AI.info

Latest Research — Page 79

Browse Latest Research on AI.info.

• SaMA: Morpho Adaptation via Asymmetric Expansion of Kronecker Product
• Procedural Graphs: Self-Evolving Execution Structures for LLM Agents
• AutoDesign: Meta-Harness Optimization for Long-Horizon Agentic Design
• TPPO: Tracking-Value PPO For Language Model Reasoning
• Training Language Models to Explain Their Own Computations
• Skill2Env: Capability-Oriented Environment Synthesis from Skills for General Agents
• NextFlow: Unified Sequential Modeling Activates Multimodal Understanding and Generation
• BI-Agent and BI-Bench: Towards Automating End-to-End Business Intelligence
• FRAMER: Frequency-Aligned Self-Distillation with Adaptive Modulation Leveraging Diffusion Priors for Real-World Image Super-Resolution
• VERA-V: Variational Inference Framework for Jailbreaking Vision-Language Models
• Flipping the Dialogue: Training and Evaluating User Language Models
• Telling Speculative Stories to Help Humans Imagine the Harms of Healthcare AI
• Graph Diffusion Transformers are In-Context Molecular Designers
• Exploration Through Introspection: A Self-Aware Reward Model
• Translation with Thought: Difficulty-Adaptive Reasoning via Reinforcement Learning for Multi-Domain Machine Translation
• Cheap2Rich: A Multi-Fidelity Framework for Data Assimilation and System Identification of Multiscale Physics -- Rotating Detonation Engines
• UniAVC: Unified Audio-Video Creation via Progressive Training and Asymmetric Scheduling
• NeAR: Coupled Neural Asset-Renderer Stack
• DroneWAM: Efficient World Action Model for Drone Visual Navigation
• AdaGuard: An Adaptive Guard Model with User-defined Policies
• Connecting the Dots: A Machine Learning Ready Dataset for Ionospheric Forecasting Models
• "Are We Done Yet?": A Vision-Based Judge for Autonomous Task Completion of Computer Use Agents
• Learning Peer Influence Probabilities with Linear Contextual Bandits
• Empowering Decision Trees via Shape Function Branching
• Towards Low-Resource Alignment to Diverse Perspectives with Sparse Feedback
• A Multi-View Pipeline and Benchmark Dataset for 3D Hand Pose Estimation in Surgery
• VLNVerse: A Benchmark for Vision-Language Navigation with Versatile, Embodied, Realistic Simulation and Evaluation
• JevAgentHarness: Benchmarking, Deploying and Training Decision Models That Share Control with LLMs
• Efficient Reasoning Training Does Not Always Harm CoT Faithfulness and Monitorability
• URDF-Anything: Constructing Articulated Objects with 3D Multimodal Language Model
• When Helpers Become Hazards: A Benchmark for Analyzing Multimodal LLM-Powered Safety in Daily Life
• What Uncertainties Do We Need for Dynamical Systems?
• EgoExo-Con: Exploring View-Invariant Video Temporal Understanding
• Enhancing Diffusion Model Guidance through Calibration and Regularization
• VLA-Arena: An Open-Source Framework for Benchmarking Vision-Language-Action Models
• Scaling to Tens of Thousands of Test-Time Iterations with Loop-Native Attention Residuals
• DexAvatar: 3D Sign Language Reconstruction with Hand and Body Pose Priors
• MassAlloc Attention: Let Attention Allocate Its Own Compute
• Words Caught in the Heart: Simulating Person-Specific Expression for Self-Opaque Clients in Psychotherapy
• TimeRSI: Towards Self-Improved Agentic Deployment for Time-Series Foundation Models