Daily Papers Arch&EAI

Imitation learning is a well-established approach for machine-learning-based control. However, its applicability depends on having access to demonstrations, which are often expensive to collect and/or suboptimal for solving the task. In this work, we present GCImOpt, an approach to learn efficient goal-conditioned policies by training on datasets generated by trajectory optimization. Our approach for dataset generation is computationally efficient, can generate thousands of optimal trajectories in minutes on a laptop computer, and produces high-quality demonstrations. Further, by means of a data augmentation scheme that treats intermediate states as goals, we are able to increase the training dataset size by an order of magnitude. Using our generated datasets, we train goal-conditioned neural network policies that can control the system towards arbitrary goals. To demonstrate the generality of our approach, we generate datasets and then train policies for various control tasks, namely cart-pole stabilization, planar and three-dimensional quadcopter stabilization, and point reaching using a 6-DoF robot arm. We show that our trained policies can achieve high success rates and near-optimal control profiles, all while being small (less than 80,000 neural network parameters) and fast enough (up to more than 6,000 times faster than a trajectory optimization solver) that they could be deployed onboard resource-constrained controllers. We provide videos, code, datasets and pre-trained policies under a free software license; see our project website https://jongoiko.github.io/gcimopt/.

Imitation learning is a well-established approach for machine-learning-based control.

Recent advances in Vision-Language-Action (VLA) models have opened new avenues for robot manipulation, yet existing methods exhibit limited efficiency and a lack of high-level knowledge and spatial awareness. To address these challenges, we propose PokeVLA, a lightweight yet powerful foundation model for embodied manipulation that effectively infuses vision-language understanding into action learning. Our framework introduces a two-stage training paradigm: first, we pre-train a compact vision-language model (PokeVLM) on a curated multimodal dataset of 2.4M samples encompassing spatial grounding, affordance, and embodied reasoning tasks; second, we inject manipulation-relevant representations into the action space through multi-view goal-aware semantics learning, geometry alignment, and a novel action expert. Extensive experiments demonstrate state-of-the-art performance on the LIBERO-Plus benchmark and in real-world deployment, outperforming comparable baselines in success rate and robustness under diverse perturbations. To foster reproducibility and community progress, we will open-source our code, model weights, and the scripts for the curated pre-training dataset. Project page: https://getterupper.github.io/PokeVLA

Recent advances in Vision-Language-Action (VLA) models have opened new avenues for robot manipulation, yet existing methods exhibit limited efficiency and a lack of high-level knowledge and spatial awareness.

End-to-end autonomous driving has evolved from the conventional paradigm based on sparse perception into vision-language-action (VLA) models, which focus on learning language descriptions as an auxiliary task to facilitate planning. In this paper, we propose an alternative Vision-Geometry-Action (VGA) paradigm that advocates dense 3D geometry as the critical cue for autonomous driving. As vehicles operate in a 3D world, we think dense 3D geometry provides the most comprehensive information for decision-making. However, most existing geometry reconstruction methods (e.g., DVGT) rely on computationally expensive batch processing of multi-frame inputs and cannot be applied to online planning. To address this, we introduce a streaming Driving Visual Geometry Transformer (DVGT-2), which processes inputs in an online manner and jointly outputs dense geometry and trajectory planning for the current frame. We employ temporal causal attention and cache historical features to support on-the-fly inference. To further enhance efficiency, we propose a sliding-window streaming strategy and use historical caches within a certain interval to avoid repetitive computations. Despite the faster speed, DVGT-2 achieves superior geometry reconstruction performance on various datasets. The same trained DVGT-2 can be directly applied to planning across diverse camera configurations without fine-tuning, including closed-loop NAVSIM and open-loop nuScenes benchmarks.

End-to-end autonomous driving has evolved from the conventional paradigm based on sparse perception into vision-language-action (VLA) models, which focus on learning language descriptions as an auxiliary task to facilitate planning.

Embodied foundation models have achieved significant breakthroughs in robotic manipulation, yet they still depend heavily on large-scale robot demonstrations. Although recent works have explored leveraging human data to alleviate this dependency, effectively extracting transferable knowledge remains a significant challenge due to the inherent embodiment gap between human and robot. We argue that the intention underlying human actions can serve as a powerful intermediate representation for bridging this gap. In this paper, we introduce a novel framework that explicitly learns and transfers human intention to facilitate robotic manipulation. Specifically, we model intention through gaze, as it naturally precedes physical actions and serves as an observable proxy for human intent. Our model is first pretrained on a large-scale egocentric human dataset to capture human intention and its synergy with action, followed by finetuning on a small set of robot and human data. During inference, the model adopts a Chain-of-Thought reasoning paradigm, sequentially predicting intention before executing the action. Extensive evaluations in simulation and real-world settings, across long-horizon and fine-grained tasks, and under few-shot and robustness benchmarks, show that our method consistently outperforms strong baselines, generalizes better, and achieves state-of-the-art performance.

