AI.info

Latest Research — Page 51

Browse Latest Research on AI.info.

• TaRA: Training-Aware Low-Rank Adaptation Initialization
• EvoTest: Evolutionary Test-Time Learning for Self-Improving Agentic Systems
• CD^2: Constrained Dataset Distillation for Few-Shot Class-Incremental Learning
• Adaptive Conformal Prediction for Quantum Machine Learning
• Teaching Pretrained Language Models to Think Deeper with Retrofitted Recurrence
• NoisyGRPO: Incentivizing Multimodal CoT Reasoning via Noise Injection and Bayesian Estimation
• MAC: A Conversion Rate Prediction Benchmark Featuring Labels Under Multiple Attribution Mechanisms
• Dynamic Capability Scoping for Enterprise AI Agents: A Synthetic Dataset and Three-Source Permission Architecture
• LLMs Are Not Good Strategists, Yet Memory-Enhanced Agency Boosts Reasoning
• Constrained Best Arm Identification with Tests for Feasibility
• Scalable Evaluation and Neural Models for Compositional Generalization
• Mint: A Simple Test-Time Adaptation of Vision-Language Models against Common Corruptions
• PRISMM-Bench: A Benchmark of Peer-Review Grounded Multimodal Inconsistencies
• DenseGRPO: From Sparse to Dense Reward for Flow Matching Model Alignment
• QAQ: Bidirectional Semantic Coherence for Selecting High-Quality Synthetic Code Instructions
• SONIC: Supersizing Motion Tracking for Natural Humanoid Whole-Body Control
• Domain-Specific Data Generation Framework for RAG Adaptation
• Efficient Stochastic Optimisation via Sequential Monte Carlo
• SciFigQual-Bench: A Benchmark for Scientific Figure Quality Assessment with Full-Manuscript Context
• DW-DGAT: Dynamically Weighted Dual Graph Attention Network for Neurodegenerative Disease Diagnosis
• Predicting Public Health Impacts of Electricity Usage
• Challenges in Evaluating Explanation Methods for Static and Evolving Data
• A Closer Look at Knowledge Distillation in Spiking Neural Network Training
• LTD-Bench: Evaluating Large Language Models by Letting Them Draw
• Do Reasoning Representations Help Humans Evaluate LLM Outputs?
• MetaGDPO: Alleviating Catastrophic Forgetting with Metacognitive Knowledge through Group Direct Preference Optimization
• OGD4All: A Framework for Accessible Interaction with Geospatial Open Government Data Based on Large Language Models
• Contrastive Spectral Rectification: Test-Time Defense towards Zero-shot Adversarial Robustness of CLIP
• Think Deep, Not Just Long: Measuring LLM Reasoning Effort via Deep-Thinking Tokens
• Model-Based Reinforcement Learning in Discrete-Action Non-Markovian Reward Decision Processes
• Asking like Socrates: Socrates helps VLMs understand remote sensing images
• Improved denoising diffusion probabilistic models with efficient non-diagonal covariance modeling
• Real-World Adverse Weather Image Restoration via Dual-Level Reinforcement Learning with High-Quality Cold Start
• Verifying Chain-of-Thought Reasoning via Its Computational Graph
• Re-Align: Structured Reasoning-guided Alignment for In-Context Image Generation and Editing
• ICDAR 2026 Competition on Information Extraction from Atomic Layer Deposition/Etching (ALD/E) Scientific Figures
• SynTSBench: Rethinking Temporal Pattern Learning in Deep Learning Models for Time Series
• Cultural Alien Sampler: Open-ended art generation balancing originality and coherence
• STEM Faculty Perspectives on Generative AI in Higher Education
• Enabling Delayed-Full Charging Through Transformer-Based Real-Time-to-Departure Modeling for EV Battery Longevity