Test-time Ego-Exo-centric Adaptation for Action Anticipation via Multi-Label Prototype Growing and Dual-Clue Consistency
• AREAL-DTA: Dynamic Tree Attention for Efficient Reinforcement Learning of Large Language Models
• Large Language Model-Driven Context-Aware Eco-Feedback Generation and Evaluation
• PhysInOne: Visual Physics Learning and Reasoning in One Suite
• E-RayZer: Self-supervised 3D Reconstruction as Spatial Visual Pre-training
• Timely Clinical Diagnosis through Active Test Selection
• With Great Capabilities Come Great Responsibilities: Introducing the Agentic Risk & Capability Framework for Governing Agentic AI Systems
• TokenPowerBench: Benchmarking the Power Consumption of LLM Inference
• Impact of Layer Norm on Memorization and Generalization in Transformers
• HiFi-RAG: Hierarchical Content Filtering and Two-Pass Generation for Open-Domain RAG
• Enhancing Node-Level Graph Domain Adaptation by Alleviating Local Dependency
• CHiQPM: Calibrated Hierarchical Interpretable Image Classification
• ELVIS: Enhance Low-Light for Video Instance Segmentation in the Dark
• Two-Layer Linear Auto-Regressive Models Estimate Latent States
• Loom: Weaving Diagnostic Strands into Free-Text Consensus via Embedding-Space Reweighting
• Data Descriptions from Large Language Models with Influence Estimation
• Certified but Fooled! Breaking Certified Defences with Ghost Certificates
• Human-Inspired Social Engagement Analysis via Interpretable Mutual Visual Attention
• Towards Adaptive Environment Generation for Training Embodied Agents
• A Unified Per-Token Gating Family for On-Policy Distillation: FKL/RKL Mixing with Multi-Channel and Bias Coefficients
• Low-N Protein Activity Optimization with FolDE
• Left-Right Symmetry Breaking in CLIP-style Vision-Language Models Trained on Synthetic Spatial-Relation Data
• FormationEval, an open multiple-choice benchmark for petroleum geoscience
• GBU-Palm: A Multimodal Video Dataset and Benchmark for Palm Presentation Attack Detection
• MetricAnything: Scaling Metric Depth Pretraining with Noisy Heterogeneous Sources
• EPSVec: Efficient and Private Synthetic Data Generation via Dataset Vectors
• DiverseDiT: Towards Diverse Representation Learning in Diffusion Transformers
• ATEX-CF: Attack-Informed Counterfactual Explanations for Graph Neural Networks
• Faithfulness Is Not Free: Auditing Offline KV-Cache Quantization in Retrieval-Augmented Generation
• Extrapolation of Periodic Functions Using Binary Encoding of Continuous Numerical Values
• Unpacking Human Preference for LLMs: Demographically Aware Evaluation with the HUMAINE Framework
• When Tokenization is Secretly Output Supervision
• Chimera: State Space Models Beyond Sequences
• JanusCoder: Towards a Foundational Visual-Programmatic Interface for Code Intelligence
• Soft Task-Aware Routing of Experts for Equivariant Representation Learning
• Real-Time Procedural Learning From Experience for AI Agents
• ViRC: Enhancing Visual Interleaved Mathematical CoT with Reason Chunking
• ConSurv: Multimodal Continual Learning for Survival Analysis
• JADE: Expert-Grounded Dynamic Evaluation for Open-Ended Professional Tasks
• Improving Information Extraction with Learned Queries