AI agents are commonly aligned with "human values" through reinforcement learning from human feedback (RLHF), where a single reward model is learned from aggregated human feedback and used to align an agent's behavior. However, human values are not homogeneous--different people hold distinct and sometimes conflicting values. Aggregating feedback into a single reward model risks disproportionately suppressing minority preferences. To address this, we present a novel reward modeling approach for learning individualized reward models. Our approach uses a language model to guide users through reflective dialogues where they critique agent behavior and construct their preferences. This personalized dialogue history, containing the user's reflections and critiqued examples, is then used as context for another language model that serves as an individualized reward function (what we call a "verbal reward model") for evaluating new trajectories. In studies with 30 participants, our method achieved a 9-12% improvement in accuracy over non-reflective verbal reward models while being more sample efficient than traditional supervised learning methods.

Although deep learning has made significant progress on fixed large-scale datasets, it typically encounters challenges regarding improperly detecting new/unseen classes in the open-world classification, over-parametrized, and overfitting small samples. In contrast, biological systems can overcome the above difficulties very well. Individuals inherit an innate gene from collective creatures that have evolved over hundreds of millions of years, and can learn new skills through a few examples. Inspired by this, we propose a practical collective-individual paradigm where open-world tasks are trained in sequence using an evolution (expandable) network. To be specific, we innovatively introduce learngene that inherits the meta-knowledge from the collective model and reconstructs a new lightweight individual model for the target task, to realize the collective-individual paradigm. Particularly, we present a novel criterion that can discover the learngene in the collective model, according to the gradient information. Finally, the individual model is trained only with a few samples in the absence of the source data. We demonstrate the effectiveness of our approach in an extensive empirical study and theoretical analysis.

Labeling nodes in a network is an important problem that has seen a growing interest. A number of methods that exploit both local and relational information have been developed for this task. Acquiring the labels for a few nodes at inference time can greatly improve the accuracy, however the question of figuring out which node labels to acquire is challenging. Previous approaches have been based on simple structural properties. Here, we present a novel technique, which we refer to as reflect and correct,that can learn and predict when the underlying classification system is likely to make mistakes and it suggests acquisitions to correct those mistakes.