Sequentially grasping multiple objects with multi-fingered hands is common in daily life, where humans can fully leverage the dexterity of their hands to enclose multiple objects. However, the diversity of object geometries and the complex contact interactions required for high-DOF hands to grasp one object while enclosing another make sequential multi-object grasping challenging for robots. In this paper, we propose SeqMultiGrasp, a system for sequentially grasping objects with a four-fingered Allegro Hand. We focus on sequentially grasping two objects, ensuring that the hand fully encloses one object before lifting it and then grasps the second object without dropping the first. Our system first synthesizes single-object grasp candidates, where each grasp is constrained to use only a subset of the hand's links. These grasps are then validated in a physics simulator to ensure stability and feasibility. Next, we merge the validated single-object grasp poses to construct multi-object grasp configurations. For real-world deployment, we train a diffusion model conditioned on point clouds to propose grasp poses, followed by a heuristic-based execution strategy. We test our system using $8 \times 8$ object combinations in simulation and $6 \times 3$ object combinations in real. Our diffusion-based grasp model obtains an average success rate of 65.8% over 1600 simulation trials and 56.7% over 90 real-world trials, suggesting that it is a promising approach for sequential multi-object grasping with multi-fingered hands. Supplementary material is available on our project website: https://hesic73.github.io/SeqMultiGrasp.

This paper presents a constraint-guided deep learning framework for developing physically consistent health indicators in bearing prognostics and health management. Conventional data-driven methods often lack physical plausibility, while physics-based models are limited by incomplete system knowledge. To address this, we integrate domain knowledge into deep learning using constraints to enforce monotonicity, bound output values between 1 and 0 (representing healthy to failed states), and ensure consistency between signal energy trends and health indicator estimates. This eliminates the need for complex loss term balancing. We implement constraint-guided gradient descent within an autoencoder architecture, creating a constrained autoencoder. However, the framework is adaptable to other architectures. Using time-frequency representations of accelerometer signals from the Pronostia dataset, our constrained model generates smoother, more reliable degradation profiles compared to conventional methods, aligning with expected physical behavior. Performance is assessed using three metrics: trendability, robustness, and consistency. Compared to a conventional baseline, the constrained model improves all three. Another baseline, incorporating monotonicity via a soft-ranking loss function, outperforms in trendability but falls short in robustness and consistency. An ablation study confirms that the monotonicity constraint enhances trendability, the boundary constraint ensures consistency, and the energy-health consistency constraint improves robustness. These findings highlight the effectiveness of constraint-guided deep learning in producing reliable, physically meaningful health indicators, offering a promising direction for future prognostic applications.

Large Language Models have revolutionized natural language processing, yet serving them efficiently in data centers remains challenging due to mixed workloads comprising latency-sensitive (LS) and best-effort (BE) jobs. Existing inference systems employ iteration-level first-come-first-served scheduling, causing head-of-line blocking when BE jobs delay LS jobs. We introduce QLLM, a novel inference system designed for Mixture of Experts (MoE) models, featuring a fine-grained, priority-aware preemptive scheduler. QLLM enables expert-level preemption, deferring BE job execution while minimizing LS time-to-first-token (TTFT). Our approach removes iteration-level scheduling constraints, enabling the scheduler to preempt jobs at any layer based on priority. Evaluations on an Nvidia A100 GPU show that QLLM significantly improves performance. It reduces LS TTFT by an average of $65.5\times$ and meets the SLO at up to $7$ requests/sec, whereas the baseline fails to do so under the tested workload. Additionally, it cuts LS turnaround time by up to $12.8\times$ without impacting throughput. QLLM is modular, extensible, and seamlessly integrates with Hugging Face MoE models.

A long-standing engineering problem, the control of soft robots is difficult because of their highly non-linear, heterogeneous, anisotropic, and distributed nature. Here, bridging engineering and biology, a neural reservoir is employed for the dynamic control of a bio-hybrid model arm made of multiple muscle-tendon groups enveloping an elastic spine. We show how the use of reservoirs facilitates simultaneous control and self-modeling across a set of challenging tasks, outperforming classic neural network approaches. Further, by implementing a spiking reservoir on neuromorphic hardware, energy efficiency is achieved, with nearly two-orders of magnitude improvement relative to standard CPUs, with implications for the on-board control of untethered, small-scale soft robots. Hyper-redundancy, underactuation, distributedness, and continuum in principle can be any dynamical system (31), integrates and mechanics are defining features of soft robots (artificial projects input data streams into a separable, high-dimensional or biological (1-8)), intrinsic to their compliant, elastic constitutive latent space that decomposes non-linear correlations. These traits are attractive in the pursuit of extreme dynamics are then sampled and recombined via linear maps reconfigurability, morphological adaptivity, delicacy and dexterity, into desired computations. Modelbased different tasks while running on the same reservoir, and can be controllers have proven effective in quasi-static settings, matched with specialized hardware (e.g., neuromorphic systems but lack accuracy when inertial effects become significant and for energy efficiency (33, 34)) or'wetware' (neural tissue used typically rely on simplifying assumptions that may overlook as bio-hybrid reservoir (35)).

Building effective and efficient Transformer-based large language models (LLMs) has recently become a research focus, requiring maximizing model language capabilities and minimizing training and deployment costs. Existing efforts have primarily described complex relationships among model performance, parameter size, and data size, as well as searched for the optimal compute allocation to train LLMs. However, they overlook the impacts of context length and attention head configuration (the number of query and key-value heads in grouped-query attention) on training and inference. In this paper, we systematically compare models with different parameter sizes, context lengths, and attention head configurations in terms of model performance, computational cost, and memory cost. Then, we extend the existing scaling methods, which are based solely on parameter size and training compute, to guide the construction of cost-optimal LLMs during both training and inference. Our quantitative scaling studies show that, when processing sufficiently long sequences, a larger model with fewer attention heads can achieve a lower loss while incurring lower computational and memory costs. Our findings provide valuable insights for developing practical LLMs, especially in long-context processing scenarios. We will publicly release our code and data.

