arXiv subjects

Yimeng Liu

Publications and source records attributed to Yimeng Liu.

At least 19 recordsLinked to original sources

arXiv2026

ForceDelta-VLA: Distilling Force-Conditioned Action Corrections for Contact-Rich Manipulation

Force-aware Vision-Language-Action (VLA) policies improve contact-rich manipulation, but typically combine task-level motion and contact-dependent adjustment in a single action prediction. Demonstrations provide no explicit labels for decomposing that prediction into a reusable reference action and a correction. We present ForceDelta-VLA, a correction-distillation framework that constructs an explicit force-correction target using paired predictions from a frozen teacher's force-conditioned and learned force-agnostic modes. A separate delay-correction target accounts for reference-action mismatch and the change in reference state. Training uses asynchronous schedule replay with the cached task context available during execution. The resulting lightweight policy adjusts the reference actions using recent force history and robot state, responding to contact changes between reference-action updates without regenerating complete action chunks. Across nine single-arm and bimanual contact-rich tasks, ForceDelta-VLA achieves an 82.2% mean success rate, compared with 54.4% for the original ForceVLA baseline. Direct execution of our Stage-1 Temporal Teacher achieves 70.6%. Relative to ForceVLA, the complete system reduces mean peak contact force over successful trials by approximately 26% on both platforms.

cs.RO↗

arXiv2026

Mimir: Physics-Grounded LLM Agents for Long-Horizon Irrigation Control

Large language model (LLM) agents increasingly combine reasoning, tool use, and action, but most evidence comes from episodic tasks with relatively immediate feedback and reset failures. Long-running physical control operates in a different regime: actions alter future states, errors compound across decisions, and an agent must improve from experience without being allowed to rewrite the physical rules that make execution safe. We study this regime through irrigation, where daily decisions interact with soil-water dynamics over entire growing seasons. We present Mimir, a physics-grounded LLM agent organized around two repair timescales. At the fast timescale, a structured physical interface and deterministic simulator turn an LLM output into a proposal that we numerically check, revise, and subject to bounded deterministic action selection before execution. At the slow timescale, recurrent failure patterns are consolidated into persistent contextual principles that condition future proposals, while the physical model, evaluator, and execution constraints remain immutable. Under a common retrospective evaluator across multiple sites, crops, and years, Mimir attains the lowest reported aggregate control cost among the evaluated references and uses about 51% less irrigation than the historical schedule replay. The ablation study show higher control cost when forward simulation, verified revision, or persistent context is removed; model-scale and model-family studies show no monotonic gain from increasing LLM size. The resulting lesson show that persistent physical agents can combine semantic reasoning with bounded, evidence-driven self-improvement while reserving physical truth and actuator authority for explicit numerical mechanisms.

cs.AI↗

arXiv2026

SPIRIT: Sparse Physics-Informed Inverse Rendering for SAR Imaging

Radio-frequency (RF) and wave-based imaging systems play a critical role in sensing under adverse conditions, but are fundamentally constrained by a trade-off between collection speed and spatial resolution. Reducing measurements to enable rapid collection of fragment geometric information across sensing dimensions, leading to highly non-injective inverse problems that destabilize conventional image-domain reconstruction and super-resolution methods. Synthetic aperture radar (SAR) exemplifies this challenge in sparse-aperture sampling. We introduce SPIRIT, a physics-informed inverse rendering framework for sparse-aperture SAR. Rather than enhancing formed images, SPIRIT reconstructs an explicit three-dimensional surface. By adopting an optimization strategy and explicitly modeling visibility, phase coherence, and wave propagation, SPIRIT aggregates distributed geometric features across sparse aperture views into a coherent reconstruction. Extensive experiments on synthetic and real-world data demonstrate that SPIRIT recovers accurate geometry and signal under extreme sparsity, reducing data collection by over 90% while achieving low geometric and signal error and consistently outperforming baselines.

cs.CV↗

arXiv2026

Working with Agentic `Teammates': When a New Organizational Actor Collides with the Human Ecosystem of Work

