AI.info

Latest Research — Page 16

Browse Latest Research on AI.info.

• Prompt Design at Scale: How Format, Instruction Count, and Context Length Shape Instruction Adherence and Hallucination in Large Language Models
• Flow of Spans: Generalizing Language Models to Dynamic Span-Vocabulary via GFlowNets
• InternVL-U: Democratizing Unified Multimodal Models for Understanding, Reasoning, Generation and Editing
• FreeInpaint: Tuning-free Prompt Alignment and Visual Rationality Enhancement in Image Inpainting
• Point-Supervised Facial Expression Spotting with Gaussian-Based Instance-Adaptive Intensity Modeling
• Sesame Plant Segmentation Dataset: A YOLO Formatted Annotated Dataset
• An Improved Model-Free Decision-Estimation Coefficient with Applications in Adversarial MDPs
• MindPower: Enabling Theory-of-Mind Reasoning in VLM-based Embodied Agents
• Class Prototypes based Contrastive Learning for Classifying Multi-Label and Fine-Grained Educational Videos
• A Fast and Flat Federated Learning Method via Weighted Momentum and Sharpness-Aware Minimization
• PerfGuard: A Performance-Aware Agent for Visual Content Generation
• Future of AI Models: A Computational perspective on Model collapse
• Multi-Agent Decision-Focused Learning via Value-Aware Sequential Communication
• Mind the Ambiguity: Aleatoric Uncertainty Quantification in LLMs for Safe Medical Question Answering
• RoRA: Role-Oriented Regional Allocation for Visual Token Pruning in MLLMs
• Lightweight Neural Networks for Affordance Segmentation: Enhancement of the Decoder Module
• Attend Before Attention: Efficient and Scalable Video Understanding via Autoregressive Gazing
• FoundationSLAM: Unleashing the Power of Depth Foundation Models for End-to-End Dense Visual SLAM
• Trellis: Learning to Compress Key-Value Memory in Attention Models
• Self-Supervised Learning via Flow-Guided Neural Operator on Time-Series Data
• Long-Horizon Model-Based Offline Reinforcement Learning Without Explicit Conservatism
• KeepLoRA: Continual Learning with Residual Gradient Adaptation
• Grounding Computer Use Agents on Human Demonstrations
• Domain-Specific Hallucination Detection in Large Language Models
• ActiShade: Activating Overshadowed Knowledge to Guide Multi-Hop Reasoning in Large Language Models
• Logit Refiner: Improving Visual Autoregressive Models via Intra-Scale Dependency Modeling
• Concept Guidance: Precise, Training-Free Latent Control for Text-to-Image Generation
• Who Does Your Algorithm Fail? Investigating Age and Ethnic Bias in the MAMA-MIA Dataset
• Systems for Scaling Accessibility Efforts in Large Computing Courses
• Caption-Driven Explainability: Probing CNNs for Bias via CLIP
• What Matters, When? Diagnosing and Improving Conditional Visual Grounding in Visuomotor Imitation Policies
• Reliability-Aware Adaptive Self-Consistency for Efficient Sampling in LLM Reasoning
• SOM Directions are Better than One: Multi-Directional Refusal Suppression in Language Models
• Dual-Robust Cross-Domain Offline Reinforcement Learning Against Dynamics Shifts
• AUHead: Realistic Emotional Talking Head Generation via Action Units Control
• Context and Symmetry in Auditing: A Case Study of Skeleton Inference in Motion Capture
• Text-to-3D by Stitching a Multi-view Reconstruction Network to a Video Generator
• SocialFusion: Addressing Social Degradation in Pre-trained Vision-Language Models
• Customizing Open Source LLMs for Quantitative Medication Attribute Extraction across Heterogeneous EHR Systems
• Speculative Sampling with Reinforcement Learning