Data and model heterogeneity are two core issues in Heterogeneous Federated Learning (HtFL). In scenarios with heterogeneous model architectures, aggregating model parameters becomes infeasible, leading to the use of prototypes (i.e., class representative feature vectors) for aggregation and guidance. However, they still experience a mismatch between the extra guiding objective and the client's original local objective when aligned with global prototypes. With theoretical guarantees, FedL2G efficiently implements the learning-to-guide process using only first-order derivatives w.r.t. We conduct extensive experiments on two data heterogeneity and six model heterogeneity settings using 14 heterogeneous model architectures (e.g., CNNs and ViTs) to demonstrate FedL2G's superior performance compared to six counterparts. With the rapid development of AI techniques (Touvron et al., 2023; Achiam et al., 2023), public data has been consumed gradually, raising the need to access local data inside devices or institutions (Ye et al., 2024). However, directly using local data often raises privacy concerns (Nguyen et al., 2021). Federated Learning (FL) is a promising privacy-preserving approach that enables collaborative model training across multiple clients (devices or institutions) in a distributed manner without the need to move the actual data outside clients (Kairouz et al., 2019; Li et al., 2020).

The advancement of robotic systems has revolutionized numerous industries, yet their operation often demands specialized technical knowledge, limiting accessibility for non-expert users. This paper introduces ROSA (Robot Operating System Agent), an AI-powered agent that bridges the gap between the Robot Operating System (ROS) and natural language interfaces. By leveraging state-of-the-art language models and integrating open-source frameworks, ROSA enables operators to interact with robots using natural language, translating commands into actions and interfacing with ROS through well-defined tools. ROSA's design is modular and extensible, offering seamless integration with both ROS1 and ROS2, along with safety mechanisms like parameter validation and constraint enforcement to ensure secure, reliable operations. While ROSA is originally designed for ROS, it can be extended to work with other robotics middle-wares to maximize compatibility across missions. ROSA enhances human-robot interaction by democratizing access to complex robotic systems, empowering users of all expertise levels with multi-modal capabilities such as speech integration and visual perception. Ethical considerations are thoroughly addressed, guided by foundational principles like Asimov's Three Laws of Robotics, ensuring that AI integration promotes safety, transparency, privacy, and accountability. By making robotic technology more user-friendly and accessible, ROSA not only improves operational efficiency but also sets a new standard for responsible AI use in robotics and potentially future mission operations. This paper introduces ROSA's architecture and showcases initial mock-up operations in JPL's Mars Yard, a laboratory, and a simulation using three different robots. The core ROSA library is available as open-source.

Not yet. We present SPACE, a benchmark that systematically evaluates spatial cognition in frontier models. Our benchmark builds on decades of research in cognitive science. It evaluates large-scale mapping abilities that are brought to bear when an organism traverses physical environments, smaller-scale reasoning about object shapes and layouts, and cognitive infrastructure such as spatial attention and memory. For many tasks, we instantiate parallel presentations via text and images, allowing us to benchmark both large language models and large multimodal models. Results suggest that contemporary frontier models fall short of the spatial intelligence of animals, performing near chance level on a number of classic tests of animal cognition.

Instruction following is a key capability for LLMs. However, recent studies have shown that LLMs often struggle with instructions containing multiple constraints (e.g. a request to create a social media post "in a funny tone" with "no hashtag"). Despite this, most evaluations focus solely on synthetic data. To address this, we introduce RealInstruct, the first benchmark designed to evaluate LLMs' ability to follow real-world multi-constrained instructions by leveraging queries real users asked AI assistants. We also investigate model-based evaluation as a cost-effective alternative to human annotation for this task. Our findings reveal that even the proprietary GPT-4 model fails to meet at least one constraint on over 21% of instructions, highlighting the limitations of state-of-the-art models. To address the performance gap between open-source and proprietary models, we propose the Decompose, Critique and Refine (DeCRIM) self-correction pipeline, which enhances LLMs' ability to follow constraints. DeCRIM works by decomposing the original instruction into a list of constraints and using a Critic model to decide when and where the LLM's response needs refinement. Our results show that DeCRIM improves Mistral's performance by 7.3% on RealInstruct and 8.0% on IFEval even with weak feedback. Moreover, we demonstrate that with strong feedback, open-source LLMs with DeCRIM can outperform GPT-4 on both benchmarks.

From common-sense reasoning to domain-specific tasks, parameter-efficient fine tuning (PEFT) methods for large language models (LLMs) have showcased significant performance improvements on downstream tasks. However, fine-tuned LLMs often struggle with overconfidence in uncertain predictions, particularly due to sparse training data. This overconfidence reflects poor epistemic uncertainty calibration, which arises from limitations in the model's ability to generalize with limited data. Existing PEFT uncertainty quantification methods for LLMs focus on the post fine-tuning stage and thus have limited capability in calibrating epistemic uncertainty. To address these limitations, we propose Functional-Level Uncertainty Quantification for Calibrated Fine-Tuning (UQ4CT), which captures and calibrates functional-level epistemic uncertainty during the fine-tuning stage via a mixture-of-expert framework. We show that UQ4CT reduces Expected Calibration Error (ECE) by more than $25\%$ while maintaining high accuracy across $5$ benchmarks. Furthermore, UQ4CT maintains superior ECE performance with high accuracy under distribution shift, showcasing improved generalizability.