Enterprise AI is transitioning from single-user, reactive tools toward proactive, multi-user 'teammates,' but our empirical understanding of this transition is limited. In this paper, we present an in-situ qualitative study of a persistent, proactive AI agent 'teammate' deployed across multiple teams in a large technology company. Our findings reveal the boundaries of the human-agent workplace are actively in flux, triggering breakdowns and negotiations across: 1) tacit rules of collaborative human workflows, 2) the relational boundaries of this new non-human actor, and 3) the redistribution of trust and human agency. We use these early micro-negotiations as signals to chart a new research, design, and organizational agenda that intentionally preserves human agency in a workplace shared with non-human organizational actors.

cs.HC↗

arXiv2026

Assessing transport resilience to heavy rainfall: An integrated modelling framework with metrics

Road passenger transport faces acute disruption risks from flash flooding from heavy rainfall. To capture system complexity and dynamics for assessing its flood resilience, this study (a) proposes performance metrics to quantify disruption impacts from physical, operational, environmental, social, and economic dimensions, and (b) develops an integrated modelling framework to assess selected metrics by combining rainfall modelling, surface water flood modelling, and agent-based transport modelling to simulate asset-level disruptions and user behavioural responses. This methodology is demonstrated through a 1-in-30-year rainfall event in Tyne and Wear, North East England. Three behavioural response scenarios to rainfall are simulated, including a worst-case scenario, an ideal-response scenario, and an early-warning scenario. Results highlight the potential of behavioural adaptation in enhancing road flood resilience by reducing journey delays, cancellations, and economic losses. The framework and multidimensional performance metrics offer a transferable approach for stress-testing rainfall scenarios and evaluating interventions, supporting evidence-based climate-resilient transport planning.

physics.soc-ph↗

arXiv2026

WiP: Characterizing and Defending Against Mobile-Agent-Driven MFA Automation

Mobile agents automate smartphone tasks by interpreting interfaces, interacting with apps, and coordinating cross-app workflows. This capability challenges the human-mediated separation assumed by passcode-based MFA, creating factor collapse: valid authentication factors are combined within one autonomous environment. Our modular pipeline com- pletes all 10 authorized MFA workflows, compared with 3/10 and 6/10 for two single-agent baselines. We also de- velop a motion-based Android risk signal that distinguishes human- from agent-driven logins with 98.4% table-top and 92.6% hand-held accuracy across 116 pilot sessions. These results demonstrate factor collapse and motivate physical- interaction sensing as complementary evidence of user pres- ence.

cs.NI↗

arXiv2026

LatentChem: From Textual CoT to Latent Thinking in Chemical Reasoning

Current chemical large language models (LLMs) predominantly rely on explicit Chain-of-Thought (CoT) to solve complex reasoning problems. However, forcing nonverbal tacit chemical logic into discrete natural language imposes a fundamental ``modality mismatch,'' creating an artificial bottleneck for reasoning. We introduce LatentChem, a reasoning interface that decouples chemical logic from linguistic generation, enabling the model to process information via continuous thought vectors and dynamic perception. Our investigation reveals a pivotal emergent behavior: spontaneous internalization, defined here as self-selected under outcome-only optimization. When optimized for task success, the model abandons verbose textual derivations in favor of implicit latent computation, suggesting that it identifies the continuous manifold as a more native substrate for chemical logic. This paradigm shift also proves to be a superior computational strategy: LatentChem achieves a 59.88\% non-tie win rate against the strong CoT baseline on the rigorous ChemCoTBench, while delivering a broad 10.84$\times$ average reduction in reasoning step overhead (5.96$\times$ wall-clock speedup) across all evaluated benchmarks. Our results provide empirical evidence that chemical reasoning is more naturally and effectively realized as continuous latent dynamics rather than discretized linguistic trajectories.

physics.chem-ph↗

arXiv2026

OmniPlan: An Adaptive Framework for Timely and Near-Optimal Network Planning Optimization

