AI.info

Latest Research — Page 127

Browse Latest Research on AI.info.

• Legibility is Not Interpretability: Comparing Judged and Actual Importance in Chain-Of-Thought Reasoning
• Optimizer choice matters for the emergence of Neural Collapse
• Reasoning Before Translation: Enhancing Legal Machine Translation with Structured Reasoning
• Reusing Latent Speech Representations for Query-Conditioned Topic Localization in Transcripts
• HVAdam: A Full-Dimension Adaptive Optimizer
• Sound Logical Explanations for Mean Aggregation Graph Neural Networks
• SurfSplat: Conquering Feedforward 2D Gaussian Splatting with Surface Continuity Priors
• Learning to Repair Lean Proofs from Compiler Feedback
• Uncertainty Quantification for Deep Regression using Contextualised Normalizing Flows
• FabDreamer: Exploring the Image-to-Physical Workflow Through AI-Assisted Layered Fabrication
• Unsupervised Adaptation of 3D CT Foundation Models for 3D CBCT Segmentation
• FedQS: Optimizing Gradient and Model Aggregation for Semi-Asynchronous Federated Learning
• Cycle-Sync: Robust Global Camera Pose Estimation through Enhanced Cycle-Consistent Synchronization
• Credit Cards, Confusion, Computation, and Consequences: What Can We Uncover About Language Model Reasoning?
• Cross-Spectral Dense Correspondence for Multimodal Spectral Medical Imaging
• WildCap: Facial Albedo Capture in the Wild via Hybrid Inverse Rendering
• A continental-scale dataset of ground beetles with high-resolution images and validated morphological trait measurements
• Sampling Control for Imbalanced Calibration in Semi-Supervised Learning
• Vero: Can AI Agents Build Formally Verified Software Repositories?
• UEmbed: Unified Sparse and Dense Multimodal Embeddings
• Reward Engineering for Spatial Epidemic Simulations: A Reinforcement Learning Platform for Individual Behavioral Learning
• RMIT-ADM+S at the MMU-RAG NeurIPS 2025 Competition
• You Really Didn't Get That? Benchmarking Social Pragmatic Inference for Indirect and Playful Chinese Online Comments
• AdaTooler-V: Adaptive Tool-Use for Images and Videos
• Alias-Free ViT: Fractional Shift Invariance via Linear Attention
• Diffusion-Driven Two-Stage Active Learning for Low-Budget Semantic Segmentation
• The Astonishing Ability of Large Language Models to Parse Jabberwockified Language
• ICONIC-444: A 3.1-Million-Image Dataset for OOD Detection Research
• Complexity-based code embeddings
• ScienceBuddy: Recursive-in-Recursive Self-Improvement for Interactive Scientific Agents
• Phys-Liquid: A Physics-Informed Dataset for Estimating 3D Geometry and Volume of Transparent Deformable Liquids
• AI Knowledge Assist: An Automated Approach for the Creation of Knowledge Bases for Conversational AI Agents
• BLISS: Bandit Layer Importance Sampling Strategy for Efficient Training of Graph Neural Networks
• WOLF: Werewolf-based Observations for LLM Deception and Falsehoods
• Learning Latent Energy-Based Models via Interacting Particle Langevin Dynamics
• Pragmatic Heterogeneous Collaborative Perception via Generative Communication Mechanism
• Picasso: Holistic Scene Reconstruction with Physics-Constrained Sampling
• Are LLMs Reliable Rankers? Rank Manipulation via Two-Stage Token Optimization
• Building Trust in Clinical LLMs: Bias Analysis and Dataset Transparency
• Edit Less, Achieve More: Dynamic Sparse Neuron Masking for Lifelong Knowledge Editing in LLMs