TMLR - arXivDaily 期刊&会议

2507.10624 2025-11-17 cs.AI cs.LG

Comprehension Without Competence: Architectural Limits of LLMs in Symbolic Computation and Reasoning

Zheng Zhang

机构 * Amazon Web Services（亚马逊网络服务）

Comments v2: Two TMLR revision rounds addressing reviewer feedback. Added real-world validation (3.4), interpretability analysis (7), computational hallucination framework, strengthened theory. v3: Sec 3.2 - added transformer architecture diagram, clarified UAT capacity vs computational limits, improved role specialization theorem presentation

2506.06127 2025-11-17 cs.LG

Flow-Attentional Graph Neural Networks

Pascal Plettenberg, Dominik Köhler, Bernhard Sick, Josephine M. Thomas

机构 * University of Kassel（卡塞尔大学）； University of Greifswald（格里夫斯瓦尔德大学）

Comments Accepted @ Transactions on Machine Learning Research (TMLR): https://openreview.net/forum?id=tOzg7UxTPD

2511.10872 2025-11-17 cs.LG cs.AI

Incorporating Spatial Information into Goal-Conditioned Hierarchical Reinforcement Learning via Graph Representations

Shuyuan Zhang, Zihan Wang, Xiao-Wen Chang, Doina Precup

机构 * McGill University（麦吉尔大学）； Mila ； Google Deepmind（谷歌DeepMind）； CIFAR Fellow

Comments Transactions on Machine Learning Research (2025)

2409.08434 2025-11-17 cs.LG

Predictive Control and Regret Analysis of Non-Stationary MDP with Look-ahead Information

Ziyi Zhang, Yorie Nakahira, Guannan Qu

机构 * Department of Electrical and Computer Engineering（电气与计算机工程系）； Carnegie Mellon University（卡内基梅隆大学）

Comments TMLR, 2025

2407.00482 2025-11-17 cs.LG cs.AI cs.CV cs.CY cs.IT math.IT

Towards Formalizing Spuriousness of Biased Datasets Using Partial Information Decomposition

Barproda Halder, Faisal Hamman, Pasan Dissanayake, Qiuyi Zhang, Ilia Sucholutsky, Sanghamitra Dutta

机构 * University of Maryland, College Park（马里兰大学学院公园分校）； Google Research（谷歌研究）； Princeton University（普林斯顿大学）

Comments Accepted at Transactions on Machine Learning Research (TMLR)