Residual Policy Gradient: A Reward View of KL-regularized Objective

Reinforcement Learning and Imitation Learning have achieved widespread success in many domains but remain constrained during real-world deployment. One of the main issues is the additional requirements that were not considered during training. To address this challenge, policy customization has been introduced, aiming to adapt a prior policy while preserving its inherent properties and meeting new task-specific requirements. A principled approach to policy customization is Residual Q-Learning (RQL), which formulates the problem as a Markov Decision Process (MDP) and derives a family of value-based learning algorithms. However, RQL has not yet been applied to policy gradient methods, which restricts its applicability, especially in tasks where policy gradient has already proven more effective. In this work, we first derive a concise form of Soft Policy Gradient as a preliminary. Building on this, we introduce Residual Policy Gradient (RPG), which extends RQL to policy gradient methods, allowing policy customization in gradient-based RL settings. With the view of RPG, we rethink the KL-regularized objective widely used in RL fine-tuning. We show that under certain assumptions, KL-regularized objective leads to a maximum-entropy policy that balances the inherent properties and task-specific requirements on a reward-level. Our experiments in MuJoCo demonstrate the effectiveness of Soft Policy Gradient and Residual Policy Gradient.

External Large Foundation Model: How to Efficiently Serve Trillions of Parameters for Online Ads Recommendation

Ads recommendation is a prominent service of online advertising systems and has been actively studied. Recent studies indicate that scaling-up and advanced design of the recommendation model can bring significant performance improvement. However, with a larger model scale, such prior studies have a significantly increasing gap from industry as they often neglect two fundamental challenges in industrial-scale applications. First, training and inference budgets are restricted for the model to be served, exceeding which may incur latency and impair user experience. Second, large-volume data arrive in a streaming mode with data distributions dynamically shifting, as new users/ads join and existing users/ads leave the system. We propose the External Large Foundation Model (ExFM) framework to address the overlooked challenges. Specifically, we develop external distillation and a data augmentation system (DAS) to control the computational cost of training/inference while maintaining high performance. We design the teacher in a way like a foundation model (FM) that can serve multiple students as vertical models (VMs) to amortize its building cost. We propose Auxiliary Head and Student Adapter to mitigate the data distribution gap between FM and VMs caused by the streaming data issue. Comprehensive experiments on internal industrial-scale applications and public datasets demonstrate significant performance gain by ExFM.

Physics-Aware Robotic Palletization with Online Masking Inference

The efficient planning of stacking boxes, especially in the online setting where the sequence of item arrivals is unpredictable, remains a critical challenge in modern warehouse and logistics management. Existing solutions often address box size variations, but overlook their intrinsic and physical properties, such as density and rigidity, which are crucial for real-world applications. We use reinforcement learning (RL) to solve this problem by employing action space masking to direct the RL policy toward valid actions. Unlike previous methods that rely on heuristic stability assessments which are difficult to assess in physical scenarios, our framework utilizes online learning to dynamically train the action space mask, eliminating the need for manual heuristic design. Extensive experiments demonstrate that our proposed method outperforms existing state-of-the-arts. Furthermore, we deploy our learned task planner in a real-world robotic palletizer, validating its practical applicability in operational settings.

MixDec Sampling: A Soft Link-based Sampling Method of Graph Neural Network for Recommendation

Graph neural networks have been widely used in recent recommender systems, where negative sampling plays an important role. Existing negative sampling methods restrict the relationship between nodes as either hard positive pairs or hard negative pairs. This leads to the loss of structural information, and lacks the mechanism to generate positive pairs for nodes with few neighbors. To overcome limitations, we propose a novel soft link-based sampling method, namely MixDec Sampling, which consists of Mixup Sampling module and Decay Sampling module. The Mixup Sampling augments node features by synthesizing new nodes and soft links, which provides sufficient number of samples for nodes with few neighbors. The Decay Sampling strengthens the digestion of graph structure information by generating soft links for node embedding learning. To the best of our knowledge, we are the first to model sampling relationships between nodes by soft links in GNN-based recommender systems. Extensive experiments demonstrate that the proposed MixDec Sampling can significantly and consistently improve the recommendation performance of several representative GNN-based models on various recommendation benchmarks.

DrugImproverGPT: A Large Language Model for Drug Optimization with Fine-Tuning via Structured Policy Optimization

Finetuning a Large Language Model (LLM) is crucial for generating results towards specific objectives. This research delves into the realm of drug optimization and introduce a novel reinforcement learning algorithm to finetune a drug optimization LLM-based generative model, enhancing the original drug across target objectives, while retains the beneficial chemical properties of the original drug. This work is comprised of two primary components: (1) DrugImprover: A framework tailored for improving robustness and efficiency in drug optimization. It includes a LLM designed for drug optimization and a novel Structured Policy Optimization (SPO) algorithm, which is theoretically grounded. This algorithm offers a unique perspective for fine-tuning the LLM-based generative model by aligning the improvement of the generated molecule with the input molecule under desired objectives. (2) A dataset of 1 million compounds, each with OEDOCK docking scores on 5 human proteins associated with cancer cells and 24 binding sites from SARS-CoV-2 virus. We conduct a comprehensive evaluation of SPO and demonstrate its effectiveness in improving the original drug across target properties. Our code and dataset will be publicly available at: https://github.com/xuefeng-cs/DrugImproverGPT.

