arXiv subjects

Fu Chen

Publications and source records attributed to Fu Chen.

9 recordsLinked to original sources

arXiv2026

Zeva: In-Context Causal Learning for Generalizable Embodied Manipulation

Generalizable embodied manipulation remains difficult to achieve through pretraining alone, due to unseen physical conditions in the real world. We argue that robots need to learn from their own physical interactions on the fly during real-world deployment and use this knowledge to inform subsequent actions. We present Zeva, the first framework that enables in-context learning from a robot's own physical interaction experience while keeping the policy model frozen. Zeva employs a Causal Interaction Extractor to encode an executed action and its induced state change into a causal interaction signal, which is stored in a dual-timescale causal memory. For subsequent actions, relevant causal interaction signals are retrieved from memory and injected into the frozen policy model as context. Experiments in simulation and real-world manipulation demonstrate that Zeva achieves the best performance among the compared frontier VLAs and WAMs and, more importantly, enables self-evolution during deployment without gradient updates. Its success rate continues to improve as the robot accumulates interaction experience. Furthermore, the acquired interaction experience can generalize across tasks.

cs.RO↗

arXiv2026

Zetta $\zeta$: An Efficient Closed-Loop Embodied Harness for Self-Evolving Physical Intelligence

Embodied agents are increasingly used to close the gap left by end-to-end policy models. Yet the agentic path has not realized closed-loop learning in physical execution: existing harnesses remain largely open-loop, following fixed skills during rollout and reflecting only after an episode completes. Such post-hoc reflection cannot govern execution as it unfolds, because physical interaction requires decisions to track rapidly changing robot-environment states at a frequency beyond today's large agentic models. We present Zetta, a closed-loop embodied harness that evolves code-based runtime critics and recovery skills online while keeping the base policy frozen. Through three timescale-separated loops, Zetta provides action-frequency governance, rollout-level critic-recovery proposal, and validation-gated skill updates. Together with Z-Infra, a rollout infrastructure decoupling agent logic from heterogeneous execution resources, Zetta achieves state-of-the-art success on LIBERO-Pro and RoboCasa under our current rollout budget, reaching 90.8% and 93.6%, with an 11.1x inference speedup; success continues to scale with self-exploration experience; learned skills transfer zero-shot, and clear robotic "Aha Moments" emerge. These results show that closed-loop harness self-evolution opens a scaling path for reliable physical intelligence.

cs.RO↗

arXiv2025

ToolExpander: Extending the Frontiers of Tool-Using Reinforcement Learning to Weak LLMs

Training Large Language Models (LLMs) with Group Relative Policy Optimization (GRPO) encounters a significant challenge: models often fail to produce accurate responses, particularly in small-scale architectures. This limitation not only diminishes performance improvements and undermines the potential of GRPO but also frequently leads to mid-training collapse, adversely affecting stability and final efficacy. To address these issues, we propose ToolExpander, a novel framework that advances tool-oriented reinforcement learning for resource-constrained LLMs through two key innovations:(1) Dynamic Multi-Round Hard Sampling, which dynamically substitutes challenging samples(those without correct outputs over 10 rollouts) with high-quality few-shot demonstrations during training, coupled with an exponential learning rate decay strategy to mitigate oscillations;(2) Self-Exemplifying Thinking, an enhanced GRPO framework that eliminates KL divergence and incorporates adjusted clipping coefficients, encouraging models to autonomously generate and analyze few-shot examples via a minimal additional reward (0.01).Experimental results demonstrate that ToolExpander significantly enhances tool-using capabilities in LLMs, especially in weaker small-scale models, improving both training stability and overall performance.

cs.CL↗

arXiv2025

Quantum Complex-Valued Self-Attention Model

Self-attention has revolutionized classical machine learning, yet existing quantum self-attention models underutilize quantum states' potential due to oversimplified or incomplete mechanisms. To address this limitation, we introduce the Quantum Complex-Valued Self-Attention Model (QCSAM), the first framework to leverage complex-valued similarities, which captures amplitude and phase relationships between quantum states more comprehensively. To achieve this, QCSAM extends the Linear Combination of Unitaries (LCUs) into the Complex LCUs (CLCUs) framework, enabling precise complex-valued weighting of quantum states and supporting quantum multi-head attention. Experiments on MNIST and Fashion-MNIST show that QCSAM outperforms recent quantum self-attention models, including QKSAN, QSAN, and GQHAN. With only 4 qubits, QCSAM achieves 100% and 99.2% test accuracies on MNIST and Fashion-MNIST, respectively. Furthermore, we evaluate scalability across 3-8 qubits and 2-4 class tasks, while ablation studies validate the advantages of complex-valued attention weights over real-valued alternatives. This work advances quantum machine learning by enhancing the expressiveness and precision of quantum self-attention in a way that aligns with the inherent complexity of quantum mechanics.

