arXiv subjects

Yi Nie

Publications and source records attributed to Yi Nie.

3 recordsLinked to original sources

arXiv2026

Harness VLA: Steering Frozen VLAs into Reliable Manipulation Primitives via Memory-Guided Agents

Language-conditioned manipulation requires both precise contact-rich control and robust reasoning over language, scenes, and long horizons. End-to-end Vision-Language-Action (VLA) models provide strong local visuomotor skills, but they are trained on in-distribution task trajectories and often fail under deployment perturbations such as semantic retargeting, goal re-binding, spatial-layout shifts, and unstable local contacts. LLM coding agents provide complementary semantic and compositional reasoning, but purely analytic primitives struggle with irregular grasping, constrained placement, and articulated-object interaction. We present Harness VLA, a memory-augmented agentic framework that exposes a frozen VLA as a retryable contact-rich primitive and composes it with a small fixed library of analytic primitives for grounding, staging, transport, navigation, and release. Rather than expanding the skill library, the harness learns the operating range of these fixed primitives from task-specific execution traces, global success rules, and failure models. By lifting semantic re-grounding, non-contact execution, and VLA re-staging to the planner while reserving the frozen VLA for local contact-rich phases, Harness VLA extends pretrained VLAs beyond their original trajectory distribution without finetuning. Across perturbed tabletop, household kitchen, and clean-to-randomized bimanual manipulation, Harness VLA improves over the strongest relevant baselines by 38.6 and 25.4 percentage points on LIBERO-Pro and RoboCasa365, respectively, and reaches 58.4% on RoboTwin C2R. Code is available at https://github.com/RLinf/RPent.

cs.RO↗

arXiv2026

Task-Centric Policy Optimization from Misaligned Motion Priors

Humanoid control often leverages motion priors from human demonstrations to encourage natural behaviors. However, such demonstrations are frequently suboptimal or misaligned with robotic tasks due to embodiment differences, retargeting errors, and task-irrelevant variations, causing na\"ive imitation to degrade task performance. Conversely, task-only reinforcement learning admits many task-optimal solutions, often resulting in unnatural or unstable motions. This exposes a fundamental limitation of linear reward mixing in adversarial imitation learning. We propose \emph{Task-Centric Motion Priors} (TCMP), a task-priority adversarial imitation framework that treats imitation as a conditional regularizer rather than a co-equal objective. TCMP maximizes task improvement while incorporating imitation signals only when they are compatible with task progress, yielding an adaptive, geometry-aware update that preserves task-feasible descent and suppresses harmful imitation under misalignment. We provide theoretical analysis of gradient conflict and task-priority stationary points, and validate our claims through humanoid control experiments demonstrating robust task performance with consistent motion style under noisy demonstrations.

cs.RO↗

arXiv2023

FOCUS : A framework for energy system optimization from prosumer to district and city scale

Decarbonizing the energy sector is one of the main challenges to combat the climate crisis. Cities play an important role to reach climate neutrality as more than 70% of global CO2 emissions originate from urban areas. Decarbonization of energy supply systems can be achieved through various means, including the use of renewable energy sources, improving the efficiency of technologies, the coupling of different energy sectors, and the use of flexibility considering individual prosumer behaviour. This leads to an increasingly decentralized energy system, which is challenging to operate in a robust and cost-effective way. The evaluation of technologies and subsystems can only be done from the perspective of the system in which it is embedded and it is highly dependent on their networking and application scenarios. Therefore, the design and operation of energy systems require adequate computation and evaluation tools, which offer a holistic view of all interconnected components. The currently available optimization tools have limitations, such as limited scope of technologies and sectors, high requirements on data, high computational cost, and difficulty in handling multi-objective optimization. To overcome these limitations a software framework called FOCUS for the flexible and dynamic modeling of any urban sector-coupled energy system is developed. The framework includes a library containing models for different technologies and offers a variety of parameter sets for each technology. FOCUS can handle multi-objective problems by returning Pareto-optimal fronts, which helps users to discover the trade-off between criteria and objectives. The developed tool can identify new flexibility potentials in the energy system, actively support companies in the respective field to optimize urban energy system planning solutions, and determine possible threads to the stable operation of such systems.

eess.SY↗