AI.info

Latest Research — Page 7

Browse Latest Research on AI.info.

• Coarse-Grained Boltzmann Generators
• REMISVFU: Vertical Federated Unlearning via Representation Misdirection for Intermediate Output Feature
• Step-Audio-EditX Technical Report
• SEMDICE: Off-policy State Entropy Maximization via Stationary Distribution Correction Estimation
• RACAS: Controlling Diverse Robots With a Single Agentic System
• WED-Net: A Weather-Effect Disentanglement Network with Causal Augmentation for Urban Flow Prediction
• AegisBench: Auditing Aerial Search-and-Rescue Person Detection Under the Conditions It Is Deployed In
• Token-Efficient Change Detection in LLM APIs
• Detecting Legend Items on Historical Maps Using GPT-4o with In-Context Learning
• OpenFrontiers: A Collection of Open Research Problems and a Benchmark for Scientific Rediscovery
• Integrating Multi-view Multi-light Surface Reconstruction into Cultural Heritage Workflows
• Zero-RAG: Towards Retrieval-Augmented Generation with Zero Redundant Knowledge
• GNN-CB: A Graph Neural Network Competition Benchmark for Human and LLM Evaluation
• CoDeQ: End-to-End Joint Model Compression with Dead-Zone Quantizer for High-Sparsity and Low-Precision Networks
• AdaTutoRank: Learning to Rerank Document Sets via Adaptive Tutoring Optimization for RAG and Deep Research
• Spectral Degeneration of Self-Attention under Isotropic Score Geometry
• Can Local Learning Match Self-Supervised Backpropagation?
• SPDMark: Selective Parameter Displacement for Robust Video Watermarking
• Bootstrap Dynamic-Aware 3D Visual Representation for Scalable Robot Learning
• A Mind Cannot Be Smeared Across Time
• Trial by Trace: Benchmarking AI Agent Harnesses for Reliability and Failure Attribution
• The Fast Lane Hypothesis: Von Economo Neurons Implement a Biological Speed-Accuracy Tradeoff
• Beyond RAG for Agent Memory: Retrieval by Decoupling and Aggregation
• From Black-Box Tuning to Guided Optimization via Hyperparameters Interaction Analysis
• QSTN: A Modular Framework for Robust Questionnaire Inference with Large Language Models
• Fairness-Regularized Online Optimization with Switching Costs
• Spoken DialogSum: An Emotion-Rich Conversational Dataset for Spoken Dialogue Summarization
• Language Models Learn to Improve Language Models
• The fixed-point bundle method over product-of-simplex domains arising from game equilibria
• Equivariant Sparse Autoencoders: Mechanistic Interpretability of Neural Networks on Symmetric Data
• T-MLA: A targeted multiscale log-exponential attack framework for neural image compression
• Continual GUI Agents
• LatentMem: Customizing Latent Memory for Multi-Agent Systems
• Aetheria: A multimodal interpretable content safety framework based on multi-agent debate and collaboration
• Clinician use of language models diverges from how the models are evaluated
• TROJail: Trajectory-Level Optimization for Multi-Turn Large Language Model Jailbreaks with Process Rewards
• AREA3D: Active Reconstruction Agent with Unified Feed-Forward 3D Perception and Vision-Language Guidance
• MemLife: Curating and Reasoning over Long-Term Egocentric Video Memories
• Saudi Sign Language Translation Using T5
• Uncertainty-Calibrated Prediction of Randomly-Timed Biomarker Trajectories with Conformal Bands