AI.info
Latest Research — Page 83
Browse Latest Research on AI.info.
- R-Align: Enhancing Generative Reward Models through Rationale-Centric Meta-Judging
- Gated Tree Cross-Attention for Checkpoint-Compatible Syntax Injection in Decoder-Only LLMs
- AI-Assisted Authoring for Transparent, Data-Driven Documents
- Continuum Attention
- TernScope: Model-Hardware Co-Design for Efficient Ternary Endoscopic Segmentation
- Saying the Unsaid: Revealing the Hidden Language of Multimodal Systems Through Telephone Games
- Beyond Monotonicity: Revisiting Factorization Principles in Multi-Agent Q-Learning
- PTOPOFL: Privacy-Preserving Personalised Federated Learning via Persistent Homology
- Text-Conditioned Background Generation for Editable Multi-Layer Documents
- Ontology-supported AI Model and Dataset Management
- Be My Eyes: Extending Large Language Models to New Modalities Through Multi-Agent Collaboration
- Energy-Aware Path Following: Comparative Analysis of Reinforcement Learning and NMPC for Electric Vehicles
- Faithful Summarization of Consumer Health Queries: A Cross-Lingual Framework with LLMs
- DiffMM: Efficient Method for Accurate Noisy and Sparse Trajectory Map Matching via One Step Diffusion
- Think Twice: Branch-and-Rethink Reasoning Reward Model
- Predicting the Emergence of Induction Heads in Language Model Pretraining
- Dual Mamba for Node-Specific Representation Learning: Tackling Over-Smoothing with Selective State Space Modeling
- FedGRPO: Privately Optimizing Foundation Models with Group-Relative Rewards from Domain Client
- Modeling Retinal Ganglion Cells with Neural Differential Equations
- WorldMemBench: A Comprehensive Benchmark for Memory of Video World Models
- IMBWatch -- a Spatio-Temporal Graph Neural Network approach to detect Illicit Massage Business
- ChronoReplay: One Answer Interface Is Not Enough for Streaming Spatial Intelligence
- MoETTA: Test-Time Adaptation Under Mixed Distribution Shifts with MoE-LayerNorm
- WalkGPT: Grounded Vision-Language Conversation with Depth-Aware Segmentation for Pedestrian Navigation
- Primal-Dual Flow Matching for Sample-Wise Constrained Generation
- Length-MAX Tokenizer for Language Models
- Vision-Language Memory for Spatial Reasoning
- UltraLED: Learning to See Everything in Ultra-High Dynamic Range Scenes
- Behavioral Embeddings of Programs: A Quasi-Dynamic Approach for Optimization Prediction
- Beyond the Lower Bound: Bridging Regret Minimization and Best Arm Identification in Lexicographic Bandits
- Partially Equivariant Reinforcement Learning in Symmetry-Breaking Environments
- LLaDA-UI: Bringing Block-wise Diffusion to Vision-Language GUI Agents
- The Zero-Step Thinking: An Empirical Study of Mode Selection as Harder Early Exit in Reasoning Models
- Panel-by-Panel Souls: A Performative Workflow for Expressive Faces in AI-Assisted Manga Creation
- Enhancing Geo-localization for Crowdsourced Flood Imagery via LLM-Guided Attention
- Pose-Free Feed-Forward 3D Inpainting via Learnable Mask Attention and Support Token Refinement
- Sparse Planning in Visual World Models via Cost Gradients
- KBQA-R1: Reinforcing Large Language Models for Knowledge Base Question Answering
- Studying Without a Syllabus: Task-Agnostic Environment Preprocessing
- Jev in the Wild: A Data-Driven Analysis of the Jev Model's Functionality, Applications and Ecosystem