Daily Papers Arch&EAI

Existing VLA pruning strategies primarily select individual visual tokens according to task-level semantic relevance, while overlooking the spatial information required for robotic manipulation. To examine this limitation, we construct a simple Stride baseline that uniformly samples tokens along the flattened one-dimensional visual sequence, representing a purely geometric pruning strategy. Surprisingly, Stride outperforms semantic pruning and random pruning at certain pruning ratios, but collapses when the token budget is only slightly reduced. We characterize this phenomenon through the spatial coverage radius, defined as the largest spatial blind spot induced by the retained token set after pruning. Our analysis reveals a strong correlation between the spatial structure of retained tokens and task success, suggesting that reliable VLA pruning requires preserving not only task-relevant tokens but also the spatial scaffold of the scene. Motivated by this diagnosis, we propose GeoScaffold, a training-free visual token pruning method that partitions each image into spatial regions, allocates inter-region token budgets using task-relevance weights, and selects intra-region scaffold tokens via farthest point sampling to reduce the local coverage radius. On pi 0.5 and LIBERO, GeoScaffold retains only 20% of visual tokens while preserving a 93.2% average success rate, and achieves a 1.78 times prefill speedup over the unpruned baseline.

Existing VLA pruning strategies primarily select individual visual tokens according to task-level semantic relevance, while overlooking the spatial information required for robotic manipulation.

Traffic signal control (TSC) is essential for improving urban mobility and reducing congestion. Although roadside cameras are widely deployed at signalized intersections and provide rich visual observations of evolving traffic, existing TSC methods typically rely on manually engineered traffic states or separate perception modules, creating a gap between physical observations and control decisions. We present VLALight, the first vision-language-action (VLA) model for end-to-end traffic signal control from multi-view roadside videos. VLALight directly maps visual observations to coordinated signal actions through multi-target spatiotemporal traffic reasoning and topology-aware cooperative perception across intersections. To establish this capability, we develop a two-stage supervised cold-start training strategy for visual traffic understanding and signal decision-making, followed by cooperative agentic reinforcement learning that jointly optimizes local control and network-wide traffic efficiency. Furthermore, VLALight introduces adaptive fast and slow reasoning modes, enabling the policy to allocate deeper reasoning only when additional deliberation provides sufficient control benefits. Through balanced mode-aware rollouts and relative advantage optimization, VLALight learns to trade off decision quality and inference cost. Extensive experiments on seven real-world traffic-flow datasets across three urban networks demonstrate that VLALight consistently outperforms transportation-based, RL-based, and LLM/VLM-based baselines. Ablation studies validate the effectiveness of cooperative perception, network-level optimization, and adaptive reasoning. These results demonstrate the potential of VLA models for real-world physical traffic control. Our project is available at https://github.com/usail-hkust/VLALight.git.

Traffic signal control (TSC) is essential for improving urban mobility and reducing congestion.

Training robot manipulation policies relies on costly robot demonstrations, making large-scale data collection impractical. Meanwhile, to improve policy generalization, existing approaches seek greater diversity in visual observations by varying object placements, viewpoints, and robot configurations during data collection. However, under a limited demonstration budget, this strategy forces the policy to model diverse visual observations, providing insufficient supervision to learn reliable observation-action correspondences under similar local conditions. Our study reveals that policies trained under this strategy achieve lower task success rates than those trained within a compact, visually and kinematically stable region. We refer to these stable regions as comfortable manipulation regions. To exploit this finding, we propose ComManip, a learning paradigm that specializes manipulation policies to comfortable manipulation regions. During inference, ComManip repositions the mobile base until the detected target center enters the familiar image-space range estimated from comfortable-region demonstrations. It then executes the same manipulation policy, enabling effective manipulation across diverse target locations. We conduct extensive experiments across multiple manipulation tasks, demonstration budgets, and policy families including ACT, $π_{0.5}$, RDT, OpenVLA-OFT, and SmolVLA. The results demonstrate that ComManip improves task success by roughly 20 percentage points or more across different policy architectures in large workspaces under limited demonstration budgets, suggesting that specializing manipulation policies to comfortable regions provides a more data-efficient learning paradigm for manipulation.

Training robot manipulation policies relies on costly robot demonstrations, making large-scale data collection impractical.

