Asymmetric Certified Robustness via Feature-Convex Neural Networks TLDR: We propose the asymmetric certified robustness problem, which requires certified robustness for only one class and reflects

Goal Representations for Instruction Following <!– Figure title. Figure caption. This image is centered and set to 50% page width. –> A longstanding goal of

Goal Representations for Instruction Following <!– Figure title. Figure caption. This image is centered and set to 50% page width. –> A longstanding goal of

Rethinking the Role of PPO in RLHF TL;DR: In RLHF, there’s tension between the reward learning phase, which uses human preference in the form of

Rethinking the Role of PPO in RLHF TL;DR: In RLHF, there’s tension between the reward learning phase, which uses human preference in the form of

Training Diffusion Models with Reinforcement Learning replay Diffusion models have recently emerged as the de facto standard for generating complex, high-dimensional outputs. You may know

Training Diffusion Models with Reinforcement Learning replay Diffusion models have recently emerged as the de facto standard for generating complex, high-dimensional outputs. You may know

Figure 1: stepwise behavior in self-supervised learning. When training common SSL algorithms, we find that the loss descends in a stepwise fashion (top left) and

Figure 1: stepwise behavior in self-supervised learning. When training common SSL algorithms, we find that the loss descends in a stepwise fashion (top left) and

<!– –> Figure 1: CoarsenConf architecture. <!– (I) The encoder $q_phi(z| X, mathcal{R})$ takes the fine-grained (FG) ground truth conformer $X$, RDKit approximate conformer $mathcal{R}$