Embodied foundation models have achieved significant breakthroughs in robotic manipulation, yet they still depend heavily on large-scale robot demonstrations.

The real-world deployment of Vision-Language-Action (VLA) models remains limited by the risk of unpredictable and irreversible physical harm. However, we currently lack effective mechanisms to proactively detect these physical safety risks before deployment. To address this gap, we propose \textbf{RedVLA}, the first red teaming framework for physical safety in VLA models. We systematically uncover unsafe behaviors through a two-stage process: (I) \textbf{Risk Scenario Synthesis} constructs a valid and task-feasible initial risk scene. Specifically, it identifies critical interaction regions from benign trajectories and positions the risk factor within these regions, aiming to entangle it with the VLA's execution flow and elicit a target unsafe behavior. (II) \textbf{Risk Amplification} ensures stable elicitation across heterogeneous models. It iteratively refines the risk factor state through gradient-free optimization guided by trajectory features. Experiments on six representative VLA models show that RedVLA uncovers diverse unsafe behaviors and achieves the ASR up to 95.5\% within 10 optimization iterations. To mitigate these risks, we further propose SimpleVLA-Guard, a lightweight safety guard built from RedVLA-generated data. Our data, assets, and code are available \href{https://redvla.github.io}{here}.

The real-world deployment of Vision-Language-Action (VLA) models remains limited by the risk of unpredictable and irreversible physical harm.

Scaling context length is reshaping large-model development, yet full-attention Transformers suffer from prohibitive computation and inference bottlenecks at long sequences. A key challenge is to design foundation models that maintain performance and long-context efficiency with minimal training overhead. We introduce SpikingBrain2.0 (SpB2.0), a 5B model that advances both architecture and training efficiency of its predecessor. Our contributions are two-fold. (1) Architectural Innovation: We propose Dual-Space Sparse Attention (DSSA), an inter-layer hybrid of Sparse Softmax Attention (MoBA) and Sparse Linear Attention (SSE), achieving an improved performance-efficiency trade-off for long-context modeling. SpB2.0 further supports dual quantization paths: INT8-Spiking coding enables sparse event-driven computation, while FP8 coding accelerates inference on modern GPUs. (2) Enhanced Training Strategy: We develop an optimized Transformer-to-Hybrid (T2H) pipeline with dual conversion paths for LLMs and VLMs using curated open-source data. Empirically, SpB2.0-5B and SpB2.0-VL-5B recover most of the base Transformer (Qwen3-4B) capability with under 7k A100 GPU hours. SpB2.0 achieves a 10.13x TTFT speedup at 4M context and supports over 10M tokens on 8 A100 GPUs under vLLM, where full-attention models exceed memory limits. It also demonstrates strong cross-platform compatibility, enabling FP8 GPU inference (2.52x speedup at 250k) and efficient neuromorphic execution (64.31% sparsity, with 70.6% and 46.5% area and power reduction at 500MHz). Overall, SpikingBrain2.0 provides a practical pathway for lightweight, multimodal, spiking foundation models, highlighting the potential of combining brain-inspired mechanisms with efficient architectures for resource-constrained and edge scenarios.

Scaling context length is reshaping large-model development, yet full-attention Transformers suffer from prohibitive computation and inference bottlenecks at long sequences.

Robot swarms navigating through unknown obstacle environments are an emerging research area that faces challenges. Performing tasks in such environments requires swarms to achieve autonomous localization, perception, decision-making, control, and planning. The limited computational resources of onboard platforms present significant challenges for planning and control. Reactive planners offer low computational demands and high re-planning frequencies but lack predictive capabilities, often resulting in local minima. Multi-step planners can make multi-step predictions to reduce deadlocks, but they require substantial computation, resulting in a lower replanning frequency. This paper proposes a novel homotopic trajectory planning framework for a robot swarm that combines centralized homotopic trajectory planning (optimal virtual tube planning) with distributed control, enabling low-computation, high-frequency replanning, thereby uniting the strengths of multi-step and reactive planners. Based on multi-parametric programming, homotopic optimal trajectories are approximated by affine functions. The resulting approximate solutions have computational complexity $O(n_t)$, where $n_t$ is the number of trajectory parameters. This low complexity makes centralized planning of a large number of optimal trajectories practical and, when combined with distributed control, enables rapid, low-cost replanning.} The effectiveness of the proposed method is validated through several simulations and experiments.

Robot swarms navigating through unknown obstacle environments are an emerging research area that faces challenges.