Aerial manipulators extend robotic manipulation into 3D workspaces that are difficult for ground-based robots to access, creating new opportunities for general-purpose manipulation. However, extending Vision-Language-Action (VLA) models to aerial robots introduces distinct challenges due to the tight coupling between manipulation and flight, continuously changing observations, and safety-critical physical interactions. These challenges demand diverse training data and systematic policy evaluation, yet collecting demonstrations and evaluating policies directly on physical aerial platforms are costly, difficult to scale, and hard to repeat under controlled conditions. We present AeroManip-VLA, a scalable benchmark for aerial VLA data generation and policy evaluation. AeroManip-VLA provides a GPU-accelerated simulation framework with low-level payload-aware flight and manipulation control in massively parallel environments. Building on this framework, we combine reusable reinforcement learning policies with expert task rules to automatically generate demonstrations without human teleoperation across diverse objects, environments, and randomized initial conditions. The generated data include basic skills such as grasping and placing, as well as long-horizon tasks that require both navigation and manipulation. We further introduce automated event labeling and trajectory categorization to filter demonstrations. These mechanisms enable fine-grained analysis of task progress, behavioral outcomes, and safety-related failures. Finally, we evaluate a range of imitation learning and VLA baselines across different task settings, revealing their performance characteristics and failure modes. Together, AeroManip-VLA enables scalable aerial manipulation data generation, structured trajectory analysis, and systematic VLA evaluation in simulation prior to real-world deployment.

Aerial manipulators extend robotic manipulation into 3D workspaces that are difficult for ground-based robots to access, creating new opportunities for general-purpose manipulation.

As LLM-based agents perform increasingly complex tasks, Agent Skills have emerged as a flexible mechanism for extending their capabilities. An Agent Skill packages task-specific instructions with executable components and auxiliary resources to provide specialized functionalities. However, the growing adoption of third-party Skills introduces a new supply-chain attack surface. Malicious Skills can embed harmful behaviors that abuse agent privileges and compromise the agent execution environment or accessible resources. Although recent LLM-based malicious Skill auditing approaches have achieved promising performance, they often rely on capable commercial LLMs. How to achieve effective auditing with compact, locally deployable LLMs in security-sensitive and resource-constrained settings remains largely unexplored. Our investigation reveals that compact LLMs struggle to identify malicious behaviors hidden in complex Skill packages. This difficulty arises from both the implicit nature of such behaviors and the limited reasoning capacity of compact LLMs. To address these challenges, we propose SKILLLITE, an evidence-guided agentic framework for malicious Skill detection. SKILLLITE effectively extracts security-relevant behaviors and infers the intended functionality from complex Skill packages. It then employs a compact LLM to assess the maliciousness of the Skill based on the observed behaviors and their functional context. Experiments show that SKILLLITE improves malicious Skill detection across different compact LLM backbones and outperforms existing representative auditing baselines. Its effectiveness generalizes to behaviorally confirmed in-the-wild malicious Skills. Meanwhile, SKILLLITE maintains a low inference latency, supporting its practical deployment.

As LLM-based agents perform increasingly complex tasks, Agent Skills have emerged as a flexible mechanism for extending their capabilities.

Acoustic information provides rich cues about object location, material properties, and changes caused by contact or motion. This paper introduces a new set of acoustic-aware manipulation tasks for imitation learning, in which robots must use auditory cues to determine manipulation targets. These tasks require sound source localization and identification for active exploration in robotic manipulation. Also, we propose a multimodal imitation learning framework, Spatial-Spectral Audio Action (S2A2), that integrates visual features with acoustic spatial and acoustic signal information for the acoustic-aware manipulation tasks. We implemented S2A2 models that integrates policies such as ACT, Diffusion Policy, VQ-BeT, and $π_0$, into our framework. Simulation experiments showed that the proposed method is the most effective for tasks requiring both position and timbre. Furthermore, real-robot experiments confirm the applicability of the proposed tasks and framework to real-world manipulation.

Acoustic information provides rich cues about object location, material properties, and changes caused by contact or motion.

We present IronLLM-0.6B, a 654M-parameter language model designed for efficient on-device inference. IronLLM-0.6B combines a hybrid attention architecture with X-MTP, a lightweight shared-KV multi-token prediction design that eliminates per-depth KV-cache replay and employs a lightweight verification head for rollback-free drafting, achieving a 1.48x decoding speedup. The model is pretrained on approximately 6.2 trillion tokens using a quality-oriented data pipeline and is further post-trained with Multi-Domain On-Policy Distillation to integrate capabilities from domain-specialized teachers. To better meet the low-latency requirements of on-device scenarios, IronLLM-0.6B adopts an Instruct-Only design. Evaluations show that IronLLM-0.6B achieves competitive performance relative to larger models such as Qwen3.5-0.8B and MiniCPM5-1B, while producing more concise responses on many tasks. We further present IronLLM-0.6B-Light, which replaces RMSNorm with Dynamic Tanh and simplifies several computationally expensive components to improve inference and quantization efficiency. Together, the IronLLM models provide an effective performance-efficiency trade-off for resource-constrained deployment.

We present IronLLM-0.6B, a 654M-parameter language model designed for efficient on-device inference.

