arXiv subjects

Tong Zhang

Publications and source records attributed to Tong Zhang.

At least 199 records · Page 11Linked to original sources

arXiv2025

Cross-Domain Few-Shot Learning via Multi-View Collaborative Optimization with Vision-Language Models

Vision-language models (VLMs) pre-trained on natural image and language data, such as CLIP, have exhibited significant potential in few-shot image recognition tasks, leading to development of various efficient transfer learning methods. These methods exploit inherent pre-learned knowledge in VLMs and have achieved strong performance on standard image datasets. However, their effectiveness is often limited when confronted with cross-domain tasks where imaging domains differ from natural images. To address this limitation, we propose Consistency-guided Multi-view Collaborative Optimization (CoMuCo), a novel fine-tuning strategy for VLMs. This strategy employs two functionally complementary expert modules to extract multi-view features, while incorporating prior knowledge-based consistency constraints and information geometry-based consensus mechanisms to enhance the robustness of feature learning. Additionally, a new cross-domain few-shot benchmark is established to help comprehensively evaluate methods on imaging domains distinct from natural images. Extensive empirical evaluations on both existing and newly proposed benchmarks suggest CoMuCo consistently outperforms current methods in few-shot tasks. The code and benchmark will be released.

cs.CV↗

arXiv2025

Preserve and Sculpt: Manifold-Aligned Fine-tuning of Vision-Language Models for Few-Shot Learning

Pretrained vision-language models (VLMs), such as CLIP, have shown remarkable potential in few-shot image classification and led to numerous effective transfer learning strategies. These methods leverage the pretrained knowledge of VLMs to enable effective domain adaptation while mitigating overfitting through parameter-efficient tuning or instance-based consistency constraints. However, such regularizations often neglect the geometric structure of data distribution, which may lead to distortion of the overall semantic representation. To overcome this limitation, we propose a novel fine-tuning method, Manifold-Preserving and Sculpting Tuning (MPS-Tuning). Regarding the data distribution in feature space as a semantic manifold, MPS-Tuning explicitly constrains the intrinsic geometry of this manifold while further sculpting it to enhance class separability. Specifically, MPS-Tuning preserves both macroscopic and microscopic topological structures of the original manifold by aligning Gram matrices of features before and after fine-tuning. Theoretically, this constraint is shown to approximate an upper bound of the Gromov-Wasserstein distance. Furthermore, features from the image and text modalities are paired, and pairwise similarities are optimized to enhance the manifold's class discriminability. Extensive experiments demonstrate that MPS-Tuning significantly improves model performance while effectively preserving the structure of the semantic manifold. The code will be released.

cs.CV↗

arXiv2025

An Efficient and Adaptive Framework for Achieving Underwater High-performance Maintenance Networks

With the development of space-air-ground-aqua integrated networks (SAGAIN), high-speed and reliable network services are accessible at any time and any location. However, the long propagation delay and limited network capacity of underwater communication networks (UCN) negatively impact the service quality of SAGAIN. To address this issue, this paper presents U-HPNF, a hierarchical framework designed to achieve a high-performance network with self-management, self-configuration, and self-optimization capabilities. U-HPNF leverages the sensing and decision-making capabilities of deep reinforcement learning (DRL) to manage limited resources in UCNs, including communication bandwidth, computational resources, and energy supplies. Additionally, we incorporate federated learning (FL) to iteratively optimize the decision-making model, thereby reducing communication overhead and protecting the privacy of node observation information. By deploying digital twins (DT) at both the intelligent sink layer and aggregation layer, U-HPNF can mimic numerous network scenarios and adapt to varying network QoS requirements. Through a three-tier network design with two-levels DT, U-HPNF provides an AI-native high-performance underwater network. Numerical results demonstrate that the proposed U-HPNF framework can effectively optimize network performance across various situations and adapt to changing QoS requirements.

cs.NI↗

arXiv2025

HuB: Learning Extreme Humanoid Balance

The human body demonstrates exceptional motor capabilities-such as standing steadily on one foot or performing a high kick with the leg raised over 1.5 meters-both requiring precise balance control. While recent research on humanoid control has leveraged reinforcement learning to track human motions for skill acquisition, applying this paradigm to balance-intensive tasks remains challenging. In this work, we identify three key obstacles: instability from reference motion errors, learning difficulties due to morphological mismatch, and the sim-to-real gap caused by sensor noise and unmodeled dynamics. To address these challenges, we propose HuB (Humanoid Balance), a unified framework that integrates reference motion refinement, balance-aware policy learning, and sim-to-real robustness training, with each component targeting a specific challenge. We validate our approach on the Unitree G1 humanoid robot across challenging quasi-static balance tasks, including extreme single-legged poses such as Swallow Balance and Bruce Lee's Kick. Our policy remains stable even under strong physical disturbances-such as a forceful soccer strike-while baseline methods consistently fail to complete these tasks. Project website: https://hub-robot.github.io

cs.RO↗

arXiv2025

Embodied Edge Intelligence Meets Near Field Communication: Concept, Design, and Verification

