AI.info

Latest Research — Page 82

Browse Latest Research on AI.info.

• FreeFlow: A Bias-free Hierarchical Transformer for Optical Flow Estimation
• Kinaema: a recurrent sequence model for memory and pose in motion
• View Space: Learning Representation across Arbitrary Graphs
• LR-DWM: Efficient Watermarking for Diffusion Language Models
• How Useful are LLMs for Grammar Engineering? Cantonese ParGram Resources and Controlled Experimental Evaluation with English Baselines
• DP-FedSOFIM: Differentially Private Federated Stochastic Optimization using Regularized Fisher Information Matrix
• SafeSearch: Do Not Trade Safety for Utility in LLM Search Agents
• UTDesign: A Unified Framework for Stylized Text Editing and Generation in Graphic Design Images
• R2-Router: A New Paradigm for LLM Routing with Reasoning
• FMA-Net++: Motion- and Exposure-Aware Joint Video Super-Resolution and Deblurring
• Perception-to-Pursuit: Track-Centric Temporal Reasoning for Open-World Drone Detection and Autonomous Chasing
• ProtocolBench: Which LLM MultiAgent Protocol to Choose?
• Can AI Agents Simulate A/B Test Outcomes? A Validation Framework for Agentic Experimentation
• Latent Chain-of-Thought for Visual Reasoning
• Scaling Near-Optimal SFT-RL Annotation Budget Allocation from Small to Large LLMs
• SEAL: Semantic-Aware Hierarchical Learning for Generalized Category Discovery
• Image Valuation in NeRF-based 3D reconstruction
• Context-aware Learned Mesh-based Simulation via Trajectory-Level Meta-Learning
• GTA: Graph Theory Agent and Benchmark for Algorithmic Graph Reasoning with LLMs
• Look Less, Think Faster: Joint Token-Compute Adaptation for Multimodal LLMs
• Learning to Clean: Reinforcement Learning for Noisy Label Correction
• How Different Tokenization Algorithms Impact LLMs and Transformer Models for Binary Code Analysis
• SURFACEBENCH: A Geometry-Aware Benchmark for Symbolic Surface Discovery
• ICL-Router: In-Context Learned Model Representations for LLM Routing
• Provable Benefit of Curriculum in Transformer Tree-Reasoning Post-Training
• The Chameleon Nature of LLMs: Quantifying Multi-Turn Stance Instability in Search-Enabled Language Models
• Reconstruction: A Blind Benchmark for Recovering Research Ideas from Pre-Publication Bibliographies
• Seeing through Light and Darkness: Sensor-Physics Grounded Deblurring HDR NeRF from Single-Exposure Images and Events
• What Does the Credential Still Certify? Cognitive Stewardship for AI-Mediated Education
• RotaryQuant: Fitting 120B MoE Models on Consumer Hardware via Fused Compressed-Space Attention
• Causal State-Space Model for Causal Inference: Estimating Longitudinal Individual Treatment Effects
• Semi-Peaucellier Linkage and Differential Mechanism for Linear Pinching and Self-Adaptive Grasping
• Objects as Audio-Visual Modal Sound Fields
• Flow Matching for Tabular Data Synthesis
• Echoes Over Time: Unlocking Length Generalization in Video-to-Audio Generation Models
• ADI-20: Arabic Dialect Identification dataset and models
• Loquetier: A Virtualized Multi-LoRA Framework for Unified LLM Fine-tuning and Serving
• Exposing Vulnerabilities in Explanation for Time Series Classifiers via Dual-Target Attacks
• Enhancing DPSGD via Per-Sample Momentum and Low-Pass Filtering
• Value-Action Alignment in Large Language Models under Privacy-Prosocial Conflict