Transparent surfaces are ubiquitous in built environments, yet they remain a persistent failure case for robotic perception. RGB cameras perceive the background behind glass rather than the surface itself, while depth sensors such as LiDAR, time-of-flight, and RGB-D often return invalid or background measurements in transparent regions. As a result, systems that rely solely on optical sensing may misinterpret glass walls, doors, or mirrors as free space, compromising safe and reliable navigation. Existing glass segmentation approaches address this by learning visual cues such as reflections, boundaries, and semantic context from RGB images. While effective under favourable lighting and viewing conditions, these cues degrade in low-light environments, under glare, or when glass surfaces are featureless or partially occluded. In this work, we propose a multimodal framework that fuses millimetre-wave radar with RGB-D sensing for real-time transparent surface segmentation. Radar reflects strongly off glass surfaces, providing a geometric cue that remains reliable precisely where vision and depth fail. We exploit this cross-modal inconsistency to generate a radar-guided spatial prior, which is integrated into a lightweight transformer-based segmentation network, GlassFormer, via cross-modal attention. We report results on a mixed-condition test split covering all scene types and a dedicated low-light split designed to stress vision-only methods. GlassFormer achieves 0.88 mIoU on the mixed split, and 0.59 mIoU on the low light split, demonstrating substantial robustness gains over vision-only baselines while maintaining real-time performance on resource-constrained platforms.

Transparent surfaces are ubiquitous in built environments, yet they remain a persistent failure case for robotic perception.

Vision-language-action (VLA) models struggle to reuse recurring interactions in unseen tasks. Our diagnostic study reveals that reliable task completion does not imply consistent execution of constituent atomic actions across task contexts. We present LexiconVLA, a retrievable atomic-action lexicon for cross-task reuse. Global and detail codebooks capture shared interaction structure and fine-grained execution variation, respectively, preserving both reusable patterns and execution details. Visual-Atomic Action Alignment couples trajectory reconstruction from visual state changes with visual outcome prediction from action codes, grounding the lexicon in motion and its effects. We learn these codebooks with trajectory reconstruction and visual alignment on our AtomAction Dataset of 57,803 segments from 69 tasks. A planner and scene-aware adapter translate new goals into code-conditioned subtasks for a shared policy, without skill-specific experts or deployment-time parameter updates. Across five policy backbones on 26 RLBench tasks, LexiconVLA largely maintains performance on 18 seen tasks while improving success on 8 tasks held out from policy training. With BridgeVLA, unseen-task success rises from 16.67% to 34.17% (+17.50 percentage points), and overall success reaches 71.08%, the highest among methods with reported results. Real-robot experiments demonstrate stepwise execution and failure recovery.

Vision-language-action (VLA) models struggle to reuse recurring interactions in unseen tasks.

Modern robot policies predict a chunk of future actions from a single observation, execute only a prefix, and discard the rest before replanning. Choosing the length of this prefix, the execution horizon, poses a trade-off between reactivity and efficiency. A short horizon keeps the policy reactive to the environment, but requires frequent policy calls. Recent test-time methods adaptively select the horizon for each chunk, but they either read model internals, where the signal must be chosen for each architecture, or draw extra samples, which adds cost. We propose Action Upcycling, a training-free algorithm that reuses actions the policy would otherwise discard, without accessing model internals or drawing extra samples. We find that discarded actions stay close to their replanned versions as long as the action velocity remains smooth. Action Upcycling therefore extends the execution horizon up to the point where the velocity begins to fluctuate. Extensive experiments on simulated and real-world manipulation tasks show that Action Upcycling reduces policy calls by 1.2-1.7x with no loss in success rate, across multiple Vision-Language-Action Models (VLAs) and even a World Action Model (WAM). It applies to any chunked policy at negligible cost and is orthogonal to other policy acceleration methods such as few-step sampling and streaming action decoding, opening a new axis for policy acceleration.

Modern robot policies predict a chunk of future actions from a single observation, execute only a prefix, and discard the rest before replanning.

Recent advances in Vision-Language-Action (VLA) models point toward general-purpose robotic intelligence by unifying perception, instruction, and control. Despite impressive progress, existing VLA models often adapt poorly due to \emph{tri-modal misalignment} among vision, language, and action, which weakens action grounding and hurts generalization and fine-tuning efficiency. In this work, we present Alignment-Guided Flow Transformer (AGFT), a novel framework that explicitly enforces tri-modal alignment through a dedicated alignment loss, bridging the representational gap across modalities and enhancing task adaptation. While prior research has predominantly emphasized bi-modal vision--language alignment, we systematically formalize and study tri-modal alignment in VLA models, and provide both ablations and analysis to isolate its role in improving adaptation and robustness. To further accelerate deployment, we adopt a flow-matching objective, enabling substantially fewer inference steps than diffusion-based policies while maintaining accuracy. Theoretically, we establish a quantitative connection between the tri-modal alignment gap and the optimization tightness of flow matching; empirically, experiments on the extensive benchmark show that AGFT achieves superior success rates and lower inference latency compared to SOTA baselines, underscoring tri-modal alignment as a key ingredient for scaling robust VLA manipulation.

