AI Papers arXiv

Don't Vibe Code, Do Skele-Code: Interactive No-Code Notebooks for Subject Matter Experts to Build Lower-Cost Agentic Workflows Sriram Gopalakrishnan arxiv.org/abs/2603.18122 [𝚌𝚜.𝙰𝙸 𝚌𝚜.𝙷𝙲 𝚌𝚜.𝙿𝙻 𝚌𝚜.𝚂𝚈 𝚎𝚎𝚜𝚜.𝚂𝚈]

Efficient Dense Crowd Trajectory Prediction Via Dynamic Clustering Antonius Bima Murti Wijaya, Paul Henderson, Marwa Mahmoud arxiv.org/abs/2603.18166 [𝚌𝚜.𝙰𝙸]

TeachingCoach: A Fine-Tuned Scaffolding Chatbot for Instructional Guidance to Instructors Isabel Molnar, Peiyu Li, Si Chen, Sugana Chawla, James Lang, Ronald Metoyer, Ting Hua, Nitesh V. Chawla arxiv.org/abs/2603.18189 [𝚌𝚜.𝙰𝙸]

Access Controlled Website Interaction for Agentic AI with Delegated Critical Tasks Sunyoung Kim, Hokeun Kim arxiv.org/abs/2603.18197 [𝚌𝚜.𝙰𝙸 𝚌𝚜.𝙲𝚁 𝚌𝚜.𝙽𝙸]

A Computationally Efficient Learning of Artificial Intelligence System Reliability Considering Error Propagation Fenglian Pan, Yinwei Zhang, Yili Hong, Larry Head, Jian Liu arxiv.org/abs/2603.18201 [𝚌𝚜.𝙰𝙸 𝚜𝚝𝚊𝚝.𝙲𝙾]

Retrieval-Augmented LLM Agents: Learning to Learn from Experience Thomas Palmeira Ferraz, Romain Deffayet, Vassilina Nikoulina, Hervé Déjean, Stéphane Clinchant arxiv.org/abs/2603.18272 [𝚌𝚜.𝙰𝙸 𝚌𝚜.𝙲𝙻]

EDM-ARS: A Domain-Specific Multi-Agent System for Automated Educational Data Mining Research Chenguang Pan, Zhou Zhang, Weixuan Xiao, Chengyuan Yao arxiv.org/abs/2603.18273 [𝚌𝚜.𝙰𝙸]

CORE: Robust Out-of-Distribution Detection via Confidence and Orthogonal Residual Scoring Jin Mo Yang, Hyung-Sin Kim, Saewoong Bahk arxiv.org/abs/2603.18290 [𝚌𝚜.𝙰𝙸]

The Validity Gap in Health AI Evaluation: A Cross-Sectional Analysis of Benchmark Composition Alvin Rajkomar, Pavan Sudarshan, Angela Lai, Lily Peng arxiv.org/abs/2603.18294 [𝚌𝚜.𝙰𝙸]

Consumer-to-Clinical Language Shifts in Ambient AI Draft Notes and Clinician-Finalized Documentation: A Multi-level Analysis Ha Na Cho, Yawen Guo, Sairam Sutari, Emilie Chow, Steven Tam, Danielle Perret, Deepti Pandita, Kai Zheng arxiv.org/abs/2603.18327 [𝚌𝚜.𝙰𝙸]

FaithSteer-BENCH: A Deployment-Aligned Stress-Testing Benchmark for Inference-Time Steering Zikang Ding, Qiying Hu, Yi Zhang, Hongji Li, Junchi Yao, Hongbo Liu, Lijie Hu arxiv.org/abs/2603.18329 [𝚌𝚜.𝙰𝙸]

MemArchitect: A Policy Driven Memory Governance Layer Lingavasan Suresh Kumar, Yang Ba, Rong Pan arxiv.org/abs/2603.18330 [𝚌𝚜.𝙰𝙸 𝚌𝚜.𝙷𝙲 𝚌𝚜.𝙻𝙶 𝚌𝚜.𝙼𝙰]

Understanding the Theoretical Foundations of Deep Neural Networks through Differential Equations Hongjue Zhao, Yizhuo Chen, Yuchen Wang, Hairong Qi, Lui Sha, Tarek Abdelzaher, Huajie Shao arxiv.org/abs/2603.18331 [𝚌𝚜.𝙰𝙸]

Large-Scale Analysis of Political Propaganda on Moltbook Julia Jose, Meghna Manoj Nair, Rachel Greenstadt arxiv.org/abs/2603.18349 [𝚌𝚜.𝙰𝙸 𝚌𝚜.𝙲𝙻]

Interpretability without actionability: mechanistic methods cannot correct language model errors despite near-perfect internal representations Sanjay Basu, Sadiq Y. Patel, Parth Sheth, Bhairavi Muralidharan, Namrata Elamaran, Aakriti Kinra, … arxiv.org/abs/2603.18353 [𝚌𝚜.𝙰𝙸]

LGESynthNet: Controlled Scar Synthesis for Improved Scar Segmentation in Cardiac LGE-MRI Imaging Athira J. Jacob, Puneet Sharma, Daniel Rueckert arxiv.org/abs/2603.18356 [𝚌𝚜.𝙰𝙸]

From Weak Cues to Real Identities: Evaluating Inference-Driven De-Anonymization in LLM Agents Myeongseob Ko, Jihyun Jeong, Sumiran Singh Thakur, Gyuhak Kim, Ruoxi Jia arxiv.org/abs/2603.18382 [𝚌𝚜.𝙰𝙸]

Reflection in the Dark: Exposing and Escaping the Black Box in Reflective Prompt Optimization Shiyan Liu, Qifeng Xia, Qiyun Xia, Yisheng Liu, Xinyu Yu, Rui Qu arxiv.org/abs/2603.18388 [𝚌𝚜.𝙰𝙸 𝚌𝚜.𝙼𝙰]

From Topic to Transition Structure: Unsupervised Concept Discovery at Corpus Scale via Predictive Associative Memory Jason Dury arxiv.org/abs/2603.18420 [𝚌𝚜.𝙰𝙸 𝚌𝚜.𝙲𝙻 𝚌𝚜.𝙸𝚁 𝚌𝚜.𝙻𝙶]

Prune-then-Quantize or Quantize-then-Prune? Understanding the Impact of Compression Order in Joint Model Compression Minjun Kim, Jaehyeon Choi, Hyunwoo Yang, Jongjin Kim, Jinho Song, U Kang arxiv.org/abs/2603.18426 [𝚌𝚜.𝙰𝙸]