AI.info

Latest Research — Page 94

Browse Latest Research on AI.info.

• Language Model Behavioral Phases are Consistent Across Architecture, Training Data, and Scale
• SynQP: A Framework and Metrics for Evaluating the Quality and Privacy Risk of Synthetic Data
• Conformal Prediction in The Loop: A Feedback-Based Uncertainty Model for Trajectory Optimization
• Offline Behavioral Data Selection
• Auto-MD-bench: Evaluating Autoresearch Agents for Real-World Medical Tasks
• Aligning Tree-Search Policies with Fixed Token Budgets in Test-Time Scaling of LLMs
• OpenMobile-2: Building Versatile Mobile Agents with Scalable Environments and App-Native Tools
• CodeGraph: Open-Taxonomy Knowledge Graph for Source Code with Wikidata Grounding
• The State and Fate of Multilingual, Contextual Evaluation in the NLP World
• Q3R: Quadratic Reweighted Rank Regularizer for Effective Low-Rank Training
• AREX-2: Advancing Self-Improving Agents through Long-Horizon Reflective Tasks
• MobileManiBench: Simplifying Model Verification for Mobile Manipulation
• REMORY: Learning Residual Memory for Context Compaction
• Position: It is Time to Virtualize Foundation Models with a Self-evolving Operating System Layer
• Can AI Agents Make Open-Ended Scientific Discovery? Evidence from Station
• Calibrating Teacher--Student Discrepancy for On-Policy Distillation
• RoadSceneVQA: Benchmarking Visual Question Answering in Roadside Perception Systems for Intelligent Transportation System
• All You Need is One: Capsule Prompt Tuning with a Single Vector
• Schrödinger's Code Repository: Have LLMs Learned SWE-bench or Memorized It?
• What matters for Representation Alignment: Global Information or Spatial Structure?
• PAN: A World Model for General, Actionable, and Long-Horizon World Simulation
• Enhancing Control Policy Smoothness by Aligning Actions with Predictions from Preceding States
• Improved Balanced Classification with Theoretically Grounded Loss Functions
• Enhancing Multimodal Misinformation Detection by Replaying the Whole Story from Image Modality Perspective
• Predictive Credit: Measuring What Scientific Explanations Add to Experimental Forecasts
• AutoMLGen: Navigating Fine-Grained Optimization for Coding Agents
• GSPN-2: Efficient Parallel Sequence Modeling
• Promoting Sustainable Web Agents: Benchmarking and Estimating Energy Consumption through Empirical and Theoretical Analysis
• SS4D: Native 4D Generative Model via Structured Spacetime Latents
• Vision-Language-Action or World-Action? Just Action Models
• EviRover: Reinforcing Agentic Perception Beyond a Glance
• QwenGyre: An Elastic Reinforcement Learning Framework for Training xLong-Horizon Agents
• Efficiently Learning To Reason or Not to Reason: Root-token Policy Optimization for Adaptive Thinking
• CANVAS: A Benchmark for Vision-Language Models on Tool-Based User Interface Design
• SweetSpot: An Analytical Model for Predicting Energy Efficiency of LLM Inference
• Inference-Time Scaling of Diffusion Models via Progressive Seed Pruning
• Super-Resolved Canopy Height Mapping from Sentinel-2 Time Series Using Airborne LiDAR HD Reference Data across Metropolitan France
• RAMEN: Resolution-Adjustable Multimodal Encoder for Earth Observation
• DF26: We Cannot Tell Fake From Real Anymore
• PhyProbe: Rethinking Physical Consistency Evaluation in Generated Videos