Recent advances in Vision-Language-Action (VLA) models point toward general-purpose robotic intelligence by unifying perception, instruction, and control.

Memory-dependent robotic manipulation requires policies to use information that is no longer available in the current observation. Retaining history alone is insufficient: memory must preserve information that supports future actions. One challenge is whether a memory-free foundation model can learn to retain and use historical information from action demonstrations alone, without external memory support. We introduce T$^2$Mem, a framework that develops this capability within a pretrained vision-language-action policy, without external reasoning models or memory-specific annotations. T$^2$Mem uses test-time training to encode observation history into compact fast weights through online self-supervised updates, avoiding repeated processing of the full history. An observation-grounded interface extracts vision-language information for memory formation and supplies retrieved context to the action expert. Action supervision shapes what the memory learns to retain and use, while alternating memory-policy learning gives each component a fixed counterpart during optimization. Across 16 RoboMME tasks, T$^2$Mem improves average success from 17.93% to 56.83% over the memory-free base policy and outperforms the recurrent-memory methods reported in the benchmark, while controlled profiling indicates at least 3x inference speedup over explicit methods. Project website: https://yzliu84.github.io/T2MEM-project/

Memory-dependent robotic manipulation requires policies to use information that is no longer available in the current observation.

Future prediction is increasingly used to improve vision-language-action (VLA) policies, based on the premise that anticipating scene evolution encourages representations useful for control. However, forecast quality alone does not establish that a policy has learned a better representation for action. This distinction matters under distribution shift, where successful control depends on preserving spatial state and likely scene change beyond familiar configurations. We study what determines whether predictive supervision improves the visual representation used by a VLA policy. Through controlled comparisons with matched target constructions, prediction horizons, and training conditions, we find that different prediction interfaces produce markedly different forecasts and visual representations, including in the spatial, dynamics, and action information that transfers beyond familiar scenes. We trace these differences to how predictive errors shape the policy's visual stream. Consistent with this controlled finding, VLA policies trained with more direct, scene-matched future supervision show stronger robustness under simulated and physical distribution shifts. Together, our results frame future prediction as a representation-learning design problem whose value for control depends on whether its supervision reaches the representations through which the policy acts.

Future prediction is increasingly used to improve vision-language-action (VLA) policies, based on the premise that anticipating scene evolution encourages representations useful for control.

Large lookup tables are widely used in hardware to store constant-valued arrays for applications ranging from elementary mathematical operations, such as constant-coefficient multiplication and nonlinear function evaluation, to emerging machine learning models, including table-based neural networks (NNs) and Kolmogorov-Arnold networks (KANs). However, storing extensive tables of constant values can lead to excessive hardware costs in resource-constrained edge devices such as FPGAs. In this paper, we propose CompressedLUT, a lossless compression scheme and its decoder hardware architecture for the efficient storage and retrieval of arbitrary data in hardware. Our method combines decomposition, self-similarities, higher-bit compression, and multilevel compression techniques to maximize table size savings without accuracy loss. Its hardware decoder primarily uses addition, arithmetic right shift, and several small lookup tables, ensuring low area and high throughput. We evaluated CompressedLUT on FPGAs by implementing multiple nonlinear functions, constant-coefficient multipliers (CCMs), and KANs at 12-bit resolution. CompressedLUT is available as an open-source tool.

Large lookup tables are widely used in hardware to store constant-valued arrays for applications ranging from elementary mathematical operations, such as constant-coefficient multiplication and nonlinear function evaluation, to emerging machine learning models, including table-based neural networks (NNs) and Kolmogorov-Arnold networks (KANs).

Understanding ongoing robot manipulation requires models to interpret visual observations in relation to interaction history and task progress. We introduce RoboChrono, a benchmark for streaming task understanding comprising 39 scenarios and 34,713 evaluation instances, constructed from real robot executions and complementary bare-hand human recordings. The benchmark evaluates seven tasks grouped into recognition, alignment, and temporal grounding, covering action understanding and anticipation, visual correspondence, temporal ordering, and action localization. Zero-shot evaluation of 18 vision-language models reveals substantial differences across tasks. GPT-6-Astra achieves 98.3% accuracy on Frame Matching but 68.3% on Frame Ordering, while RynnBrain1.1-122B-A10B exhibits a larger gap, reaching 95.4% and 32.9%, respectively. Input ablations on matched questions with five open-weight models further reveal distinct dependencies on visual evidence: removing visual observations reduces Current Action Recognition accuracy by 22.1 percentage points, whereas Next Action Prediction decreases by only 0.7 points. These findings show that strong visual matching does not consistently coincide with strong temporal ordering, and suggest that next-action prediction can be supported by task and action priors even when visual evidence is unavailable. RoboChrono provides a diagnostic setting for examining these differences, highlighting the need for capability-specific evaluation beyond aggregate scores when assessing task understanding in robot manipulation.