Fairness in artificial intelligence and machine learning (AI/ML) models is becoming critically important, especially as decisions made by these systems impact diverse groups. In education, a vital sector for all countries, the widespread application of AI/ML systems raises specific concerns regarding fairness. Current research predominantly focuses on fairness for individual sensitive features, which limits the comprehensiveness of fairness assessments. This paper introduces FAIREDU, a novel and effective method designed to improve fairness across multiple sensitive features. Through extensive experiments, we evaluate FAIREDU effectiveness in enhancing fairness without compromising model performance. The results demonstrate that FAIREDU addresses intersectionality across features such as gender, race, age, and other sensitive features, outperforming state-of-the-art methods with minimal effect on model accuracy. The paper also explores potential future research directions to enhance further the method robustness and applicability to various machine-learning models and datasets.

The Abstraction and Reasoning Corpus (ARC) is a popular benchmark focused on visual reasoning in the evaluation of Artificial Intelligence systems. In its original framing, an ARC task requires solving a program synthesis problem over small 2D images using a few input-output training pairs. In this work, we adopt the recently popular data-driven approach to the ARC and ask whether a Vision Transformer (ViT) can learn the implicit mapping, from input image to output image, that underlies the task. We show that a ViT -- otherwise a state-of-the-art model for images -- fails dramatically on most ARC tasks even when trained on one million examples per task. This points to an inherent representational deficiency of the ViT architecture that makes it incapable of uncovering the simple structured mappings underlying the ARC tasks. Building on these insights, we propose ViTARC, a ViT-style architecture that unlocks some of the visual reasoning capabilities required by the ARC. Specifically, we use a pixel-level input representation, design a spatially-aware tokenization scheme, and introduce a novel object-based positional encoding that leverages automatic segmentation, among other enhancements. Our task-specific ViTARC models achieve a test solve rate close to 100% on more than half of the 400 public ARC tasks strictly through supervised learning from input-output grids. This calls attention to the importance of imbuing the powerful (Vision) Transformer with the correct inductive biases for abstract visual reasoning that are critical even when the training data is plentiful and the mapping is noise-free. Hence, ViTARC provides a strong foundation for future research in visual reasoning using transformer-based architectures.

Learning from human feedback has gained traction in fields like robotics and natural language processing in recent years. While prior works mostly rely on human feedback in the form of comparisons, language is a preferable modality that provides more informative insights into user preferences. In this work, we aim to incorporate comparative language feedback to iteratively improve robot trajectories and to learn reward functions that encode human preferences. To achieve this goal, we learn a shared latent space that integrates trajectory data and language feedback, and subsequently leverage the learned latent space to improve trajectories and learn human preferences. To the best of our knowledge, we are the first to incorporate comparative language feedback into reward learning. Our simulation experiments demonstrate the effectiveness of the learned latent space and the success of our learning algorithms. We also conduct human subject studies that show our reward learning algorithm achieves a 23.9% higher subjective score on average and is 11.3% more time-efficient compared to preference-based reward learning, underscoring the superior performance of our method. Our website is at https://liralab.usc.edu/comparative-language-feedback/

Artificial intelligence plays a crucial role in robotics, with reinforcement learning (RL) emerging as one of the most promising approaches for robot control. However, several key challenges hinder its broader application. First, many RL methods rely on online learning, which requires either real-world hardware or advanced simulation environments--both of which can be costly, time-consuming, and impractical. Offline reinforcement learning offers a solution, enabling models to be trained without ongoing access to physical robots or simulations. A second challenge is learning multi-goal tasks, where robots must achieve multiple objectives simultaneously. This adds complexity to the training process, as the model must generalize across different goals. At the same time, transformer architectures have gained significant popularity across various domains, including reinforcement learning. Yet, no existing methods effectively combine offline training, multi-goal learning, and transformer-based architectures. In this paper, we address these challenges by introducing a novel adaptation of the decision transformer architecture for offline multi-goal reinforcement learning in robotics. Our approach integrates goal-specific information into the decision transformer, allowing it to handle complex tasks in an offline setting. To validate our method, we developed a new offline reinforcement learning dataset using the Panda robotic platform in simulation. Our extensive experiments demonstrate that the decision transformer can outperform state-of-the-art online reinforcement learning methods.

Reinforcement Learning from Human Feedback (RLHF) has emerged as a pivotal tool for aligning large language models (LLMs) with human preferences. Direct Preference Optimization (DPO), one of the most popular approaches, formulates RLHF as a policy optimization problem without explicitly estimating the reward function. It overcomes the stability and efficiency issues of two-step approaches, which typically involve first estimating the reward function and then optimizing the policy via proximal policy optimization (PPO). Since RLHF is essentially an optimization problem, and it is well-known that momentum techniques can accelerate optimization both theoretically and empirically, a natural question arises: Can RLHF be accelerated by momentum? This paper answers this question in the affirmative. In detail, we first show that the iterative preference optimization method can be viewed as a proximal point method. Based on this observation, we propose a general Accelerated Preference Optimization (APO) framework, which unifies many existing preference optimization algorithms and employs Nesterov's momentum technique to speed up the alignment of LLMs. Theoretically, we demonstrate that APO can achieve a faster convergence rate than the standard iterative preference optimization methods, including DPO and Self-Play Preference Optimization (SPPO). Empirically, we show the superiority of APO over DPO, iterative DPO, and other strong baselines for RLHF on the AlpacaEval 2.0 benchmark.