AI.info

Latest Research — Page 27

Browse Latest Research on AI.info.

• When Users Are Happy but Agents Are Wrong: Multi-Dimensional Evaluation of Tool-Augmented Dialogue
• RePro: Training Language Models to Faithfully Recycle the Web for Pretraining
• Efficient semantic uncertainty quantification in language models via diversity-steered sampling
• Progressive Learned Image Compression for Machine Perception
• Reflective Translation: Improving Low-Resource Machine Translation via Structured Self-Reflection
• EPRBench: A High-Quality Benchmark Dataset for Event Stream Based Visual Place Recognition
• CoT is Not the Chain of Truth: An Empirical Internal Analysis of Reasoning LLMs for Fake News Generation
• FastJAM: a Fast Joint Alignment Model for Images
• Disentangling Direction and Magnitude in Transformer Representations: A Double Dissociation Through L2-Matched Perturbation Analysis
• Robust Generalization with Adaptive Optimal Transport Priors for Decision-Focused Learning
• ECSEL: Explainable Classification via Signomial Equation Learning
• Discrete Bayesian Sample Inference for Graph Generation
• Breaking the Dyadic Barrier: Rethinking Fairness in Link Prediction Beyond Demographic Parity
• Robust Graph Condensation via Classification Complexity Mitigation
• Retaining by Doing: The Role of On-Policy Data in Mitigating Forgetting
• Beyond Superficial Forgetting: Thorough Unlearning through Knowledge Density Estimation and Block Re-insertion
• Predicting Microbial Interactions Using Graph Neural Networks
• Failure Prediction at Runtime for Generative Robot Policies
• MoEMeta: Mixture-of-Experts Meta Learning for Few-Shot Relational Learning
• RA-VLA: Retrieval-Augmented VLA for Test-Time Adaptation
• Generative Neural Operators through Diffusion Last Layer
• Do LLMs Encode Functional Importance of Reasoning Tokens?
• HugAgent: Evaluating LLMs in Simulating Human-Like Individual Reasoning on Open-Ended Tasks
• Evaluation Principles for MRI-MRA Registration in Trigeminal Neuralgia: An ROI-Centered Neurovascular Benchmark
• What We Don't C: Manifold Disentanglement for Structured Discovery
• FI-TW: An Open Train-Weather Dataset for Railway Delay Analysis in Finland
• AUVIC: Adversarial Unlearning of Visual Concepts for Multi-modal Large Language Models
• Mitra: Mixed Synthetic Priors for Enhancing Tabular Foundation Models
• Bootstrapping Audiovisual Speech Recognition in Zero-AV-Resource Scenarios with Synthetic Visual Data
• HiPHI: A Large-Scale Benchmark for High-Precision Human Motion and Object-Interaction
• Disentangled Representation Learning via Modular Compositional Bias
• Evolve Vision-Language-Action Model into an Agent with On-the-fly Tool-use
• TTP: Test-Time Padding for Adversarial Detection and Robust Adaptation on Vision-Language Models
• Adversarial Attack on Black-Box Multi-Agent by Adaptive Perturbation
• Quo Vadis? Scientific Discovery in the Age of Artificial Intelligence
• Learning to Solve PDEs on Neural Shape Representations
• Linear time small coresets for k-mean clustering of segments with applications
• Lightweight Optimal-Transport Harmonization on Edge Devices
• ScenicRules: An Autonomous Driving Benchmark with Multi-Objective Specifications and Abstract Scenarios
• If You Want to Be Robust, Be Wary of Initialization