Thanks to the latest advances in learning and robotics, domestic robots are beginning to enter homes, aiming to execute household chores autonomously. However, robots still struggle to perform autonomous manipulation tasks in open-ended environments. In this context, this paper presents a method that enables a robot to manipulate a wide spectrum of articulated objects. In this paper, we automatically generate different robot low-level trajectory primitives to manipulate given object articulations. A very important point when it comes to generating expert trajectories is to consider the diversity of solutions to achieve the same goal. Indeed, knowing diverse low-level primitives to accomplish the same task enables the robot to choose the optimal solution in its real-world environment, with live constraints and unexpected changes. To do so, we propose a method based on Quality-Diversity algorithms that leverages sparse reward exploration in order to generate a set of diverse and high-performing trajectory primitives for a given manipulation task. We validated our method, QDTraj, by generating diverse trajectories in simulation and deploying them in the real world. QDTraj generates at least 5 times more diverse trajectories for both hinge and slider activation tasks, outperforming the other methods we compared against. We assessed the generalization of our method over 30 articulations of the PartNetMobility articulated object dataset, with an average of 704 different trajectories by task. Code is publicly available at: https://kappel.web.isir.upmc.fr/trajectory_primitive_website

Thanks to the latest advances in learning and robotics, domestic robots are beginning to enter homes, aiming to execute household chores autonomously.

Effective contact-rich manipulation requires robots to synergistically leverage vision, force, and proprioception. However, Reinforcement Learning agents struggle to learn in such multisensory settings, especially amidst sensory noise and dynamic changes. We propose MultiSensory Dynamic Pretraining (MSDP), a novel framework for learning expressive multisensory representations tailored for task-oriented policy learning. MSDP is based on masked autoencoding and trains a transformer-based encoder by reconstructing multisensory observations from only a subset of sensor embeddings, leading to cross-modal prediction and sensor fusion. For downstream policy learning, we introduce a novel asymmetric architecture, where a cross-attention mechanism allows the critic to extract dynamic, task-specific features from the frozen embeddings, while the actor receives a stable pooled representation to guide its actions. Our method demonstrates accelerated learning and robust performance under diverse perturbations, including sensor noise, and changes in object dynamics. Evaluations in multiple challenging, contact-rich robot manipulation tasks in simulation and the real world showcase the effectiveness of MSDP. Our approach exhibits strong robustness to perturbations and achieves high success rates on the real robot with as few as 6,000 online interactions, offering a simple yet powerful solution for complex multisensory robotic control. Website: https://msdp-pearl.github.io/

Effective contact-rich manipulation requires robots to synergistically leverage vision, force, and proprioception.

Visual-Language-Action (VLA) models represent a paradigm shift in embodied AI, yet existing frameworks often struggle with imprecise spatial perception, suboptimal multimodal fusion, and instability in reinforcement learning. To bridge these gaps, we propose OmniVLA-RL, a novel architecture that leverages a Mix-of-Transformers (MoT) design to synergistically integrate reasoning, spatial, and action experts. Furthermore, we introduce Flow-GSPO, which reformulates flow matching as a Stochastic Differential Equation (SDE) process and integrates it with Group Segmented Policy Optimization (GSPO) to enhance action precision and training robustness. Extensive evaluations on the LIBERO and LIBERO-Plus benchmarks demonstrate that OmniVLA-RL achieves decent overall performance and surpasses mainstream existing methods, effectively overcoming the fundamental limitations of current VLA models.

Visual-Language-Action (VLA) models represent a paradigm shift in embodied AI, yet existing frameworks often struggle with imprecise spatial perception, suboptimal multimodal fusion, and instability in reinforcement learning.

Household environments present one of the most common, impactful yet challenging application domains for robotics. Within household scenarios, manipulating deformable objects is particularly difficult, both in simulation and real-world execution, due to varied categories and shapes, complex dynamics, and diverse material properties, as well as the lack of reliable deformable-object support in existing simulations. We introduce LeHome, a comprehensive simulation environment designed for deformable object manipulation in household scenarios. LeHome covers a wide spectrum of deformable objects, such as garments and food items, offering high-fidelity dynamics and realistic interactions that existing simulators struggle to simulate accurately. Moreover, LeHome supports multiple robotic embodiments and emphasizes low-cost robots as a core focus, enabling end-to-end evaluation of household tasks on resource-constrained hardware. By bridging the gap between realistic deformable object simulation and practical robotic platforms, LeHome provides a scalable testbed for advancing household robotics. Webpage: https://lehome-web.github.io/ .

Household environments present one of the most common, impactful yet challenging application domains for robotics.

Annotated 3D scene data is scarce and expensive to acquire, while abundant unlabeled videos are readily available on the internet. In this paper, we demonstrate that carefully designed data engines can leverage web-curated, unlabeled videos to automatically generate training data, to facilitate end-to-end models in 3D scene understanding alongside human-annotated datasets. We identify and analyze bottlenecks in automated data generation, revealing critical factors that determine the efficiency and effectiveness of learning from unlabeled data. To validate our approach across different perception granularities, we evaluate on three tasks spanning low-level perception, i.e., 3D object detection and instance segmentation, to high-evel reasoning, i.e., 3D spatial Visual Question Answering (VQA) and Vision-Lanugage Navigation (VLN). Models trained on our generated data demonstrate strong zero-shot performance and show further improvement after finetuning. This demonstrates the viability of leveraging readily available web data as a path toward more capable scene understanding systems.