Network planning optimization is a fundamental problem across diverse domains, including transportation systems, communication networks, and power grids. It requires simultaneous optimization of multiple competing objectives under complex constraints. Existing network planning optimization frameworks rely on mixed integer programming (MIP) solvers, heuristics, and deep reinforcement learning (DRL) models to compute planning decisions. However, they lack effective adaptability to diverse and dynamic user intents, thus leading to the trade-off between execution time and optimality. In this paper, we propose OmniPlan, an adaptive framework that achieves both timeliness and near-optimality in network planning optimization. To achieve the adaptability lacking in existing solutions, OmniPlan employs a large language model (LLM)-based interpreter to convert heterogeneous natural-language intents into a unified and quantifiable user-preference vector. Then it employs a mixture-of-experts architecture that integrates MIP solvers, heuristics, and DRL models as specialized experts, where OmniPlan adapts to diverse intents by dynamically selecting timely and near-optimal experts. Finally, it incorporates a DRL-based expert configuration module that fine-tunes optimization objective weights to align planning decisions with user-specific preferences. We evaluate OmniPlan with a representative real-world workload, i.e., distributed machine learning (ML), where we leverage OmniPlan to offload a wide spectrum of ML inference tasks, e.g., decision trees, SVM, naive Bayes, XGBoost, and random forests, onto a network of hardware devices. Our experiments on a real-world testbed indicate that OmniPlan achieves near-optimal and low-execution-time offloading for real-world ML inference tasks, reducing latency by up to 97.8\% and network device resource consumption by up to 11.5\%.

cs.NI↗

arXiv2026

TaskLens: Generating Task-Conditioned Scaffolded Interfaces for Learning Professional Creative Software

Professional creative software has steep learning curves for novices due to complex interfaces, limited guidance, and unfamiliar terminology. To support educators and tool creators in addressing learner challenges, we introduce TaskLens, an LLM-based method that automatically generates task-conditioned scaffolded UIs from natural language task descriptions. Our method uses LLMs to identify workflow stages and domain concepts, select task-relevant tools, generate implementation code, and execute the code to produce scaffolded interfaces. The interfaces surface relevant tools, organize them by workflow stage, link them to domain concepts, and progressively disclose advanced features. We evaluate TaskLens by deploying two LLM-generated scaffolded interfaces in Blender, a professional 3D modeling software. A user study with beginners (n=32) showed that our scaffolded interfaces significantly reduced perceived task load, improved task performance through embedded workflow guidance, and increased domain concept learning in Blender during task execution. A second study with experts (n=8) showed improved task efficiency and potential to create personalized UIs for productivity and creativity.

cs.HC↗

arXiv2026

AIoT-based Continuous, Contextualized, and Explainable Driving Assessment for Older Adults

The world is undergoing a major demographic shift as older adults become a rapidly growing share of the population, creating new challenges for driving safety. In car-dependent regions such as the United States, driving remains essential for independence, access to services, and social participation. At the same time, aging can introduce gradual changes in vision, attention, reaction time, and driving control that quietly reduce safety. Today's assessment methods rely largely on infrequent clinic visits or simple screening tools, offering only a brief snapshot and failing to reflect how an older adult actually drives on the road. Our work starts from the observation that everyday driving provides a continuous record of functional ability and captures how a driver responds to traffic, navigates complex roads, and manages routine behavior. Leveraging this insight, we propose AURA, an Artificial Intelligence of Things (AIoT) framework for continuous, real-world assessment of driving safety among older adults. AURA integrates richer in-vehicle sensing, multi-scale behavioral modeling, and context-aware analysis to extract detailed indicators of driving performance from routine trips. It organizes fine-grained actions into longer behavioral trajectories and separates age-related performance changes from situational factors such as traffic, road design, or weather. By integrating sensing, modeling, and interpretation within a privacy-preserving edge architecture, AURA provides a foundation for proactive, individualized support that helps older adults drive safely. This paper outlines the design principles, challenges, and research opportunities needed to build reliable, real-world monitoring systems that promote safer aging behind the wheel.

cs.AI↗

arXiv2026

AlignUI: A Method for Designing LLM-Generated UIs Aligned with User Preferences

Designing user interfaces that align with user preferences is a time-consuming process, which requires iterative cycles of prototyping, user testing, and refinement. Recent advancements in LLM-based UI generation have enabled efficient UI generation to assist the UI design process. We introduce AlignUI, a method that aligns LLM-generated UIs with user tasks and preferences by using a user preference dataset to guide the LLM's reasoning process. The dataset was crowdsourced from 50 general users (the target users of generated UIs) and contained 720 UI control preferences on eight image-editing tasks. We evaluated AlignUI by generating UIs for six unseen tasks and conducting a user study with 72 additional general users. The results showed that the generated UIs closely align with multiple dimensions of user preferences. We conclude by discussing the applicability of our method to support user-aligned UI design for multiple task domains and user groups, as well as personalized user needs.