quant-ph↗

arXiv2025

VMTS: Vision-Assisted Teacher-Student Reinforcement Learning for Multi-Terrain Locomotion in Bipedal Robots

Bipedal robots, due to their anthropomorphic design, offer substantial potential across various applications, yet their control is hindered by the complexity of their structure. Currently, most research focuses on proprioception-based methods, which lack the capability to overcome complex terrain. While visual perception is vital for operation in human-centric environments, its integration complicates control further. Recent reinforcement learning (RL) approaches have shown promise in enhancing legged robot locomotion, particularly with proprioception-based methods. However, terrain adaptability, especially for bipedal robots, remains a significant challenge, with most research focusing on flat-terrain scenarios. In this paper, we introduce a novel mixture of experts teacher-student network RL strategy, which enhances the performance of teacher-student policies based on visual inputs through a simple yet effective approach. Our method combines terrain selection strategies with the teacher policy, resulting in superior performance compared to traditional models. Additionally, we introduce an alignment loss between the teacher and student networks, rather than enforcing strict similarity, to improve the student's ability to navigate diverse terrains. We validate our approach experimentally on the Limx Dynamic P1 bipedal robot, demonstrating its feasibility and robustness across multiple terrain types.

cs.RO↗

arXiv2024

OTO Planner: An Efficient Only Travelling Once Exploration Planner for Complex and Unknown Environments

Autonomous exploration in complex and cluttered environments is essential for various applications. However, there are many challenges due to the lack of global heuristic information. Existing exploration methods suffer from the repeated paths and considerable computational resource requirement in large-scale environments. To address the above issues, this letter proposes an efficient exploration planner that reduces repeated paths in complex environments, hence it is called "Only Travelling Once Planner". OTO Planner includes fast frontier updating, viewpoint evaluation and viewpoint refinement. A selective frontier updating mechanism is designed, saving a large amount of computational resources. In addition, a novel viewpoint evaluation system is devised to reduce the repeated paths utilizing the enclosed sub-region detection. Besides, a viewpoint refinement approach is raised to concentrate the redundant viewpoints, leading to smoother paths. We conduct extensive simulation and real-world experiments to validate the proposed method. Compared to the state-of-the-art approach, the proposed method reduces the exploration time and movement distance by 10%-20% and improves the speed of frontier detection by 6-9 times.

cs.RO↗

arXiv2024

Quantum Mixed-State Self-Attention Network

Attention mechanisms have revolutionized natural language processing. Combining them with quantum computing aims to further advance this technology. This paper introduces a novel Quantum Mixed-State Self-Attention Network (QMSAN) for natural language processing tasks. Our model leverages quantum computing principles to enhance the effectiveness of self-attention mechanisms. QMSAN uses a quantum attention mechanism based on mixed state, allowing for direct similarity estimation between queries and keys in the quantum domain. This approach leads to more effective attention coefficient calculations. We also propose an innovative quantum positional encoding scheme, implemented through fixed quantum gates within the circuit, improving the model's ability to capture sequence information without additional qubit resources. In numerical experiments of text classification tasks on public datasets, QMSAN outperforms Quantum Self-Attention Neural Network (QSANN). Furthermore, we demonstrate QMSAN's robustness in different quantum noise environments, highlighting its potential for near-term quantum devices.

quant-ph↗

arXiv2023

Improvements on Recommender System based on Mathematical Principles

In this article, we will research the Recommender System's implementation about how it works and the algorithms used. We will explain the Recommender System's algorithms based on mathematical principles, and find feasible methods for improvements. The algorithms based on probability have its significance in Recommender System, we will describe how they help to increase the accuracy and speed of the algorithms. Both the weakness and the strength of two different mathematical distance used to describe the similarity will be detailed illustrated in this article.

cs.IR↗

arXiv2021

Explainable Enterprise Credit Rating via Deep Feature Crossing Network

Due to the powerful learning ability on high-rank and non-linear features, deep neural networks (DNNs) are being applied to data mining and machine learning in various fields, and exhibit higher discrimination performance than conventional methods. However, the applications based on DNNs are rare in enterprise credit rating tasks because most of DNNs employ the "end-to-end" learning paradigm, which outputs the high-rank representations of objects and predictive results without any explanations. Thus, users in the financial industry cannot understand how these high-rank representations are generated, what do they mean and what relations exist with the raw inputs. Then users cannot determine whether the predictions provided by DNNs are reliable, and not trust the predictions providing by such "black box" models. Therefore, in this paper, we propose a novel network to explicitly model the enterprise credit rating problem using DNNs and attention mechanisms. The proposed model realizes explainable enterprise credit ratings. Experimental results obtained on real-world enterprise datasets verify that the proposed approach achieves higher performance than conventional methods, and provides insights into individual rating results and the reliability of model training.

cs.LG↗