AI.info

Latest Research — Page 100

Browse Latest Research on AI.info.

• SuS: Strategy-aware Surprise for Intrinsic Exploration
• VL-SAE: Interpreting and Enhancing Vision-Language Alignment with a Unified Concept Set
• The Denario project: Deep knowledge AI agents for scientific discovery
• CASTLE: A Comprehensive Benchmark for Evaluating Student-Tailored Personalized Safety in Large Language Models
• Ignorance or Incompetence? Constructing Knowledge-Gated, Verifiable Tasks for LLM Agents
• NAF: Zero-Shot Feature Upsampling via Neighborhood Attention Filtering
• BADiff: Bandwidth Adaptive Diffusion Model
• MeanFuser: Fast One-Step Multi-Modal Trajectory Generation and Adaptive Reconstruction via MeanFlow for End-to-End Autonomous Driving
• On Stealing Graph Neural Network Models
• One Stone with Two Birds: A Null-Text-Null Frequency-Aware Diffusion Models for Text-Guided Image Inpainting
• Last Translation Benchmark
• ALLMTS: Automated LLM-based Medical Triage System
• Multi-Task Pre-Finetuning of Lightweight Transformer Encoders for Text Classification and NER
• Animus3D: Text-driven 3D Animation via Motion Score Distillation
• CLIP-FTI: Fine-Grained Face Template Inversion via CLIP-Driven Attribute Conditioning
• From Pixels to Structure: Lightweight Vision-Language Models for Document OCR and Structured JSON Extraction
• VepAgent: Bridging Causal-Transition via Tool-Augmented Reinforcement Learning for Video Event Prediction
• Where Is a Prediction Model Miscalibrated, and for Whom? Local Audits in Outcome-Learned Pockets
• Spatiotemporal-Untrammelled Mixture of Experts for Multi-Person Motion Prediction
• Learning to Collaborate: An Orchestrated-Decentralized Framework for Peer-to-Peer LLM Federation
• ReGround: Grounding Reviewer Comments in Multimodal Evidence
• Decoy Direction Optimization: A Post-Hoc Defense Against LLM Abliteration
• CoEM: Empowering Long-Context Reasoning with Commit-on-Evidence Memory
• ClusterMine: Robust Label-Free Visual Out-Of-Distribution Detection via Concept Mining from Text Corpora
• Exploring Single Domain Generalization of LiDAR-based Semantic Segmentation under Imperfect Labels
• Diagnosing the Reliability of LLM-as-a-Judge via Item Response Theory
• CaMiT: A Time-Aware Car Model Dataset for Classification and Generation
• MobileOcc: A Human-Aware Semantic Occupancy Dataset for Mobile Robots
• FlashMask-3: Efficient and Expressive Mask-Aware Distributed Attention
• Visual Sync: Multi-Camera Synchronization via Cross-View Object Motion
• AskStream: When and What to Ask Proactively under Uncertainty in Streaming Video Understanding
• Towards Unsupervised Open-Set Graph Domain Adaptation via Dual Reprogramming
• Value-Rich, Differently Organized: Measuring Value Expression and Organization in AI-Generated Debates
• KORE: Enhancing Knowledge Injection for Large Multimodal Models via Knowledge-Oriented Controls
• Uni-Hand: Universal Hand Motion Forecasting in Egocentric Views
• Towards Transparent Reasoning: What Drives Faithfulness in Large Language Models?
• CalibForge: Adversarial Solver Calibration for Scaling Learnable Terminal Tasks
• Tri-PvP: Exposing Modality Bias in Omni-Modal Large Language Models through Perceptual-Propositional Evidence Conflicts
• The Robustness of Differentiable Causal Discovery in Misspecified Scenarios
• Longitudinal Vestibular Schwannoma Dataset with Consensus-based Human-in-the-loop Annotations