Active Advantage-Aligned Online Reinforcement Learning with Offline Data

Online reinforcement learning (RL) enhances policies through direct interactions with the environment, but faces challenges related to sample efficiency. In contrast, offline RL leverages extensive pre-collected data to learn policies, but often produces suboptimal results due to limited data coverage. Recent efforts have sought to integrate offline and online RL in order to harness the advantages of both approaches. However, effectively combining online and offline RL remains challenging due to issues that include catastrophic forgetting, lack of robustness and sample efficiency. In an effort to address these challenges, we introduce A3 RL , a novel method that actively selects data from combined online and offline sources to optimize policy improvement. We provide theoretical guarantee that validates the effectiveness our active sampling strategy and conduct thorough empirical experiments showing that our method outperforms existing state-of-the-art online RL techniques that utilize offline data. Our code will be publicly available at: https://github.com/xuefeng-cs/A3RL.

Online Reward-Weighted Fine-Tuning of Flow Matching with Wasserstein Regularization

Recent advancements in reinforcement learning (RL) have achieved great success in fine-tuning diffusion-based generative models. However, fine-tuning continuous flow-based generative models to align with arbitrary user-defined reward functions remains challenging, particularly due to issues such as policy collapse from overoptimization and the prohibitively high computational cost of likelihoods in continuous-time flows. In this paper, we propose an easy-to-use and theoretically sound RL fine-tuning method, which we term Online Reward-Weighted Conditional Flow Matching with Wasserstein-2 Regularization (ORW-CFM-W2). Our method integrates RL into the flow matching framework to fine-tune generative models with arbitrary reward functions, without relying on gradients of rewards or filtered datasets. By introducing an online reward-weighting mechanism, our approach guides the model to prioritize high-reward regions in the data manifold. To prevent policy collapse and maintain diversity, we incorporate Wasserstein-2 (W2) distance regularization into our method and derive a tractable upper bound for it in flow matching, effectively balancing exploration and exploitation of policy optimization. We provide theoretical analyses to demonstrate the convergence properties and induced data distributions of our method, establishing connections with traditional RL algorithms featuring Kullback-Leibler (KL) regularization and offering a more comprehensive understanding of the underlying mechanisms and learning behavior of our approach. Extensive experiments on tasks including target image generation, image compression, and text-image alignment demonstrate the effectiveness of our method, where our method achieves optimal policy convergence while allowing controllable trade-offs between reward maximization and diversity preservation.

Beyond Reward Hacking: Causal Rewards for Large Language Model Alignment

Recent advances in large language models (LLMs) have demonstrated significant progress in performing complex tasks. While Reinforcement Learning from Human Feedback (RLHF) has been effective in aligning LLMs with human preferences, it is susceptible to spurious correlations in reward modeling. Consequently, it often introduces biases-such as length bias, sycophancy, conceptual bias, and discrimination that hinder the model's ability to capture true causal relationships. To address this, we propose a novel causal reward modeling approach that integrates causal inference to mitigate these spurious correlations. Our method enforces counterfactual invariance, ensuring reward predictions remain consistent when irrelevant variables are altered. Through experiments on both synthetic and real-world datasets, we show that our approach mitigates various types of spurious correlations effectively, resulting in more reliable and fair alignment of LLMs with human preferences. As a drop-in enhancement to the existing RLHF workflow, our causal reward modeling provides a practical way to improve the trustworthiness and fairness of LLM finetuning.

FDPP: Fine-tune Diffusion Policy with Human Preference

Imitation learning from human demonstrations enables robots to perform complex manipulation tasks and has recently witnessed huge success. However, these techniques often struggle to adapt behavior to new preferences or changes in the environment. To address these limitations, we propose Fine-tuning Diffusion Policy with Human Preference (FDPP). FDPP learns a reward function through preference-based learning. This reward is then used to fine-tune the pre-trained policy with reinforcement learning (RL), resulting in alignment of pre-trained policy with new human preferences while still solving the original task. Our experiments across various robotic tasks and preferences demonstrate that FDPP effectively customizes policy behavior without compromising performance. Additionally, we show that incorporating Kullback-Leibler (KL) regularization during fine-tuning prevents over-fitting and helps maintain the competencies of the initial policy.

Testing Human-Hand Segmentation on In-Distribution and Out-of-Distribution Data in Human-Robot Interactions Using a Deep Ensemble Model

Reliable detection and segmentation of human hands are critical for enhancing safety and facilitating advanced interactions in human-robot collaboration. Current research predominantly evaluates hand segmentation under in-distribution (ID) data, which reflects the training data of deep learning (DL) models. However, this approach fails to address out-of-distribution (OOD) scenarios that often arise in real-world human-robot interactions. In this study, we present a novel approach by evaluating the performance of pre-trained DL models under both ID data and more challenging OOD scenarios. To mimic realistic industrial scenarios, we designed a diverse dataset featuring simple and cluttered backgrounds with industrial tools, varying numbers of hands (0 to 4), and hands with and without gloves. For OOD scenarios, we incorporated unique and rare conditions such as finger-crossing gestures and motion blur from fast-moving hands, addressing both epistemic and aleatoric uncertainties. To ensure multiple point of views (PoVs), we utilized both egocentric cameras, mounted on the operator's head, and static cameras to capture RGB images of human-robot interactions. This approach allowed us to account for multiple camera perspectives while also evaluating the performance of models trained on existing egocentric datasets as well as static-camera datasets. For segmentation, we used a deep ensemble model composed of UNet and RefineNet as base learners. Performance evaluation was conducted using segmentation metrics and uncertainty quantification via predictive entropy. Results revealed that models trained on industrial datasets outperformed those trained on non-industrial datasets, highlighting the importance of context-specific training. Although all models struggled with OOD scenarios, those trained on industrial datasets demonstrated significantly better generalization.