AI.info

Latest Research — Page 35

Browse Latest Research on AI.info.

• Focused Omni-Modal GUI Agents
• UA-Code-Bench: A Competitive Programming Benchmark for Evaluating LLM Code Generation in Ukrainian
• HiFusion: Hierarchical Intra-Spot Alignment and Regional Context Fusion for Spatial Gene Expression Prediction from Histopathology
• Not What a Child Expressed: Auditing the Sign-to-Text Safety Interface in Child-Facing AI
• Humans Introduce, Models Elaborate: Asymmetric Narrative Agency in Human-LLM Co-Writing
• End2Reg: Learning Task-Specific Segmentation for Markerless Registration in Spine Surgery
• From Single to Societal: Analyzing Persona-Induced Bias in Multi-Agent Interactions
• Scaling Laws Meet Model Architecture: Toward Inference-Efficient LLMs
• WebClipper: Efficient Evolution of Web Agents with Graph-based Trajectory Pruning
• Assessing Quality of Experience in Natural Language Generation of German Text
• Transformer-based Parameter Fitting of Models derived from Bloch-McConnell Equations for CEST MRI Analysis
• GroundLoc: Efficient Large-Scale Outdoor LiDAR-Only Localization
• Isaac Sim-to-Real: Reinforcement Learning based Locomotion for Quadrupeds
• Towards Bridging the Gap between Large-Scale Pretraining and Efficient Finetuning for Humanoid Control
• Bypassing the Rationale: Causal Auditing of Implicit Reasoning in Language Models
• ACE-G: Improving Generalization of Scene Coordinate Regression Through Query Pre-Training
• ACE-Align: Attribute Causal Effect Alignment for Cultural Values under Varying Persona Granularities
• Toward Human-Centered Readability Evaluation
• Dataless Weight Disentanglement in Task Arithmetic via Kronecker-Factored Approximate Curvature
• Energy-based Autoregressive Generation for Neural Population Dynamics
• Segment-Level Attribution for Selective Learning of Long Reasoning Traces
• CRoPS: A Training-Free Hallucination Mitigation Framework for Vision-Language Models
• TSRBench: A Comprehensive Multi-task Multi-modal Time Series Reasoning Benchmark for Generalist Models
• MosaicDiff: Training-free Structural Pruning for Diffusion Model Acceleration Reflecting Pretraining Dynamics
• VibeGame: Prompt-to-Game Development with AI-Native Engine and Self-Evolving Adversarial Agent Team
• FinMMDocR: Benchmarking Financial Multimodal Reasoning with Scenario Awareness, Document Understanding, and Multi-Step Computation
• VULCA-Bench: A Multicultural Vision-Language Benchmark for Evaluating Cultural Understanding
• Out-of-Distribution Generalization with a SPARC: Racing 100 Unseen Vehicles with a Single Policy
• Model Inversion with Layer-Specific Modeling and Alignment for Data-Free Continual Learning
• Self-supervised Multiplex Consensus Mamba for General Image Fusion
• Multimodal Taxonomic Conditioning for Generative Plankton Imagery
• Curvature-Aware Zeroth-Order Optimization for Memory-Efficient Test-Time Adaptation
• PANINI-DR: Controlled Deep Research through Generative Semantic Workspaces
• Deep networks learn to parse uniform-depth context-free languages from local statistics
• STREAM-VAE: Dual-Path Routing for Slow and Fast Dynamics in Vehicle Telemetry Anomaly Detection
• The Symbol of the Surrogate: Measuring Numerical Provenance in Neural PDE Solvers
• Measuring Pragmatic Influence in Large Language Model Instructions
• SAMTok: Representing Any Mask with Two Words
• When LLMs get significantly worse: A statistical approach to detect model degradations
• PRET is a few-shot system for pan-cancer recognition without example training