Summary of Extractive Structures Learned in Pretraining Enable Generalization on Finetuned Facts, by Jiahai Feng et al.
Extractive Structures Learned in Pretraining Enable Generalization on Finetuned Factsby Jiahai Feng, Stuart Russell, Jacob…
Extractive Structures Learned in Pretraining Enable Generalization on Finetuned Factsby Jiahai Feng, Stuart Russell, Jacob…
Sometimes I am a Tree: Data Drives Unstable Hierarchical Generalizationby Tian Qin, Naomi Saphra, David…
Mind the Gap: Towards Generalizable Autonomous Penetration Testing via Domain Randomization and Meta-Reinforcement Learningby Shicheng…
Weak-to-Strong Generalization Through the Data-Centric Lensby Changho Shin, John Cooper, Frederic SalaFirst submitted to arxiv…
The broader spectrum of in-context learningby Andrew Kyle Lampinen, Stephanie C. Y. Chan, Aaditya K.…
How to Correctly do Semantic Backpropagation on Language-based Agentic Systemsby Wenyi Wang, Hisham A. Alyahya,…
Tight PAC-Bayesian Risk Certificates for Contrastive Learningby Anna Van Elst, Debarghya GhoshdastidarFirst submitted to arxiv…
Surveying the Effects of Quality, Diversity, and Complexity in Synthetic Data From Large Language Modelsby…
UTSD: Unified Time Series Diffusion Modelby Xiangkai Ma, Xiaobin Hong, Wenzhong Li, Sanglu LuFirst submitted…
Few-Shot Learning with Adaptive Weight Masking in Conditional GANsby Jiacheng Hu, Zhen Qi, Jianjun Wei,…