Annotated 3D scene data is scarce and expensive to acquire, while abundant unlabeled videos are readily available on the internet.

Vision-Language-Action (VLA) models promise generalist robot manipulation, but are typically trained and deployed as short-horizon policies that assume the latest observation is sufficient for action reasoning. This assumption breaks in non-Markovian long-horizon tasks, where task-relevant evidence can be occluded or appear only earlier in the trajectory, and where clutter and distractors make fine-grained visual grounding brittle. We present CodeGraphVLP, a hierarchical framework that enables reliable long-horizon manipulation by combining a persistent semantic-graph state with an executable code-based planner and progress-guided visual-language prompting. The semantic-graph maintains task-relevant entities and relations under partial observability. The synthesized planner executes over this semantic-graph to perform efficient progress checks and outputs a subtask instruction together with subtask-relevant objects. We use these outputs to construct clutter-suppressed observations that focus the VLA executor on critical evidence. On real-world non-Markovian tasks, CodeGraphVLP improves task completion over strong VLA baselines and history-enabled variants while substantially lowering planning latency compared to VLM-in-the-loop planning. We also conduct extensive ablation studies to confirm the contributions of each component.

Vision-Language-Action (VLA) models promise generalist robot manipulation, but are typically trained and deployed as short-horizon policies that assume the latest observation is sufficient for action reasoning.

Classical robot ethics is often framed around obedience, most famously through Asimov's laws. This framing is too narrow for contemporary AI systems, which are increasingly adaptive, generative, embodied, and embedded in physical, psychological, and social worlds. We argue that future human-AI relations should not be understood as master-tool obedience. A better framework is conditional mutualism under governance: a co-evolutionary relationship in which humans and AI systems can develop, specialize, and coordinate, while institutions keep the relationship reciprocal, reversible, psychologically safe, and socially legitimate. We synthesize work from computability, automata theory, statistical machine learning, neural networks, deep learning, transformers, generative and foundation models, world models, embodied AI, alignment, human-robot interaction, ecological mutualism, biological markets, coevolution, and polycentric governance. We then formalize coexistence as a multiplex dynamical system across physical, psychological, and social layers, with reciprocal supply-demand coupling, conflict penalties, developmental freedom, and governance regularization. The framework yields a coexistence model with conditions for existence, uniqueness, and global asymptotic stability of equilibria. It shows that reciprocal complementarity can strengthen stable coexistence, while ungoverned coupling can produce fragility, lock-in, polarization, and domination basins. Human-AI coexistence should therefore be designed as a co-evolutionary governance problem, not as a one-shot obedience problem. This shift supports a scientifically grounded and normatively defensible charter of coexistence: one that permits bounded AI development while preserving human dignity, contestability, collective safety, and fair distribution of gains.

Classical robot ethics is often framed around obedience, most famously through Asimov's laws.

Recent Vision-Language-Action (VLA) models have made impressive progress toward general-purpose robotic manipulation by post-training large Vision-Language Models (VLMs) for action prediction. Yet most VLAs entangle perception and control in a monolithic pipeline optimized purely for action, which can erode language-conditioned grounding. In our real-world tabletop tests, policies over-grasp when the target is absent, are distracted by clutter, and overfit to background appearance. To address these issues, we propose OBEYED-VLA (OBject-centric and gEometrY groundED VLA), a framework that explicitly disentangles perceptual grounding from action reasoning. Instead of operating directly on raw RGB, OBEYED-VLA augments VLAs with a perception module that grounds multi-view inputs into task-conditioned, object-centric, and geometry-aware observations. This module includes a VLM-based object-centric grounding stage that selects task-relevant object regions across camera views, along with a complementary geometric grounding stage that emphasizes the 3D structure of these objects over their appearance. The resulting grounded views are then fed to a pretrained VLA policy, which we fine-tune exclusively on single-object demonstrations collected without environmental clutter or non-target objects. On a real-world UR10e tabletop setup, OBEYED-VLA substantially improves robustness over strong VLA baselines across four challenging regimes and multiple difficulty levels: distractor objects, absent-target rejection, background appearance changes, and cluttered manipulation of unseen objects. Ablation studies confirm that both semantic grounding and geometry-aware grounding are critical to these gains. Overall, the results indicate that making perception an explicit, object-centric component is an effective way to strengthen and generalize VLA-based robotic manipulation.

Recent Vision-Language-Action (VLA) models have made impressive progress toward general-purpose robotic manipulation by post-training large Vision-Language Models (VLMs) for action prediction.