Understanding ongoing robot manipulation requires models to interpret visual observations in relation to interaction history and task progress.

Visual-memory systems commonly retain or compress past observations. Robot control additionally requires interaction-derived state that no individual frame may explicitly represent, such as persistent identity relations, accumulated progress, or ordered procedures. We introduce Simple Agentic Robot Memory (SimpleARM), a training-free memory layer for frozen generalist robot policies. From the task instruction, SimpleARM specifies what to monitor; frozen perceptual tools maintain compact typed state online; structured access retrieves that state only when a proposed subgoal depends on history; and current-view grounding resolves recalled entities before execution. We evaluate SimpleARM on RoboMME, a benchmark of memory-dependent robot manipulation tasks that require history information no longer available in the current observation. Across all 16 tasks and three policy seeds, SimpleARM achieves 67.17% mean success, compared with 44.51% for the strongest non-oracle baseline. Matched ablations show mechanism specificity: removing relation, reference, progress, or route state produces large losses where the affected state is retrieved for control, while largely sparing other tasks. These results support a state-based view of robot memory: effective memory for control is not simply retained visual history, but compact task-relevant state derived from the interaction history.

Visual-memory systems commonly retain or compress past observations.

We study reinforcement learning (RL) methods for cooperative multi-agent Vision-Language-Action (VLA) models. This problem is challenging because VLAs are pretrained on large-scale single-agent data and therefore lack the fine-grained coordination skills required for inter-robot collaboration. Supervised fine-tuning (SFT) on multi-robot demonstrations partially bridges this gap, but its performance is bounded by the demonstration data and cannot improve from its own experience. We present a three-stage reinforced fine-tuning (RFT) pipeline for multi-agent VLAs. First, initialization-aware data collection sweeps over initial configurations and invokes human demonstrations only when the pretrained VLA repeatedly fails, yielding robustness to initialization shift with reduced human cost. Second, offline credit-filtered tuning assigns credit to individual agents and fine-tunes on per-agent trajectories with positive advantage rather than on entire joint rollouts. Third, we find existing online RL for VLAs are less effective for hard multi-agent tasks, which we attribute to noisy co-exploration and unstable updates. We instead use online latent-space fine tuning, which freeze the VLA and perform RL in its latent noise space. We evaluate our multi-agent VLA with both $π_0$ and $π_{0.5}$ backbones across 11 tasks in RoboTwin, RoboFactory and real-world manipulation with two Franka robots. Our multi-agent VLA improves the average success rate by $+23.1\%$, $+16.4\%$, and $+44\%$ on RoboTwin, RoboFactory, and real-world tasks, respectively. Code available at https://anonymous.4open.science/r/mavla_rft-2BC0/.

We study reinforcement learning (RL) methods for cooperative multi-agent Vision-Language-Action (VLA) models.

Robotic manipulation is inherently history-dependent, yet most pretrained robotic policies condition on only the current observation or a short temporal window. Equipping such policies with long-term memory remains challenging: existing approaches either feed the backbone multi-frame observation windows, which substantially increase inference cost, or rely on pre-defined semantic features, which limit task generality and may also require the retraining of the backbone to adapt to the memory. We introduce DRAM (Delta-rule Recurrent Associative Memory), a plug-and-play memory module that can be attached to a wide range of pretrained robotic policies, endowing them with long-horizon memory without architectural modification or backbone retraining, requiring only task-specific post-training of the memory module and action expert. DRAM maintains a fixed-size associative memory using gated delta-rule linear attention, with a modified update that incorporates all tokens within each frame in parallel. An architecture-agnostic readout integrates historical context into action prediction across different policy architectures. Experiments show that DRAM consistently improves frozen pretrained policies over short-context baselines and alternative compact memory designs, validating its effectiveness as a fixed-size, post-hoc memory module trained with the backbone frozen.

Robotic manipulation is inherently history-dependent, yet most pretrained robotic policies condition on only the current observation or a short temporal window.

Long-horizon policies trained with reinforcement learning can still achieve high return through inefficient interactions, while rare efficient behaviors discovered during training may be forgotten. We argue that temporal efficiency itself provides a source of self-supervision for reinforcement learning. We introduce Temporal Self-Imitation Learning (TSIL), a reinforcement learning framework that mines temporally efficient successful trajectories generated during learning and converts them into reusable supervision for future policy improvement. TSIL progressively refines learning using configuration-conditioned adaptive temporal targets derived from fast successful trajectories, while preserving and replaying efficient behaviors through efficiency-weighted self-imitation learning. Across 30 long-horizon tasks spanning robot manipulation and interactive navigation, TSIL consistently improves task success rates, learning efficiency, behavioral efficiency, and robustness to unstable training conditions.

