AI.info

Latest Research — Page 123

Browse Latest Research on AI.info.

• A Unifying View of Coverage in Linear Off-Policy Evaluation
• Visual Prompt-Agnostic Evolution
• Mixture of Horizons in Action Chunking
• ACTG-ARL: Differentially Private Conditional Text Generation with RL-Boosted Control
• Segment and Matte Anything in a Unified Model
• MASFIN: A Multi-Agent System for Decomposed Financial Reasoning and Forecasting
• Dynamical modeling of nonlinear latent factors in multiscale neural activity with real-time inference
• FairImagen: Post-Processing for Bias Mitigation in Text-to-Image Models
• HiMo-CLIP: Modeling Semantic Hierarchy and Monotonicity in Vision-Language Alignment
• RynnValue: Scaling Robotic Value Foundation Models with Temporal Distance
• Variational Grey-Box Dynamics Matching
• Per-Matrix Optimality Is Not Enough: Three-Level Optimization for Low-Rank LLM Compression
• Data Pyramid for Embodied Manipulation
• Measuring the Tokenization Premium: A Cost Audit for Underserved Language Communities
• Real-World Reinforcement Learning of Active Perception Behaviors
• Preference-based Reinforcement Learning beyond Pairwise Comparisons: Benefits of Multiple Options
• Instruction Duplication as an Inference-Time Control Primitive
• iFlip: Iterative Feedback-driven Counterfactual Example Refinement
• Human-AI Co-design for Clinical Prediction Models
• Recurrent Sinusoidal INRs for Efficient High-Fidelity Representation
• OmniMotion-X: Versatile Multimodal Whole-Body Motion Generation
• White Box Evidence Packages for Policy Audit Reports
• Adapting Self-Supervised Representations as a Latent Space for Efficient Generation
• Refining Context-Entangled Content Segmentation via Curriculum Selection and Anti-Curriculum Promotion
• Maximizing Local Entropy Where It Matters: Prefix-Aware Localized LLM Unlearning
• Atomic Diffusion Models for Small Molecule Structure Elucidation from NMR Spectra
• Rethinking Benign Relearning: Syntax as the Hidden Driver of Unlearning Failures
• LLM-Guided Neural Architecture Search for Robust Co-Design of Physical Neural Networks
• On-Policy Self-Distillation in Diffusion Models
• Why LVLMs Are More Prone to Hallucinations in Longer Responses: The Role of Context
• The Impact and Outlook of 3D Gaussian Splatting
• KV Admission: Learning What to Write for Efficient Long-Context LLM Inference
• WaterMod: Modular Token-Rank Partitioning for Probability-Balanced LLM Watermarking
• Delta-XAI: A Unified Framework for Explaining Prediction Changes in Online Time Series Monitoring
• Enhancing Demand-Oriented Regionalization with Agentic AI and Local Heterogeneous Data for Adaptation Planning
• Improved Object-Centric Diffusion Learning with Registers and Contrastive Alignment
• CogFlow: Bridging Perception and Reasoning through Knowledge Internalization for Visual Mathematical Problem Solving
• Latent Harmony: Synergistic Unified UHD Image Restoration via Latent Space Regularization and Controllable Refinement
• Studying People to Study AI: Expert Perspectives on the Epistemic Fit and Barriers of Human Research in AI Safety & Ethics
• RoMa v2: Harder Better Faster Denser Feature Matching