Coordinating the motions of multiple autonomous vehicles (AVs) requires planning frameworks that ensure safety while making efficient use of space and time. This paper presents a new approach, termed variable-time-step spatio-temporal corridor (V-STC), that enhances the temporal efficiency of multi-vehicle coordination. An optimization model is formulated to construct a V-STC for each AV, in which both the spatial configuration of the corridor cubes and their time durations are treated as decision variables. By allowing the corridor's spatial position and time step to vary, the constructed V-STC reduces the overall temporal occupancy of each AV while maintaining collision-free separation in the spatio-temporal domain. Based on the generated V-STC, a dynamically feasible trajectory is then planned independently for each AV. Simulation studies demonstrate that the proposed method achieves safe multi-vehicle coordination and yields more time-efficient motion compared with existing STC approaches.

Coordinating the motions of multiple autonomous vehicles (AVs) requires planning frameworks that ensure safety while making efficient use of space and time.

Robotic foundation models, or generalist robot policies, hold immense potential to enable flexible, general-purpose and dexterous robotic systems. Despite their advancements, our empirical experiments reveal that existing robot policies are prone to learning spurious correlations from pre-training trajectories, adversely affecting their generalization capabilities beyond the training data. To tackle this, we propose a novel Policy Contrastive Decoding (PCD) approach, which redirects the robot policy's focus toward object-relevant visual clues by contrasting action probability distributions derived from original and object-masked visual inputs. As a training-free method, our PCD can be used as a plugin to improve different types of robot policies without needing to finetune or access model weights. We conduct extensive experiments on top of three open-source robot policies, including the autoregressive policy OpenVLA and the diffusion-based policies Octo and $π_0$. The obtained results in both simulation and real-world environments prove PCD's flexibility and effectiveness, e.g., PCD enhances the state-of-the-art policy $π_0$ by 8.9% in the simulation environment and by 108% in the real-world environment. Code and demos are publicly available at: https://koorye.github.io/PCD.

Robotic foundation models, or generalist robot policies, hold immense potential to enable flexible, general-purpose and dexterous robotic systems.

Low-cost FPGA platforms can broaden access to neuromorphic systems research, but current spiking neural network (SNN) workflows remain divided between hardware-first implementations, which are difficult to integrate with PyTorch-style development, and software-first frameworks, which often stop at simulation or GPU execution. This paper presents a semantics-preserving hardware-software co-design framework for the deterministic deployment of PyTorch-defined SNNs to event-driven FPGA execution. A single exported artifact carries weights, thresholds, connectivity descriptors, and grouped time-to-first-spike (TTFS) decoding metadata from software definition to board execution and is reused unchanged by both the software reference and the board runtime. A 10-class MNIST TTFS classifier implemented in the routed 80 MHz design achieves 87.40\% accuracy and matches the software reference on all 10,000 test images. The programmable-logic path delivers a service latency of 0.1375 μs/image and an estimated dynamic energy of 31.6 nJ/image, while scope-aware comparisons with matched GPU and CPU baselines keep accelerator-only and system-level measurements distinct. These results show that low-cost event-driven FPGA hardware can provide a direct and reproducible software-to-board path for software-defined SNN models.

Low-cost FPGA platforms can broaden access to neuromorphic systems research, but current spiking neural network (SNN) workflows remain divided between hardware-first implementations, which are difficult to integrate with PyTorch-style development, and software-first frameworks, which often stop at simulation or GPU execution.

SANDO is a safe trajectory planner for 3D dynamic unknown environments, where obstacle locations and motions are unknown a priori and a collision-free plan can become unsafe at any moment, requiring fast replanning. Existing soft-constraint planners are fast but cannot guarantee collision-free paths, while hard-constraint methods ensure safety at the cost of longer computation. SANDO addresses this trade-off through three contributions. First, a heat map-based A* global planner steers paths away from high-risk regions using soft costs, and a spatiotemporal safe flight corridor (STSFC) generator produces time-layered polytopes that inflate obstacles only by their worst-case reachable set at each time layer, rather than by the worst case over the entire horizon. Second, trajectory optimization is formulated as a Mixed-Integer Quadratic Program (MIQP) with hard collision-avoidance constraints, and a variable elimination technique reduces the number of decision variables, enabling fast computation. Third, a formal safety analysis establishes collision-free guarantees under explicit velocity-bound and estimation-error assumptions. Ablation studies show that variable elimination yields up to 7.4x speedup in optimization time, and that STSFCs are critical for feasibility in dense dynamic environments. Benchmark simulations against state-of-the-art methods across standardized static benchmarks, obstacle-rich static forests, and dynamic environments show that SANDO consistently achieves the highest success rate with no constraint violations across all difficulty levels; perception-only experiments without ground truth obstacle information confirm robust performance under realistic sensing. Hardware experiments on a UAV with fully onboard planning, perception, and localization demonstrate six safe flights in static environments and ten safe flights among dynamic obstacles.

SANDO is a safe trajectory planner for 3D dynamic unknown environments, where obstacle locations and motions are unknown a priori and a collision-free plan can become unsafe at any moment, requiring fast replanning.