The rapid development and large body of literature on machine learning interatomic potentials (MLIPs) can make it difficult to know how to proceed for researchers who are not experts but wish to use these tools. The spirit of this review is to help such researchers by serving as a practical, accessible guide to the state-of-the-art in MLIPs. This review paper covers a broad range of topics related to MLIPs, including (i) central aspects of how and why MLIPs are enablers of many exciting advancements in molecular modeling, (ii) the main underpinnings of different types of MLIPs, including their basic structure and formalism, (iii) the potentially transformative impact of universal MLIPs for both organic and inorganic systems, including an overview of the most recent advances, capabilities, downsides, and potential applications of this nascent class of MLIPs, (iv) a practical guide for estimating and understanding the execution speed of MLIPs, including guidance for users based on hardware availability, type of MLIP used, and prospective simulation size and time, (v) a manual for what MLIP a user should choose for a given application by considering hardware resources, speed requirements, energy and force accuracy requirements, as well as guidance for choosing pre-trained potentials or fitting a new potential from scratch, (vi) discussion around MLIP infrastructure, including sources of training data, pre-trained potentials, and hardware resources for training, (vii) summary of some key limitations of present MLIPs and current approaches to mitigate such limitations, including methods of including long-range interactions, handling magnetic systems, and treatment of excited states, and finally (viii) we finish with some more speculative thoughts on what the future holds for the development and application of MLIPs over the next 3-10+ years.

-- We introduce Vision-Language Attention Distillation (Vi-LAD), a novel approach for distilling socially compliant navigation knowledge from a large Vision-Language Model (VLM) into a lightweight transformer model for real-time robotic navigation. Unlike traditional methods that rely on expert demonstrations or human-annotated datasets, Vi-LAD performs knowledge distillation and fine-tuning at the intermediate layer representation level (i.e., attention maps) by leveraging the backbone of a pre-trained vision-action model. These attention maps highlight key navigational regions in a given scene, which serve as implicit guidance for socially aware motion planning. Vi-LAD fine-tunes a transformer-based model using intermediate attention maps extracted from the pre-trained vision-action model, combined with attention-like semantic maps constructed from a large VLM. T o achieve this, we introduce a novel attention-level distillation loss that fuses knowledge from both sources, generating augmented attention maps with enhanced social awareness. These refined attention maps are then utilized as a traversability costmap within a socially aware model predictive controller (MPC) for navigation. We validate our approach through real-world experiments on a Husky wheeled robot, demonstrating significant improvements over state-of-the-art (SOT A) navigation methods. Our results show up to 14.2% - 50% improvement in success rate, which highlights the effectiveness of Vi-LAD in enabling socially compliant and efficient robot navigation. I NTRODUCTION As autonomous robots become increasingly integrated into human-centered environments, ensuring safe, efficient, and socially compliant navigation is a critical challenge [1].

Algorithmic fairness has received considerable attention due to the failures of various predictive AI systems that have been found to be unfairly biased against subgroups of the population. Many approaches have been proposed to mitigate such biases in predictive systems, however, they often struggle to provide accurate estimates and transparent correction mechanisms in the case where multiple sensitive variables, such as a combination of gender and race, are involved. This paper introduces a new open source Python package, EquiPy, which provides a easy-to-use and model agnostic toolbox for efficiently achieving fairness across multiple sensitive variables. It also offers comprehensive graphic utilities to enable the user to interpret the influence of each sensitive variable within a global context. EquiPy makes use of theoretical results that allow the complexity arising from the use of multiple variables to be broken down into easier-to-solve sub-problems. We demonstrate the ease of use for both mitigation and interpretation on publicly available data derived from the US Census and provide sample code for its use.

We propose a 3D U-Net model to predict the spatial distribution of electromagnetic fields inside a radio-frequency (RF) coil with a subject present, using the phase, amplitude, and position of the coils, along with the density, permittivity, and conductivity of the surrounding medium as inputs. To improve accuracy, we introduce a physics-augmented variant, U-Net Phys, which incorporates Gauss's law of magnetism into the loss function using finite differences. We train our models on electromagnetic field simulations from CST Studio Suite for an eight-channel dipole array RF coil at 7T MRI. Experimental results show that U-Net Phys significantly outperforms the standard U-Net, particularly in predicting fields within the subject, demonstrating the advantage of integrating physical constraints into deep learning-based field prediction.

Predictive models of the future are fundamental for an agent's ability to reason and plan. A common strategy learns a world model and unrolls it step-by-step at inference, where small errors can rapidly compound. Geometric Horizon Models (GHMs) offer a compelling alternative by directly making predictions of future states, avoiding cumulative inference errors. While GHMs can be conveniently learned by a generative analog to temporal difference (TD) learning, existing methods are negatively affected by bootstrapping predictions at train time and struggle to generate high-quality predictions at long horizons. This paper introduces Temporal Difference Flows (TD-Flow), which leverages the structure of a novel Bellman equation on probability paths alongside flow-matching techniques to learn accurate GHMs at over 5x the horizon length of prior methods. Theoretically, we establish a new convergence result and primarily attribute TD-Flow's efficacy to reduced gradient variance during training. We further show that similar arguments can be extended to diffusion-based methods. Empirically, we validate TD-Flow across a diverse set of domains on both generative metrics and downstream tasks including policy evaluation. Moreover, integrating TD-Flow with recent behavior foundation models for planning over pre-trained policies demonstrates substantial performance gains, underscoring its promise for long-horizon decision-making.