arXiv subjects

Bo Mao

Publications and source records attributed to Bo Mao.

12 recordsLinked to original sources

arXiv2026

Learning While Acting: A Skill-Enhanced Test-Time Co-Evolution Framework for Online Lifelong Learning Agents

Lifelong learning is essential for Large Language Model (LLM) agents operating in dynamic, interactive environments. However, existing lifelong learning agents for long-horizon tasks typically depend on discrete skill or past experiences retrieval with static parameters during inference, which prevents them from continuously internalizing test-time feedback like human learners. To bridge this gap, we propose Skill-enhanced Test-Time Co-Evolution (\texttt{LifeSkill}), a two-stage reinforcement learning framework for Online Lifelong Learning Agents. Specifically, we design Verifier-Guided Skill Learning that addresses the lack of direct supervision for skill extraction by rewarding candidate skills according to the average verifier success of multiple skill-conditioned policy rollouts, encouraging the model to generate skills that are useful for solving tasks rather than merely plausible in text. Furthermore, we introduce Online Skill Internalization, which continuously improves the policy model during test-time interaction by transforming skill-conditioned trajectories into reward signals. This enables the agent to directly internalize reasoning capabilities into its parameters, avoiding the context bloat of experience retrieval. Experiments on LifelongAgentBench show that LifeSkill improves average performance by 7 absolute points by comparing with existing lifelong agent baselines.

cs.LG↗

arXiv2026

GPU Acceleration of TFHE-Based High-Precision Nonlinear Layers for Encrypted LLM Inference

Deploying large language models (LLMs) as cloud services raises privacy concerns as inference may leak sensitive data. Fully Homomorphic Encryption (FHE) allows computation on encrypted data, but current FHE methods struggle with efficient and precise nonlinear function evaluation. Specifically, CKKS-based approaches require high-degree polynomial approximations, which are costly when target precision increases. Alternatively, TFHE's Programmable Bootstrapping (PBS) outperforms CKKS by offering exact lookup-table evaluation. But it lacks high-precision implementations of LLM nonlinear layers and underutilizes GPU resources. We propose \emph{TIGER}, the first GPU-accelerated framework for high-precision TFHE-based nonlinear LLM layer evaluation. TIGER offers: (1) GPU-optimized WoP-PBS method combined with numerical algorithms to surpass native lookup-table precision limits on nonlinear functions; (2) high-precision and efficient implementations of key nonlinear layers, enabling practical encrypted inference; (3) batch-driven design exploiting inter-input parallelism to boost GPU efficiency. TIGER achieves 7.17$\times$, 16.68$\times$, and 17.05$\times$ speedups over a CPU baseline for GELU, Softmax, and LayerNorm, respectively.

cs.CR↗

arXiv2025

Transforming Monolithic Foundation Models into Embodied Multi-Agent Architectures for Human-Robot Collaboration

Foundation models have become central to unifying perception and planning in robotics, yet real-world deployment exposes a mismatch between their monolithic assumption that a single model can handle all cognitive functions and the distributed, dynamic nature of practical service workflows. Vision-language models offer strong semantic understanding but lack embodiment-aware action capabilities while relying on hand-crafted skills. Vision-Language-Action policies enable reactive manipulation but remain brittle across embodiments, weak in geometric grounding, and devoid of proactive collaboration mechanisms. These limitations indicate that scaling a single model alone cannot deliver reliable autonomy for service robots operating in human-populated settings. To address this gap, we present InteractGen, an LLM-powered multi-agent framework that decomposes robot intelligence into specialized agents for continuous perception, dependency-aware planning, decision and verification, failure reflection, and dynamic human delegation, treating foundation models as regulated components within a closed-loop collective. Deployed on a heterogeneous robot team and evaluated in a three-month open-use study, InteractGen improves task success, adaptability, and human-robot collaboration, providing evidence that multi-agent orchestration offers a more feasible path toward socially grounded service autonomy than further scaling standalone models.