Human visual perception offers valuable insights for understanding computational principles of motion-based scene interpretation. Humans robustly detect and segment moving entities that constitute independently moveable chunks of matter, whether observing sparse moving dots, textured surfaces, or naturalistic scenes. In contrast, existing computer vision systems lack a unified approach that works across these diverse settings. Inspired by principles of human perception, we propose a generative model that hierarchically groups low-level motion cues and high-level appearance features into particles (small Gaussians representing local matter), and groups particles into clusters capturing coherently and independently moveable physical entities. We develop a hardware-accelerated inference algorithm based on parallelized block Gibbs sampling to recover stable particle motion and groupings. Our model operates on different kinds of inputs (random dots, stylized textures, or naturalistic RGB video), enabling it to work across settings where biological vision succeeds but existing computer vision approaches do not. We validate this unified framework across three domains: on 2D random dot kinematograms, our approach captures human object perception including graded uncertainty across ambiguous conditions; on a Gestalt-inspired dataset of camouflaged rotating objects, our approach recovers correct 3D structure from motion and thereby accurate 2D object segmentation; and on naturalistic RGB videos, our model tracks the moving 3D matter that makes up deforming objects, enabling robust object-level scene understanding. This work thus establishes a general framework for motion-based perception grounded in principles of human vision.

Human visual perception offers valuable insights for understanding computational principles of motion-based scene interpretation.

Most learning-based approaches to complex physical reasoning sidestep the crucial problem of parameter identification (e.g., mass, friction) that governs scene dynamics, despite its importance in real-world applications such as collision avoidance and robotic manipulation. In this paper, we present LLMPhy, a black-box optimization framework that integrates large language models (LLMs) with physics simulators for physical reasoning. The core insight of LLMPhy is to bridge the textbook physical knowledge embedded in LLMs with the world models implemented in modern physics engines, enabling the construction of digital twins of input scenes via latent parameter estimation. Specifically, LLMPhy decomposes digital twin construction into two subproblems: (i) a continuous problem of estimating physical parameters and (ii) a discrete problem of estimating scene layout. For each subproblem, LLMPhy iteratively prompts the LLM to generate computer programs encoding parameter estimates, executes them in the physics engine to reconstruct the scene, and uses the resulting reconstruction error as feedback to refine the LLM's predictions. As existing physical reasoning benchmarks rarely account for parameter identifiability, we introduce three new datasets designed to evaluate physical reasoning in zero-shot settings. Our results show that LLMPhy achieves state-of-the-art performance on our tasks, recovers physical parameters more accurately, and converges more reliably than prior black-box methods. See the LLMPhy project page for details: https://www.merl.com/research/highlights/LLMPhy

Most learning-based approaches to complex physical reasoning sidestep the crucial problem of parameter identification (e.g., mass, friction) that governs scene dynamics, despite its importance in real-world applications such as collision avoidance and robotic manipulation.

Long-horizon manipulation remains challenging for vision-language-action (VLA) policies: real tasks are multi-step, progress-dependent, and brittle to compounding execution errors. We present LoHo-Manip, a modular framework that scales short-horizon VLA execution to long-horizon instruction following via a dedicated task-management VLM. The manager is decoupled from the executor and is invoked in a receding-horizon manner: given the current observation, it predicts a progress-aware remaining plan that combines (i) a subtask sequence with an explicit done + remaining split as lightweight language memory, and (ii) a visual trace -- a compact 2D keypoint trajectory prompt specifying where to go and what to approach next. The executor VLA is adapted to condition on the rendered trace, thereby turning long-horizon decision-making into repeated local control by following the trace. Crucially, predicting the remaining plan at each step yields an implicit closed loop: failed steps persist in subsequent outputs, and traces update accordingly, enabling automatic continuation and replanning without hand-crafted recovery logic or brittle visual-history buffers. Extensive experiments spanning embodied planning, long-horizon reasoning, trajectory prediction, and end-to-end manipulation in simulation and on a real Franka robot demonstrate strong gains in long-horizon success, robustness, and out-of-distribution generalization. Project page: https://www.liuisabella.com/LoHoManip

Long-horizon manipulation remains challenging for vision-language-action (VLA) policies: real tasks are multi-step, progress-dependent, and brittle to compounding execution errors.

Recently, end-to-end robotic manipulation models have gained significant attention for their generalizability and scalability. However, they often suffer from limited robustness to camera viewpoint changes when training with a fixed camera. In this paper, we propose VistaBot, a novel framework that integrates feed-forward geometric models with video diffusion models to achieve view-robust closed-loop manipulation without requiring camera calibration at test time. Our approach consists of three key components: 4D geometry estimation, view synthesis latent extraction, and latent action learning. VistaBot is integrated into both action-chunking (ACT) and diffusion-based ($π_0$) policies and evaluated across simulation and real-world tasks. We further introduce the View Generalization Score (VGS) as a new metric for comprehensive evaluation of cross-view generalization. Results show that VistaBot improves VGS by 2.79$\times$ and 2.63$\times$ over ACT and $π_0$, respectively, while also achieving high-quality novel view synthesis. Our contributions include a geometry-aware synthesis model, a latent action planner, a new benchmark metric, and extensive validation across diverse environments. The code and models will be made publicly available.