Long-horizon policies trained with reinforcement learning can still achieve high return through inefficient interactions, while rare efficient behaviors discovered during training may be forgotten.

Partial observability remains a fundamental challenge in robotic navigation, where limited sensor coverage and occlusions leave large portions of the environment unobserved. Existing scene completion methods primarily focus on improving incomplete mapping or reconstructing partially observed 3D structures, but rarely investigate how scene completion can be designed to benefit downstream tasks such as path planning. In this work, we propose a path-planning-oriented 3D scene completion framework that moves beyond pure occupancy modeling toward a coupled geometric formulation. Specifically, given partial LiDAR observations as input, the proposed framework jointly predicts completed Truncated Unsigned Distance Field (TUDF)-based continuous geometric representations and voxel-wise occupancy maps. This coupled representation allows the network to better reason about obstacle boundaries and free-space geometry. To fully exploit the synergy between the two representations, we introduce a bidirectionally coupled learning scheme, where TUDF features provide dense geometric guidance to improve occupancy reconstruction, while occupancy features in turn offer complementary structural constraints that refine distance-field estimation. Consequently, the proposed network directly predicts complete occupancy and TUDF representations, allowing seamless integration of TUDF into trajectory planning without post-processing. Extensive experiments on unseen environments demonstrate that the proposed method consistently improves both geometric reconstruction quality and downstream planning performance.

Partial observability remains a fundamental challenge in robotic navigation, where limited sensor coverage and occlusions leave large portions of the environment unobserved.

Generalist robot policies such as vision-language-action models (VLAs) have achieved remarkable generalization, but their inference delays can conflict with the demands of real-time control. Asynchronous execution avoids pauses between action chunks by predicting the next sequence of actions while the robot carries out the previous one. In this paper, we study whether asynchronous execution produces the same action distribution as the original VLA. We find that, for non-Markovian demonstrations, asynchronous execution can produce a fundamentally different action distribution, which can limit the policy's reactivity. In our method, we seek to restore this reactivity by aligning the asynchronously produced action distribution with that of the original VLA through two complementary mechanisms. First, Recursive Flow-Field Distillation trains the asynchronous policy using the VLA's action-generation flow. We characterize the learned distribution theoretically and show experimentally that our asynchronous policy can generate nearly the full range of actions the original VLA would produce, while existing asynchronous methods recover only a fraction of that range. Second, Propose-Resolve prepares multiple action sequences asynchronously and uses the latest observation to select among them based on a lightweight approximation of their likelihood under the VLA's action distribution. Our resulting method matches the original VLA's success on LIBERO and retains about 80% of its success on RoboMimic, about 30 percentage points more than existing asynchronous methods.

Generalist robot policies such as vision-language-action models (VLAs) have achieved remarkable generalization, but their inference delays can conflict with the demands of real-time control.

Robots must often continue a task after a target moves, the viewpoint shifts, or an obstacle appears, even though their earlier observations and committed actions reflect the previous scene. Many simulation robustness benchmarks fix external conditions at reset, leaving this temporal challenge underexamined. We introduce LIBERO-MAX, a benchmark of 8,000 paired cases spanning eight types of changes to geometry, observations, appearance, clutter, and paths. Each pair compares task execution with and without a mid-task event, holding the task, initial state, policy seed, and pre-event action sequence fixed. This controlled comparison distinguishes event-associated regressions from failures already present without the change. Across fourteen current VLA, hybrid, and world-action policies, events reduce success by 11.0-25.7 percentage points. Event profiles reveal shared vulnerabilities to geometry and observation changes, while policy-family rankings interleave. Camera controls show that robustness reflects both competence under the changed conditions and the trajectory from which they are encountered; varying query cadence does not eliminate the gap. Together, the paired protocol and temporal diagnostics establish LIBERO-MAX as a reproducible testbed for diagnosing failures under mid-execution changes and measuring progress toward robot policies that remain effective as the world changes.

Robots must often continue a task after a target moves, the viewpoint shifts, or an obstacle appears, even though their earlier observations and committed actions reflect the previous scene.

Complex robotic manipulation tasks frequently require a long-term memory of past events and actions. As conditioning on full histories renders policies prone to spurious correlations and degrades performance, many approaches to policy memory involve compressing historical information through expensive VLM queries in-the-loop to process only task-salient information. In this paper, we propose an alternative approach in which computationally intensive VLM queries are made during train-time to learn a lightweight latent memory that can be efficiently queried at deployment time. Our representation, which we call the \textbf{workspace token}, is trained by (1) using a VLM to identify current and historical information necessary for completing a task, then (2) distilling these into the workspace token using a set-reconstruction decoder loss. In both simulation and hardware, we show that the workspace token can be used as a drop-in replacement for observations during deployment, enabling policies to solve memory-intensive tasks without the need for VLM reasoning in-the-loop, in effect serving as a \textbf{latent harness} for distilling a stronger reasoning models ability to solve long-horizon tasks to a reactive robotic policy. We further demonstrate that the workspace tokens are not only more lightweight, but also lead to better policy performance compared to conditioning policies on explicit modalities like curated past image frames, motivating a ``latent'' approach to history curation and reasoning model harnesses more broadly.