cs.HC↗

arXiv2026

ReUseIt: Synthesizing Reusable AI Agent Workflows for Web Automation

AI-powered web agents have the potential to automate repetitive tasks, such as form filling, information retrieval, and scheduling, but they struggle to reliably execute these tasks without human intervention, requiring users to provide detailed guidance during every run. We address this limitation by automatically synthesizing reusable workflows from an agent's successful and failed attempts. These workflows incorporate execution guards that help agents detect and fix errors while keeping users informed of progress and issues. Our approach enables agents to successfully complete repetitive tasks of the same type with minimal user intervention, increasing the success rates from 24.2% to 70.1% across fifteen tasks. To evaluate this approach, we invited nine users and found that our agent helped them complete web tasks with a higher success rate and less guidance compared to two baseline methods, as well as allowed users to easily monitor agent behavior and understand its failures.

cs.HC↗

arXiv2025

Generalized Jeffreys's approximate objective Bayes factor: Model-selection consistency, finite-sample accuracy, and statistical evidence in 71,126 clinical trial findings

Concerns about the misuse and misinterpretation of p-values and statistical significance have motivated alternatives for quantifying evidence. We define a generalized form of Jeffreys's approximate objective Bayes factor (eJAB), a one-line calculation that is a function of the p-value, sample size, and parameter dimension. We establish conditions under which eJAB is model-selection consistent and verify them for ten statistical tests. We assess finite-sample accuracy by comparing eJAB with Markov chain Monte Carlo computed Bayes factors in 12 simulation studies. We then apply eJAB to 71,126 results from ClinicalTrials.gov (CTG) and find that the proportion of findings with $\text{p-value} \le α$ yet $eJAB_{01}>1$ (favoring the null) closely tracks the significance level $α$, suggesting that such contradictions are pointing to the type I errors. We catalog 4,088 such candidate type I errors and provide details for 131 with reported $\text{p-value} \le 0.01$. We also identify 487 instances of the Jeffreys-Lindley paradox. Finally, we estimate that 75% (6%) of clinical trial plans from CTG set $α\ge 0.05$ as the target evidence threshold, and that 35.5% (0.22%) of results significant at $α=0.05$ correspond to evidence that is no stronger than anecdotal under eJAB.

stat.ME↗

arXiv2025

Hydra: Accurate Multi-Modal Leaf Wetness Sensing with mm-Wave and Camera Fusion

Leaf Wetness Duration (LWD), the time that water remains on leaf surfaces, is crucial in the development of plant diseases. Existing LWD detection lacks standardized measurement techniques, and variations across different plant characteristics limit its effectiveness. Prior research proposes diverse approaches, but they fail to measure real natural leaves directly and lack resilience in various environmental conditions. This reduces the precision and robustness, revealing a notable practical application and effectiveness gap in real-world agricultural settings. This paper presents Hydra, an innovative approach that integrates millimeter-wave (mm-Wave) radar with camera technology to detect leaf wetness by determining if there is water on the leaf. We can measure the time to determine the LWD based on this detection. Firstly, we design a Convolutional Neural Network (CNN) to selectively fuse multiple mm-Wave depth images with an RGB image to generate multiple feature images. Then, we develop a transformer-based encoder to capture the inherent connection among the multiple feature images to generate a feature map, which is further fed to a classifier for detection. Moreover, we augment the dataset during training to generalize our model. Implemented using a frequency-modulated continuous-wave (FMCW) radar within the 76 to 81 GHz band, Hydra's performance is meticulously evaluated on plants, demonstrating the potential to classify leaf wetness with up to 96% accuracy across varying scenarios. Deploying Hydra in the farm, including rainy, dawn, or poorly light nights, it still achieves an accuracy rate of around 90%.

