AI.info

Latest Research — Page 13

Browse Latest Research on AI.info.

• CARVQ: Corrective Adaptor with Group Residual Vector Quantization for LLM Embedding Compression
• LaGen: Towards Autoregressive LiDAR Scene Generation
• Conditional Trajectory Peaks: Single-Pass Multimodal Policies over Action Chunks
• CareMedEval dataset: Evaluating Critical Appraisal and Reasoning in the Biomedical Field
• Regret Lower Bounds for Decentralized Multi-Agent Stochastic Shortest Path Problems
• When CNNs Outperform Transformers and Mambas: Revisiting Deep Architectures for Dental Caries Segmentation
• PointRAFT: 3D deep learning for high-throughput prediction of potato tuber weight from partial point clouds
• TTCS: Test-Time Curriculum Synthesis for Self-Evolving
• UI-Ins: Enhancing GUI Grounding with Multi-Perspective Instruction-as-Reasoning
• Beta Distribution Learning for Reliable Roadway Crash Risk Assessment
• Vision Transformers are Circulant Attention Learners
• GryphOne: Symbol-Aware Masked Diffusion for Structural Refinement in Offline Handwritten Mathematical Expression Recognition
• Stable diffusion models reveal a persisting human and AI gap in visual creativity
• OwlerLite: Scope- and Freshness-Aware Web Retrieval for LLM Assistants
• Learning Engagement Assistant (LEA): Cross-Course Scalability and Classroom Evaluation of an Agentic AI Tutoring System
• RAFT: A Stateful Retrieval-Augmented Framework for Troubleshooting Agents
• The Low-Rank Structure of VLA Reinforcement Learning
• Rectified LpJEPA: Joint-Embedding Predictive Architectures with Sparse and Maximum-Entropy Representations
• Safety for Whom? Boundary-Aware Self-Distillation for Controlled LLM Safety Refusal
• MIMO: A medical vision language model with visual referring multimodal input and pixel grounding multimodal output
• MEND: RL For Flow Models via Proximal Velocity Matching
• Less Is More, but Where? Dynamic Token Compression via LLM-Guided Keyframe Prior
• GRAVER: Generative Graph Vocabularies for Robust Graph Foundation Models Fine-tuning
• BOFA: Bridge-Layer Orthogonal Low-Rank Fusion for CLIP-Based Class-Incremental Learning
• PeakFlow: Peak-Guided Coarse-to-Refined Modeling for EEG-Based Dynamic Affective Trajectory Prediction
• Attacks by Content: Automated Fact-checking is an AI Security Issue
• Manipulation-Proof Oblivious Audits against Deceptive Model Providers
• Generative adversarial networks with Monte Carlo dropout uncertainty quantification for structural health monitoring: Fibre Bragg grating data augmentation comparison in Hawk T1A aircraft
• Quantized FCA: Efficient Zero-Shot Texture Anomaly Detection
• Structure-Conditional Minimum Bayes Risk Decoding
• Building Reasonable Inference for Vision-Language Models in Blind Image Quality Assessment
• LensErase: A Foundation Model for Erase Anything
• DeflectBench: A Benchmark for Evaluating Rhetorical Fallacy Generation in LLMs
• MetaWorld: Skill Transfer and Composition in a Hierarchical World Model for Grounding High-Level Instructions
• Singular Vectors of Attention Heads Align with Features
• Why Attention Patterns Exist: A Unifying Temporal Perspective Analysis
• Total Variation Distance Estimation in Autoregressive Models
• Can Protective Watermarking Safeguard the Copyright of 3D Gaussian Splatting?
• RankLLM: Weighted Ranking of LLMs by Quantifying Question Difficulty
• Hail to the Thief: Exploring Attacks and Defenses in Decentralised GRPO