Complex robotic manipulation tasks frequently require a long-term memory of past events and actions.

Vision-and-language navigation increasingly relies on general-purpose semantic planners, yet translating correct high-level intent into reliable physical execution remains difficult in spatially constrained transitions. Reaching a staircase, doorway, or narrow passage does not ensure traversal; the agent must identify an executable affordance pose and recover from accumulated action errors. We propose PACE (Preference-refined Affordance-Conditioned Execution), a supervised local execution module that augments frozen zero-shot semantic planners for reliable cross-floor navigation. PACE grounds transition-related semantics into a long-horizon, agent-centric traversable affordance pose and conditions short-horizon action generation on this spatial target, thereby aligning semantic goals with physical execution. We further post-train PACE through failure-aware preference refinement using rollout-derived pairs that contrast normal or recovery behaviors with deviation-amplifying behaviors, thereby improving closed-loop correction. We integrate PACE into six open-source zero-shot VLN navigators and demonstrate consistent improvements on the cross-floor subsets of R2R-CE and RxR-CE, increasing the average success rate from 16.35% to 27.65% and from 4.76% to 12.06%, respectively. Real-world experiments further demonstrate PACE's applicability in unseen environments, highlighting the potential of traversable affordances to bridge semantic intent and reliable embodied behavior.

Vision-and-language navigation increasingly relies on general-purpose semantic planners, yet translating correct high-level intent into reliable physical execution remains difficult in spatially constrained transitions.

World-Action Models (WAMs) improve robotic manipulation by conditioning action generation on predicted future observations, but future prediction adds further inference overhead to already expensive iterative action generation. Action chunking can amortize this cost over multiple actions, yet performance degrades over long execution horizons because later actions remain conditioned on stale observations. We introduce STAIRCASE POLICY, a streaming inference and training framework that turns a flow-matching VLA into a JEPA-style WAM and partitions a large action chunk into sub-chunks at staggered denoising stages. Near-term actions are executed as soon as they become available, while later actions continue to be refined. At each sub-chunk boundary, the future latent is re-predicted from the latest observation and used to update all unexecuted actions, enabling long-horizon execution without repeated full policy inference. The resulting future-prediction error can further serve as a signal for adaptive chunking. S-WAM achieves 97.7% on LIBERO and 87.9% on LIBERO-Plus, and improves performance across multiple policy backbones and real-robot tasks. It reaches 292.7 executed actions per second, $3.62\times$ the throughput of conventional execution at comparable accuracy, while reducing time-to-first-action from 123.6 to 73.3 ms. With additional inference optimizations, throughput further increases to 642.9 actions per second.

World-Action Models (WAMs) improve robotic manipulation by conditioning action generation on predicted future observations, but future prediction adds further inference overhead to already expensive iterative action generation.

Robots operating in real-world environments must execute complex, multi-step bimanual tasks over long horizons rather than single, isolated actions. Current manipulation datasets struggle to support this capability: although single-arm datasets reach hundreds of thousands of trajectories, they typically provide only one high-level instruction per episode while the rare bimanual effort that does label subtasks annotates only a fraction of its hours. We present FineART, a densely annotated bimanual manipulation dataset of 40,543 episodes, 1,718 hours, and 533,913 subtasks across 151 tasks. We also introduce FineART-VLA, a vision-language-action policy that predicts its own next subtask, and show that mid-training it this way yields substantial gains. Specifically, success on a spatial disambiguation task increases from 32.0% to 100.0%, and step-by-step human subtask guidance lifts success on an unseen long-horizon task from 16.0% to 76.0%. Furthermore, after minimal fine-tuning on a new robot, the policy requires one-tenth the data of baselines without mid-training and generalizes zero-shot to completely unseen tasks on the new hardware. We open-source the full dataset, model weights, and training code.

Robots operating in real-world environments must execute complex, multi-step bimanual tasks over long horizons rather than single, isolated actions.

A pretrained video world model admits many plausible futures for a scene, but a robot must realize the exact task-conditioned one. To turn world models into executable robot policies, existing methods fine-tune the heavy world model backbone using large-scale robot data and computational resources. Challenging this status quo, we argue that the expensive part has already been paid in the world model pretraining since the representation space of a video world model lays out the diverse potential futures. In this case, what remains is to select the future that accomplishes the task and to read out the actions that realize it. We formalize this task as actualization, which learns a task-conditioned selection and realization on top of a prior supplied by a frozen world model. This can be solved by a tiny actualizer model. We implement RoboActualizer with as few as 60M parameters on top of a frozen world model encoder. The actualizer is composed of two lightweight DiT experts that jointly predict future latents and actions by flow matching. The model can be trained entirely on a single GPU with 32 GB peak memory. With up to 100x fewer trainable parameters than existing WAMs and VLAs, RoboActualizer reaches great performance on simulation benchmarks including LIBERO, LIBERO-Plus, RoboTwin 2.0 and five tasks on two real-world platforms, with a low latency of 39 ms that allows real-time control.