cs.RO↗

arXiv2024

A Novel Extensible Simulation Framework for CXL-Enabled Systems

Compute Express Link (CXL) serves as a rising industry standard, delivering high-speed cache-coherent links to a variety of devices, including host CPUs, computational accelerators, and memory devices. It is designed to promote system scalability, enable peer-to-peer exchanges, and accelerate data transmissions. To achieve these objectives, the most recent CXL protocol has brought forth several innovative features, such as port-focused routing, device-handled coherence, and PCIe 6.0 compatibility. However, due to the limited availability of hardware prototypes and simulators compatible with CXL, earlier CXL research has largely depended on emulating CXL devices using remote NUMA nodes. Unfortunately, these NUMA-based emulators have difficulties in accurately representing the new features due to fundamental differences in hardware and protocols. Moreover, the absence of support for non-tree topology and PCIe links makes it complex to merely adapt existing simulators for CXL simulation. To overcome these problems, we introduce ESF, a simulation framework specifically designed for CXL systems. ESF has been developed to accurately reflect the unique features of the latest CXL protocol from the ground up. It uses a specialized interconnect layer to facilitate connections within a wide range of system topologies and also includes key components to carry out specific functions required by these features. By utilizing ESF, we thoroughly investigate various aspects of CXL systems, including system topology, device-handled coherence, and the effects of PCIe characteristics, leading to important findings that can guide the creation of high-performance CXL systems. The ESF source codes are fully open-source and can be accessed at https://anonymous.4open.science/r/ESF-1CE3.

cs.AR↗

arXiv2024

AssistantX: An LLM-Powered Proactive Assistant in Collaborative Human-Populated Environment

Current service robots suffer from limited natural language communication abilities, heavy reliance on predefined commands, ongoing human intervention, and, most notably, a lack of proactive collaboration awareness in human-populated environments. This results in narrow applicability and low utility. In this paper, we introduce AssistantX, an LLM-powered proactive assistant designed for autonomous operation in realworld scenarios with high accuracy. AssistantX employs a multi-agent framework consisting of 4 specialized LLM agents, each dedicated to perception, planning, decision-making, and reflective review, facilitating advanced inference capabilities and comprehensive collaboration awareness, much like a human assistant by your side. We built a dataset of 210 real-world tasks to validate AssistantX, which includes instruction content and status information on whether relevant personnel are available. Extensive experiments were conducted in both text-based simulations and a real office environment over the course of a month and a half. Our experiments demonstrate the effectiveness of the proposed framework, showing that AssistantX can reactively respond to user instructions, actively adjust strategies to adapt to contingencies, and proactively seek assistance from humans to ensure successful task completion. More details and videos can be found at https://assistantx-agent.github.io/AssistantX/.

cs.RO↗

arXiv2023

LearnedFTL: A Learning-Based Page-Level FTL for Reducing Double Reads in Flash-Based SSDs

We present LearnedFTL, a new on-demand page-level flash translation layer (FTL) design, which employs learned indexes to improve the address translation efficiency of flash-based SSDs. The first of its kind, it reduces the number of double reads induced by address translation in random read accesses. LearnedFTL proposes three key techniques: an in-place-update linear model to build learned indexes efficiently, a virtual PPN representation to obtain contiguous PPNs for sorted LPNs, and a group-based allocation and model training via GC/rewrite strategy to reduce the training overhead. By tightly integrating the aforementioned key techniques, LearnedFTL considerably speeds up address translation while reducing the number of flash read accesses caused by the address translation. Our extensive experiments on a FEMU-based prototype show that LearnedFTL can reduce up to 55.5\% address translation-induced double reads. As a result, LearnedFTL reduces the P99 tail latency by 2.9$\times$ $\sim$ 12.2$\times$ with an average of 5.5$\times$ and 8.2$\times$ compared to the state-of-the-art TPFTL and LeaFTL schemes, respectively.