Recently, end-to-end robotic manipulation models have gained significant attention for their generalizability and scalability.

In the emerging field of Fault Tolerant Quantum Computation (FTQC), resource estimation is an important tool for quantitatively comparing prospective architectures, identifying hardware bottlenecks and informing which research paths are most valuable. Despite a recent increase in attention on FTQC, there is currently a lack of resource estimation research for architectures that can realistically offer quantum advantage. In particular, current modelling efforts focus on monolithic quantum computers where all qubits reside on a single device. Constraints on fabrication yield, wiring density, and cooling power make monolithic devices unlikely to scale to fault-tolerant sizes in the foreseeable future. Distributed quantum supercomputers offer a path to overcome these limitations. We propose a prospective distributed quantum computing architecture based on lattice surgery with support for modular and distributed operations, with a focus on superconducting qubits. We develop a resource-estimation framework and software tool tailored to distributed FTQC, enabling end-to-end analysis of practical quantum algorithms on our proposed architecture with various hardware configurations, spanning different node sizes, inter-node entanglement generation rates and distillation protocols. Our extensive benchmarking across eight applications and thousands of hardware configurations, shows that resource estimation driven architecture design is crucial for scalability. We provide concrete design configurations that have feasible resource requirements, recommendations for hardware design and system organization. More broadly, our work provides a rigorous methodology for architectural pathfinding, capable of informing system designs and guiding future research priorities.

In the emerging field of Fault Tolerant Quantum Computation (FTQC), resource estimation is an important tool for quantitatively comparing prospective architectures, identifying hardware bottlenecks and informing which research paths are most valuable.

Robotic scene understanding increasingly relies on Vision-Language Models (VLMs) to generate natural language descriptions of the environment. In this work, we systematically evaluate single-view object captioning for tabletop scenes captured by a robotic manipulator, introducing a controlled physical domain shift that contrasts real-world tools with geometrically similar 3D-printed counterparts that differ in texture, colour, and material. We benchmark a suite of state-of-the-art, locally deployable VLMs across multiple metrics to assess semantic alignment and factual grounding. Our results demonstrate that while VLMs describe common real-world objects effectively, performance degrades markedly on 3D-printed items despite their structurally familiar forms. We further expose critical vulnerabilities in standard evaluation metrics, showing that some fail to detect domain shifts entirely or reward fluent but factually incorrect captions. These findings highlight the limitations of deploying foundation models for embodied agents and the need for more robust architectures and evaluation protocols in physical robotic applications.

Robotic scene understanding increasingly relies on Vision-Language Models (VLMs) to generate natural language descriptions of the environment.

Bayesian Optimal Experimental Design (BOED) provides a rigorous framework for decision-making tasks in which data acquisition is often the critical bottleneck, especially in resource-constrained settings. Traditionally, BOED typically selects designs by maximizing expected information gain (EIG), commonly defined through the Kullback-Leibler (KL) divergence. However, classical evaluation of EIG often involves challenging nested expectations, and even advanced variational methods leave the underlying log-density-ratio objective unchanged. As a result, support mismatch, tail underestimation, and rare-event sensitivity remain intrinsic concerns for KL-based BOED. To address these fundamental bottlenecks, we introduce an IPM-based BOED framework that replaces density-based divergences with integral probability metrics (IPMs), including the Wasserstein distance, Maximum Mean Discrepancy, and Energy Distance, resulting in a highly flexible plug-and-play BOED framework. We establish theoretical guarantees showing that IPM-based utilities provide stronger geometry-aware stability under surrogate-model error and prior misspecification than classical EIG-based utilities. We also validate the proposed framework empirically, demonstrating that IPM-based designs yield highly concentrated credible sets. Furthermore, by extending the same sample-based BOED template in a plug-and-play manner to geometry-aware discrepancies beyond the IPM class, illustrated by a neural optimal transport estimator, we achieve accurate optimal designs in high-dimensional settings where conventional nested Monte Carlo estimators and advanced variational methods fail.

Bayesian Optimal Experimental Design (BOED) provides a rigorous framework for decision-making tasks in which data acquisition is often the critical bottleneck, especially in resource-constrained settings.