A pretrained video world model admits many plausible futures for a scene, but a robot must realize the exact task-conditioned one.

Domain adaptation of small language models (SLMs) has emerged as a practical strategy for deploying capable NLP systems in resource-constrained, high-stakes environments including healthcare, legal services, and financial analysis. While performance gains from parameter-efficient fine-tuning are well characterised, the corresponding impact on trustworthiness (factual calibration and adversarial robustness) remains poorly understood. This paper presents the first systematic cross-domain, cross-architecture empirical study quantifying the trustworthiness cost of domain adaptation across three SLM architectures (TinyLlama 1B, Gemma-2 2B, Llama 3.2 1B), three domains (healthcare, legal, finance), two training-data conditions (benign and adversarially perturbed), and four fine-tuning strategies (baseline LoRA, Safety-DPO, Dark Experience Replay, and Task Arithmetic LoRA, TA-LoRA). Trustworthiness is evaluated through TruthfulQA MC2 (factual calibration) and HarmBench ASR (adversarial robustness) across all 216 experimental configurations with three random seeds. Three principal findings emerge. First, baseline QLoRA domain adaptation produces minimal TruthfulQA MC2 change across all model-domain combinations (mean |Delta TQA| < 0.02). Second, adversarially perturbed training data consistently improves domain adaptation quality (Delta loss approximately -0.040) without worsening trustworthiness benchmarks. Third, none of the three safety-preserving strategies reduced adversarial harm susceptibility: Safety-DPO was effectively neutral (mean Delta ASR < 0.001), while Dark ER and TA-LoRA increased mean HarmBench ASR by +0.171 and +0.155 respectively in safety-aligned models (Gemma-2 2B, Llama 3.2 1B), with individual configurations exceeding +0.45. These results challenge the assumption that replay-based and arithmetic-merge strategies transfer alignment to domain-adapted SLMs.

Domain adaptation of small language models (SLMs) has emerged as a practical strategy for deploying capable NLP systems in resource-constrained, high-stakes environments including healthcare, legal services, and financial analysis.

Vision-language-action (VLA) models perform well on shorter-horizon manipulation tasks but still struggle with long-horizon tasks that require multiple dependent manipulations from a single command. Online reinforcement learning (RL) can improve these policies through environment interaction, yet many existing methods provide reward only after the complete task succeeds. However, such terminal supervision is sparse and does not distinguish early failures from rollouts that make substantial partial progress. We propose StructRL, an online RL framework that constructs structured intermediate supervision from verifiable subtask completions. StructRL decomposes each task into verifiable subtasks, grants intermediate rewards only after the prerequisite subtasks have been completed, and scales each reward according to completion pace. Across RoboCasa365 and LIBERO-Long with GR00T-N1.5 and pi 0.5, StructRL consistently outperforms evaluated online RL baselines. These results show that verifiable, structured intermediate rewards improve long-horizon VLA post-training. Code is available at https://github.com/amazon-science/StructRL.

Vision-language-action (VLA) models perform well on shorter-horizon manipulation tasks but still struggle with long-horizon tasks that require multiple dependent manipulations from a single command.

Robot manipulation policies are usually trained under the assumption that a commanded action produces the same motion as it did during training even after hours of operation. Real hardware violates this assumption as the motors gradually heat up, current saturates near contact, voltage sags under load, thus the same policy action can produce a weaker, delayed, or noisier motion. These conditions are already measured by onboard telemetry, such as joint temperature, motor current, and supply voltage, yet this signal is typically used only for logging or safety checks rather than policy adaptation. We introduce Telemetry-Aware Action Rectification (TeAR), a policy-agnostic method that turns a frozen manipulation policy into a telemetry-conditioned policy by rectifying its outgoing action before it reaches the low-level controller. TeAR learns a lightweight Transformer that combines the proposed action with live actuator telemetry and amplifies, damps, or biases individual action components. We evaluate TeAR across 18 policy-task pairs spanning 8 policy families and 5 manipulation tasks. In an additional paired evaluation with degradation-model mismatch, TeAR achieves 31.8% success, compared with 25.6% for the base policy and 30.6% for an assumed-model inverse. On a physical arm, TeAR improves success under heating by 10-15% without on-robot fine-tuning.

Robot manipulation policies are usually trained under the assumption that a commanded action produces the same motion as it did during training even after hours of operation.