AI.info

Latest Research — Page 29

Browse Latest Research on AI.info.

• VLASH: Real-Time VLAs via Future-State-Aware Asynchronous Inference
• ZGCM-1: A Fully Open and Extremely Efficient Foundation Model for Math and Agentic Search
• Is Forgetting Unified? Exploring Concept Unlearning in Unified Multimodal Models
• Surgical Activation Steering via Generative Causal Mediation
• Monocular Normal Estimation via Shading Sequence Estimation
• PSTF-AttControl: Per-Subject-Tuning-Free Personalized Image Generation with Controllable Face Attributes
• XInsight: Integrative Stage-Consistent Psychological Counseling Support Agents for Digital Well-Being
• You don't need long context
• AdaPRA: Process Reward Agents with Adaptive Rubrics for Data Science
• FoMo: Forking Moment in Generative Trajectory as a Perceptual Distance
• AgentODRL: A Large Language Model-based Multi-agent System for ODRL Generation
• Scalable GPU-Accelerated Euler Characteristic Curves: Optimization and Differentiable Learning for PyTorch
• It depends: Incorporating correlations for joint aleatoric and epistemic uncertainties of high-dimensional output spaces
• EpiAgent: An Agent-Centric System for Ancient Inscription Restoration
• Tool Zero: Training Tool-Augmented LLMs via Pure RL from Scratch
• AdmTree: Compressing Lengthy Context with Adaptive Semantic Trees
• SparkMe: Adaptive Semi-Structured Interviewing for Qualitative Insight Discovery
• MeanVoiceFlow: One-step Nonparallel Voice Conversion with Mean Flows
• Typhoon ASR Real-time: FastConformer-Transducer for Thai Automatic Speech Recognition
• Towards general embodied intelligence: integrating large language models, knowledge bases, and reasoning capabilities to build the next generation of AI agents
• Think-While-Generating: On-the-Fly Reasoning for Personalized Long-Form Generation
• MAD-Fact: A Multi-Agent Debate Framework for Long-Form Factuality Evaluation in LLMs
• BEAT: Visual Backdoor Attacks on VLM-based Embodied Agents via Contrastive Trigger Learning
• ReflectRL: Learning from Golden Negative Trajectories via Reflective-to-Direct Reasoning
• OSS-CRS: Liberating AIxCC Cyber Reasoning Systems for Real-World Open-Source Security
• Kernel Learning with Adversarial Features: Numerical Efficiency and Adaptive Regularization
• DINOv3-Guided Cross Fusion Framework for Semantic-aware CT generation from MRI and CBCT
• Breaking the Safety-Capability Tradeoff: Reinforcement Learning with Verifiable Rewards Maintains Safety Guardrails in LLMs
• BIRDTurk: Adaptation of the BIRD Text-to-SQL Dataset to Turkish
• CORE: Comprehensive Ontological Relation Evaluation for Large Language Models
• A Theoretical Study on Bridging Internal Probability and Self-Consistency for LLM Reasoning
• CompTrack: Information Bottleneck-Guided Low-Rank Dynamic Token Compression for Point Cloud Tracking
• Performance Evaluation of Transfer Learning Based Medical Image Classification Techniques for Disease Detection
• InsideOut: Integrated RGB-Radiative Gaussian Splatting for Comprehensive 3D Object Representation
• Seeing Is Believing? A Benchmark for Multimodal Large Language Models on Visual Illusions and Anomalies
• CLIFT: Conformal Self-Verification for Web Agent Training and Test-Time Scaling
• Generative Classifiers Avoid Shortcut Solutions
• G3Ego: Gaze-Guided Graphs for Egocentric Action Understanding
• EM-KD: Distilling Efficient Multimodal Large Language Model with Unbalanced Vision Tokens
• Learning Hierarchical Patch Splitting Policies for Faster Vision Transformers