Reinforcement learning with human feedback (RLHF) fine-tunes a pretrained large language model (LLM) using preference datasets, enabling the LLM to generate outputs that align with human preferences. Given the sensitive nature of these preference datasets held by various clients, there is a need to implement RLHF within a federated learning (FL) framework, where clients are reluctant to share their data due to privacy concerns. To address this, we introduce a feasible framework in which clients collaboratively train a binary selector with their preference datasets using our proposed FedBis. With a well-trained selector, we can further enhance the LLM that generates human-preferred completions. Meanwhile, we propose a novel algorithm, FedBiscuit, that trains multiple selectors by organizing clients into balanced and disjoint clusters based on their preferences. Compared to the FedBis, FedBiscuit demonstrates superior performance in simulating human preferences for pairwise completions. Our extensive experiments on federated human preference datasets -- marking the first benchmark to address heterogeneous data partitioning among clients -- demonstrate that FedBiscuit outperforms FedBis and even surpasses traditional centralized training.

Mobile manipulators always need to determine feasible base positions prior to carrying out navigation-manipulation tasks. Real-world environments are often cluttered with various furniture, obstacles, and dozens of other objects. Efficiently computing base positions poses a challenge. In this work, we introduce a framework named MoMa-Pos to address this issue. MoMa-Pos first learns to predict a small set of objects that, taken together, would be sufficient for finding base positions using a graph embedding architecture. MoMa-Pos then calculates standing positions by considering furniture structures, robot models, and obstacles comprehensively. We have extensively evaluated the proposed MoMa-Pos across different settings (e.g., environment and algorithm parameters) and with various mobile manipulators. Our empirical results show that MoMa-Pos demonstrates remarkable effectiveness and efficiency in its performance, surpassing the methods in the literature. %, but also is adaptable to cluttered environments and different robot models. Supplementary material can be found at \url{https://yding25.com/MoMa-Pos}.