Machine Learning

UMKC Announces New Master of Science in Artificial Intelligence

UMKC announces a new Master of Science in Artificial Intelligence program aimed at addressing workforce demand for AI expertise, set to l...

Improving AI models’ ability to explain their predictions

New technique makes AI models leaner and faster while they’re still learning

[2604.04561] Mapping the Exploitation Surface: A 10,000-Trial Taxonomy of What Makes LLM Agents Exploit Vulnerabilities

Abstract page for arXiv paper 2604.04561: Mapping the Exploitation Surface: A 10,000-Trial Taxonomy of What Makes LLM Agents Exploit Vuln...

[2604.04552] StableTTA: Training-Free Test-Time Adaptation that Improves Model Accuracy on ImageNet1K to 96%

Abstract page for arXiv paper 2604.04552: StableTTA: Training-Free Test-Time Adaptation that Improves Model Accuracy on ImageNet1K to 96%

[2604.04490] RAVEN: Radar Adaptive Vision Encoders for Efficient Chirp-wise Object Detection and Segmentation

Abstract page for arXiv paper 2604.04490: RAVEN: Radar Adaptive Vision Encoders for Efficient Chirp-wise Object Detection and Segmentation

[2604.04450] Conversational Control with Ontologies for Large Language Models: A Lightweight Framework for Constrained Generation

Abstract page for arXiv paper 2604.04450: Conversational Control with Ontologies for Large Language Models: A Lightweight Framework for C...

[2604.04440] Training Transformers in Cosine Coefficient Space

Abstract page for arXiv paper 2604.04440: Training Transformers in Cosine Coefficient Space

[2604.04418] Justified or Just Convincing? Error Verifiability as a Dimension of LLM Quality

Abstract page for arXiv paper 2604.04418: Justified or Just Convincing? Error Verifiability as a Dimension of LLM Quality

[2604.04411] Responses Fall Short of Understanding: Revealing the Gap between Internal Representations and Responses in Visual Document Understanding

Abstract page for arXiv paper 2604.04411: Responses Fall Short of Understanding: Revealing the Gap between Internal Representations and R...

[2604.04384] Compressible Softmax-Attended Language under Incompressible Attention

Abstract page for arXiv paper 2604.04384: Compressible Softmax-Attended Language under Incompressible Attention

[2604.04359] GROUNDEDKG-RAG: Grounded Knowledge Graph Index for Long-document Question Answering

Abstract page for arXiv paper 2604.04359: GROUNDEDKG-RAG: Grounded Knowledge Graph Index for Long-document Question Answering

[2604.04353] ReFinE: Streamlining UI Mockup Iteration with Research Findings

Abstract page for arXiv paper 2604.04353: ReFinE: Streamlining UI Mockup Iteration with Research Findings

[2604.04306] HighFM: Towards a Foundation Model for Learning Representations from High-Frequency Earth Observation Data

Abstract page for arXiv paper 2604.04306: HighFM: Towards a Foundation Model for Learning Representations from High-Frequency Earth Obser...

[2604.04299] A Persistent Homology Design Space for 3D Point Cloud Deep Learning

Abstract page for arXiv paper 2604.04299: A Persistent Homology Design Space for 3D Point Cloud Deep Learning

[2604.04289] Poisoned Identifiers Survive LLM Deobfuscation: A Case Study on Claude Opus 4.6

Abstract page for arXiv paper 2604.04289: Poisoned Identifiers Survive LLM Deobfuscation: A Case Study on Claude Opus 4.6

[2604.04263] Commercial Persuasion in AI-Mediated Conversations

Abstract page for arXiv paper 2604.04263: Commercial Persuasion in AI-Mediated Conversations

[2604.04211] LOCARD: An Agentic Framework for Blockchain Forensics

Abstract page for arXiv paper 2604.04211: LOCARD: An Agentic Framework for Blockchain Forensics

[2604.04172] GENFIG1: Visual Summaries of Scholarly Work as a Challenge for Vision-Language Models

Abstract page for arXiv paper 2604.04172: GENFIG1: Visual Summaries of Scholarly Work as a Challenge for Vision-Language Models

[2604.04144] Many Preferences, Few Policies: Towards Scalable Language Model Personalization

Abstract page for arXiv paper 2604.04144: Many Preferences, Few Policies: Towards Scalable Language Model Personalization

[2604.04133] Learning Robust Visual Features in Computed Tomography Enables Efficient Transfer Learning for Clinical Tasks

Abstract page for arXiv paper 2604.04133: Learning Robust Visual Features in Computed Tomography Enables Efficient Transfer Learning for ...

[2604.04089] From Paper to Program: A Multi-Stage LLM-Assisted Workflow for Accelerating Quantum Many-Body Algorithm Development

Abstract page for arXiv paper 2604.04089: From Paper to Program: A Multi-Stage LLM-Assisted Workflow for Accelerating Quantum Many-Body A...

[2604.04064] Extracting and Steering Emotion Representations in Small Language Models: A Methodological Comparison

Abstract page for arXiv paper 2604.04064: Extracting and Steering Emotion Representations in Small Language Models: A Methodological Comp...