cs.CV↗

arXiv2025

Hydra-Bench: A Benchmark for Multi-Modal Leaf Wetness Sensing

Leaf wetness detection is a crucial task in agricultural monitoring, as it directly impacts the prediction and protection of plant diseases. However, existing sensing systems suffer from limitations in robustness, accuracy, and environmental resilience when applied to natural leaves under dynamic real-world conditions. To address these challenges, we introduce a new multi-modal dataset specifically designed for evaluating and advancing machine learning algorithms in leaf wetness detection. Our dataset comprises synchronized mmWave raw data, Synthetic Aperture Radar (SAR) images, and RGB images collected over six months from five diverse plant species in both controlled and outdoor field environments. We provide detailed benchmarks using the Hydra model, including comparisons against single modality baselines and multiple fusion strategies, as well as performance under varying scan distances. Additionally, our dataset can serve as a benchmark for future SAR imaging algorithm optimization, enabling a systematic evaluation of detection accuracy under diverse conditions.

cs.CV↗

arXiv2025

TR-LLM: Integrating Trajectory Data for Scene-Aware LLM-Based Human Action Prediction

Accurate prediction of human behavior is crucial for AI systems to effectively support real-world applications, such as autonomous robots anticipating and assisting with human tasks. Real-world scenarios frequently present challenges such as occlusions and incomplete scene observations, which can compromise predictive accuracy. Thus, traditional video-based methods often struggle due to limited temporal and spatial perspectives. Large Language Models (LLMs) offer a promising alternative. Having been trained on a large text corpus describing human behaviors, LLMs likely encode plausible sequences of human actions in a home environment. However, LLMs, trained primarily on text data, lack inherent spatial awareness and real-time environmental perception. They struggle with understanding physical constraints and spatial geometry. Therefore, to be effective in a real-world spatial scenario, we propose a multimodal prediction framework that enhances LLM-based action prediction by integrating physical constraints derived from human trajectories. Our experiments demonstrate that combining LLM predictions with trajectory data significantly improves overall prediction performance. This enhancement is particularly notable in situations where the LLM receives limited scene information, highlighting the complementary nature of linguistic knowledge and physical constraints in understanding and anticipating human behavior.

cs.HC↗

arXiv2025

SiCo: An Interactive Size-Controllable Virtual Try-On Approach for Informed Decision-Making

Virtual try-on (VTO) applications aim to replicate the in-store shopping experience and enhance online shopping by enabling users to interact with garments. However, many existing tools adopt a one-size-fits-all approach when visualizing clothing items. This approach limits user interaction with garments, particularly regarding size and fit adjustments, and fails to provide direct insights for size recommendations. As a result, these limitations contribute to high return rates in online shopping. To address this, we introduce SiCo, a new online VTO system that allows users to upload images of themselves and interact with garments by visualizing how different sizes would fit their bodies. Our user study demonstrates that our approach significantly improves users' ability to assess how outfits will appear on their bodies and increases their confidence in selecting clothing sizes that align with their preferences. Based on our evaluation, we believe that SiCo has the potential to reduce return rates and transform the online clothing shopping experience.

cs.HC↗

arXiv2025

Designing Scaffolded Interfaces for Enhanced Learning and Performance in Professional Software

Professional software offers immense power but also presents significant learning challenges. Its complex interfaces, as well as insufficient built-in structured guidance and unfamiliar terminology, often make newcomers struggle with task completion. To address these challenges, we introduce ScaffoldUI, a method for scaffolded interface design to reduce interface complexity, provide structured guidance, and enhance software learnability. The scaffolded interface presents task-relevant tools, progressively discloses tool complexity, and organizes tools based on domain concepts, aiming to assist task performance and software learning. To evaluate the feasibility of our interface design method, we present a technical pipeline for scaffolded interface implementation in professional 3D software, i.e., Blender, and conduct user studies with beginners (N=32) and experts (N=8). Study results demonstrate that our scaffolded interfaces significantly reduce perceived task load caused by interface complexity, support task performance through structured guidance, and augment learning by clearly connecting concepts and tools within the taskflow context. Based on a discussion of the user study findings, we offer insights for future research on designing scaffolded interfaces to support instruction, productivity, creativity, and cross-software workflows.

cs.HC↗