Reinforcement learning (RL) is a promising approach for solving robotic manipulation tasks. However, it is challenging to apply the RL algorithms directly in the real world. For one thing, RL is data-intensive and typically requires millions of interactions with environments, which are impractical in real scenarios. For another, it is necessary to make heavy engineering efforts to design reward functions manually. To address these issues, we leverage foundation models in this paper. We propose Reinforcement Learning with Foundation Priors (RLFP) to utilize guidance and feedback from policy, value, and success-reward foundation models. Within this framework, we introduce the Foundation-guided Actor-Critic (FAC) algorithm, which enables embodied agents to explore more efficiently with automatic reward functions. The benefits of our framework are threefold: (1) \textit{sample efficient}; (2) \textit{minimal and effective reward engineering}; (3) \textit{agnostic to foundation model forms and robust to noisy priors}. Our method achieves remarkable performances in various manipulation tasks on both real robots and in simulation. Across 5 dexterous tasks with real robots, FAC achieves an average success rate of 86\% after one hour of real-time learning. Across 8 tasks in the simulated Meta-world, FAC achieves 100\% success rates in 7/8 tasks under less than 100k frames (about 1-hour training), outperforming baseline methods with manual-designed rewards in 1M frames. We believe the RLFP framework can enable future robots to explore and learn autonomously in the physical world for more tasks. Visualizations and code are available at https://yewr.github.io/rlfp.

Reinforcement learning (RL) is a promising approach for solving robotic manipulation tasks.

Dexterous robotic manipulation requires comprehensive perception across all phases of interaction: pre-contact, contact initiation, and post-contact. Such continuous feedback allows a robot to adapt its actions throughout interaction. However, many existing tactile sensors, such as GelSight and its variants, only provide feedback after contact is established, limiting a robot's ability to precisely initiate contact. We introduce FingerEye, a compact and cost-effective sensor that provides continuous vision-tactile feedback throughout the interaction process. FingerEye integrates binocular RGB cameras to provide close-range visual perception with implicit stereo depth. Upon contact, external forces and torques deform a compliant ring structure; these deformations are captured via marker-based pose estimation and serve as a proxy for contact wrench sensing. This design enables a perception stream that smoothly transitions from pre-contact visual cues to post-contact tactile feedback. Building on this sensing capability, we develop a vision-tactile imitation learning policy that fuses signals from multiple FingerEye sensors to learn dexterous manipulation behaviors from limited real-world data. We further develop a digital twin of our sensor and robot platform to improve policy generalization. By combining real demonstrations with visually augmented simulated observations for representation learning, the learned policies become more robust to object appearance variations. Together, these design aspects enable dexterous manipulation across diverse object properties and interaction regimes, including coin standing, chip picking, letter retrieving, and syringe manipulation. The hardware design, code, appendix, and videos are available on our project website: https://nus-lins-lab.github.io/FingerEyeWeb/

Dexterous robotic manipulation requires comprehensive perception across all phases of interaction: pre-contact, contact initiation, and post-contact.

The paradigm shift towards local and on-device inference under stringent resource constraints is represented by the tiny machine learning (TinyML) domain. The primary goal of TinyML is to integrate intelligence into tiny, low-cost devices under strict resource, energy, and latency constraints. However, the ultra-resource-constrained nature of these devices can lead to increased inference execution time, which can be detrimental in latency critical applications. At the same time, TinyML applications are often associated with sensitive data. As such, latency optimization approaches that rely on training samples are infeasible when such data is unavailable, proprietary, or sensitive, highlighting a pressing need for optimization approaches that do not require access to the training dataset and can be applied directly to pre-trained models. Replacing costly multiplications with more hardware-efficient operations, such as shifts and additions, has been proposed as an effective method for reducing inference latency. However, post-training power-of-two (Po2) approaches are scarce and, in many cases, lead to unacceptable accuracy loss. In this work, we propose a framework that applies approximate matrix decomposition to a given CNN in order to optimize hardware implementations subject to strict constraints and without any need of re-training or fine-tuning steps. The genetic algorithm-driven framework explores different matrix decompositions and resulting multiplier-less CNN accelerator designs for FPGA targets. A comprehensive evaluation of different TinyML benchmarks demonstrates our framework's efficacy in generating latency-optimized implementations that satisfy strict accuracy and resource constraints, achieving an average 33% latency improvement with an average accuracy loss of 1.3% compared to typical systolic array-based FPGA accelerators.

The paradigm shift towards local and on-device inference under stringent resource constraints is represented by the tiny machine learning (TinyML) domain.

Denoising of periodic signals and accurate waveform estimation are core tasks across many signal processing domains, including speech, music, medical diagnostics, radio, and sonar. Although deep learning methods have recently shown performance improvements over classical approaches, they require substantial computational resources and are usually trained separately for each signal observation. This study proposes a computationally efficient method based on DCNN and Re-sampling, termed R-DCNN, designed for operation under strict power and resource constraints. The approach targets signals with varying fundamental frequencies and requires only a single observation for training. It generalizes to additional signals via a lightweight resampling step that aligns time scales in signals with different frequencies to re-use the same network weights. Despite its low computational complexity, R-DCNN achieves performance comparable to state-of-the-art classical methods, such as autoregressive (AR)-based techniques, as well as conventional DCNNs trained individually for each observation. This combination of efficiency and performance makes the proposed method particularly well suited for deployment in resource-constrained environments without sacrificing denoising or estimation accuracy.

Denoising of periodic signals and accurate waveform estimation are core tasks across many signal processing domains, including speech, music, medical diagnostics, radio, and sonar.