Improving Steering Vectors by Targeting Sparse Autoencoder Featuresby Sviatoslav Chalnev, Matthew Siu, Arthur ConmyFirst submitted…
Ask, and it shall be given: On the Turing completeness of promptingby Ruizhong Qiu, Zhe…
In-Context Fine-Tuning for Time-Series Foundation Modelsby Abhimanyu Das, Matthew Faw, Rajat Sen, Yichen ZhouFirst submitted…
RAGraph: A General Retrieval-Augmented Graph Learning Frameworkby Xinke Jiang, Rihong Qiu, Yongxin Xu, Wentao Zhang,…
Can Language Models Perform Robust Reasoning in Chain-of-thought Prompting with Noisy Rationales?by Zhanke Zhou, Rong…
RL-STaR: Theoretical Analysis of Reinforcement Learning Frameworks for Self-Taught Reasonerby Fu-Chieh Chang, Yu-Ting Lee, Hui-Ying…
Wide Two-Layer Networks can Learn from Adversarial Perturbationsby Soichiro Kumano, Hiroshi Kera, Toshihiko YamasakiFirst submitted…
ProTransformer: Robustify Transformers via Plug-and-Play Paradigmby Zhichao Hou, Weizhi Gao, Yuchen Shen, Feiyi Wang, Xiaorui…
Are Large-Language Models Graph Algorithmic Reasoners?by Alexander K Taylor, Anthony Cuturrufo, Vishal Yathish, Mingyu Derek…
Reliable and Compact Graph Fine-tuning via GraphSparse Promptingby Bo Jiang, Hao Wu, Beibei Wang, Jin…