cs.AR↗

arXiv2019

Automated Detecting and Placing Road Objects from Street-level Images

Navigation services utilized by autonomous vehicles or ordinary users require the availability of detailed information about road-related objects and their geolocations, especially at road intersections. However, these road intersections are mainly represented as point elements without detailed information, or are even not available in current versions of crowdsourced mapping databases including OpenStreetMap(OSM). This study develops an approach to automatically detect road objects and place them to right location from street-level images. Our processing pipeline relies on two convolutional neural networks: the first segments the images, while the second detects and classifies the specific objects. Moreover, to locate the detected objects, we establish an attributed topological binary tree(ATBT) based on urban grammar for each image to depict the coherent relations of topologies, attributes and semantics of the road objects. Then the ATBT is further matched with map features on OSM to determine the right placed location. The proposed method has been applied to a case study in Berlin, Germany. We validate the effectiveness of our method on two object classes: traffic signs and traffic lights. Experimental results demonstrate that the proposed approach provides near-precise localization results in terms of completeness and positional accuracy. Among many potential applications, the output may be combined with other sources of data to guide autonomous vehicles

cs.CV↗

arXiv2013

Fabrication of Nb/Al2O3/Nb Josephson Junctions using in situ Magnetron Sputtering and Atomic Layer Deposition

Atomic layer deposition (ALD) provides a promising approach for deposition of ultrathin low-defect-density tunnel barriers, and it has been implemented in a high-vacuum magnetron sputtering system for in situ deposition of ALD-Al2O3 tunnel barriers in superconductor-insulator-superconductor (SIS) Josephson junctions. A smooth ALD-Al2O3 barrier layer was grown on a Al-wetted Nb bottom electrode and was followed with a top Nb electrode growth using sputtering. Preliminary low temperature measurements of current-voltage characteristics (IVC) of the Josephson junctions made from these trilayers confirmed the integrity of the ALD-Al2O3 barrier layer. However, the IcRN product of the junctions is much smaller than the value expected from the Ambegaokar-Baratoff formula suggesting a significant pair-breaking mechanism at the interfaces.

cond-mat.mtrl-sci↗

arXiv2012

Entanglement dynamics of a superconducting phase qubit coupled to a two-level system

We report the observation and quantitative characterization of driven and spontaneous oscillations of quantum entanglement, as measured by concurrence, in a bipartite system consisting of a macroscopic Josephson phase qubit coupled to a microscopic two-level system. The data clearly show the behavior of entanglement dynamics such as sudden death and revival, and the effect of decoherence and ac driving on entanglement.

cond-mat.mes-hall↗

arXiv2011

Landau-Zener-Stückelberg Interference of Microwave Dressed States of a Superconducting Phase Qubit

We present the first observation of Landau-Zener-Stückelberg (LZS) interference of the dressed states arising from an artificial atom, a superconducting phase qubit, interacting with a microwave field. The dependence of LZS interference fringes on various external parameters and the initial state of the qubit agrees quantitatively very well with the theoretical prediction. Such LZS interferometry between the dressed states enables us to control the quantum states of a tetrapartite solid-state system with ease, demonstrating the feasibility of implementing efficient multipartite quantum logic gates with this unique approach.

cond-mat.mes-hall↗

arXiv2010

Tunable Quantum Beam Splitters for Coherent Manipulation of a Solid-State Tripartite Qubit System

Coherent control of quantum states is at the heart of implementing solid-state quantum processors and testing quantum mechanics at the macroscopic level. Despite significant progress made in recent years in controlling single- and bi-partite quantum systems, coherent control of quantum wave function in multipartite systems involving artificial solid-state qubits has been hampered due to the relatively short decoherence time and lacking of precise control methods. Here we report the creation and coherent manipulation of quantum states in a tripartite quantum system, which is formed by a superconducting qubit coupled to two microscopic two-level systems (TLSs). The avoided crossings in the system's energy-level spectrum due to the qubit-TLS interaction act as tunable quantum beam splitters of wave functions. Our result shows that the Landau-Zener-Stückelberg interference has great potential in the precise control of the quantum states in the tripartite system.

cond-mat.supr-con↗

arXiv2010

Quantum Dynamics of a Microwave Driven Superconducting Phase Qubit Coupled to a Two-Level System

We present an analytical and comprehensive description of the quantum dynamics of a microwave resonantly driven superconducting phase qubit coupled to a microscopic two-level system (TLS), covering a wide range of the external microwave field strength. Our model predicts several interesting phenomena in such an ac driven four-level bipartite system including anomalous Rabi oscillations, high-contrast beatings of Rabi oscillations, and extraordinary two-photon transitions. Our experimental results in a coupled qubit-TLS system agree quantitatively very well with the predictions of the theoretical model.

cond-mat.supr-con↗