arXiv:2608. 12892v1 Announce Type: new Abstract: Activation steering turns localized representations into control directions, but localization alone does not reveal whether a direction has a selective operating regime.
By Jinhao Jing, Tian Zeyu, Lucas Qingyang Fang, Zhisheng Chen, Shuang Chen, Yuhao Luo, Qiannian Zhao
arXiv:2606. 31208v1 Announce Type: new Abstract: Large tabular models (LTMs), i.
By Francesco Capano, Jonas B\"ohler
arXiv:2608. 01130v1 Announce Type: new Abstract: A broad range of models face the mismatch where they are updated through trajectory losses but are evaluated by downstream task reward.
By Yuyang Shen
Dataset distillation (DD) condenses large corpora into compact, information-rich subsets for efficient training and reuse. However, under noisy supervision, DD risks condensing corrupted associations together with useful signals, degrading robustness.
arXiv:2606. 08365v1 Announce Type: cross Abstract: Sparse autoencoder (SAE) features are increasingly used to steer language models, but feature steering is rarely clean: the same intervention can behave inconsistently across contexts and perturb unrelated features.
By Evan Duan
arXiv:2606. 09396v1 Announce Type: cross Abstract: Supervised fine-tuning (SFT) is an efficient approach for downstream task adaptation and often serves as the initialization stage for reinforcement learning (RL), but it can show weaker generalization than RL.
By Ke Wang, Shuangqi Li, Mathieu Salzmann, Pascal Frossard