Realizing embodied artificial intelligence is challenging due to the huge computation demands of large models (LMs). To support LMs while ensuring real-time inference, embodied edge intelligence (EEI) is a promising paradigm, which leverages an LM edge to provide computing powers in close proximity to embodied robots. Due to embodied data exchange, EEI requires higher spectral efficiency, enhanced communication security, and reduced inter-user interference. To meet these requirements, near-field communication (NFC), which leverages extremely large antenna arrays as its hardware foundation, is an ideal solution. Therefore, this paper advocates the integration of EEI and NFC, resulting in a near-field EEI (NEEI) paradigm. However, NEEI also introduces new challenges that cannot be adequately addressed by isolated EEI or NFC designs, creating research opportunities for joint optimization of both functionalities. To this end, we propose radio-friendly embodied planning for EEI-assisted NFC scenarios and view-guided beam-focusing for NFC-assisted EEI scenarios. We also elaborate how to realize resource-efficient NEEI through opportunistic collaborative navigation. Experimental results are provided to confirm the superiority of the proposed techniques compared with various benchmarks.

cs.RO↗

arXiv2025

P/D-Device: Disaggregated Large Language Model between Cloud and Devices

Serving disaggregated large language models has been widely adopted in industrial practice for enhanced performance. However, too many tokens generated in decoding phase, i.e., occupying the resources for a long time, essentially hamper the cloud from achieving a higher throughput. Meanwhile, due to limited on-device resources, the time to first token (TTFT), i.e., the latency of prefill phase, increases dramatically with the growth on prompt length. In order to concur with such a bottleneck on resources, i.e., long occupation in cloud and limited on-device computing capacity, we propose to separate large language model between cloud and devices. That is, the cloud helps a portion of the content for each device, only in its prefill phase. Specifically, after receiving the first token from the cloud, decoupling with its own prefill, the device responds to the user immediately for a lower TTFT. Then, the following tokens from cloud are presented via a speed controller for smoothed TPOT (the time per output token), until the device catches up with the progress. On-device prefill is then amortized using received tokens while the resource usage in cloud is controlled. Moreover, during cloud prefill, the prompt can be refined, using those intermediate data already generated, to further speed up on-device inference. We implement such a scheme P/D-Device, and confirm its superiority over other alternatives. We further propose an algorithm to decide the best settings. Real-trace experiments show that TTFT decreases at least 60%, maximum TPOT is about tens of milliseconds, and cloud throughput increases by up to 15x.

cs.DC↗

arXiv2025

Traffic Load-Aware Resource Management Strategy for Underwater Wireless Sensor Networks

Underwater Wireless Sensor Networks (UWSNs) represent a promising technology that enables diverse underwater applications through acoustic communication. However, it encounters significant challenges including harsh communication environments, limited energy supply, and restricted signal transmission. This paper aims to provide efficient and reliable communication in underwater networks with limited energy and communication resources by optimizing the scheduling of communication links and adjusting transmission parameters (e.g., transmit power and transmission rate). The efficient and reliable communication multi-objective optimization problem (ERCMOP) is formulated as a decentralized partially observable Markov decision process (Dec-POMDP). A Traffic Load-Aware Resource Management (TARM) strategy based on deep multi-agent reinforcement learning (MARL) is presented to address this problem. Specifically, a traffic load-aware mechanism that leverages the overhear information from neighboring nodes is designed to mitigate the disparity between partial observations and global states. Moreover, by incorporating a solution space optimization algorithm, the number of candidate solutions for the deep MARL-based decision-making model can be effectively reduced, thereby optimizing the computational complexity. Simulation results demonstrate the adaptability of TARM in various scenarios with different transmission demands and collision probabilities, while also validating the effectiveness of the proposed approach in supporting efficient and reliable communication in underwater networks with limited resources.

cs.NI↗

arXiv2025

Achieving Fair-Effective Communications and Robustness in Underwater Acoustic Sensor Networks: A Semi-Cooperative Approach

This paper investigates the fair-effective communication and robustness in imperfect and energy-constrained underwater acoustic sensor networks (IC-UASNs). Specifically, we investigate the impact of unexpected node malfunctions on the network performance under the time-varying acoustic channels. Each node is expected to satisfy Quality of Service (QoS) requirements. However, achieving individual QoS requirements may interfere with other concurrent communications. Underwater nodes rely excessively on the rationality of other underwater nodes when guided by fully cooperative approaches, making it difficult to seek a trade-off between individual QoS and global fair-effective communications under imperfect conditions. Therefore, this paper presents a SEmi-COoperative Power Allocation approach (SECOPA) that achieves fair-effective communication and robustness in IC-UASNs. The approach is distributed multi-agent reinforcement learning (MARL)-based, and the objectives are twofold. On the one hand, each intelligent node individually decides the transmission power to simultaneously optimize individual and global performance. On the other hand, advanced training algorithms are developed to provide imperfect environments for training robust models that can adapt to the time-varying acoustic channels and handle unexpected node failures in the network. Numerical results are presented to validate our proposed approach.

cs.NI↗

arXiv2025

Joint link scheduling and power allocation in imperfect and energy-constrained underwater wireless sensor networks

Underwater wireless sensor networks (UWSNs) stand as promising technologies facilitating diverse underwater applications. However, the major design issues of the considered system are the severely limited energy supply and unexpected node malfunctions. This paper aims to provide fair, efficient, and reliable (FER) communication to the imperfect and energy-constrained UWSNs (IC-UWSNs). Therefore, we formulate a FER-communication optimization problem (FERCOP) and propose ICRL-JSA to solve the formulated problem. ICRL-JSA is a deep multi-agent reinforcement learning (MARL)-based optimizer for IC-UWSNs through joint link scheduling and power allocation, which automatically learns scheduling algorithms without human intervention. However, conventional RL methods are unable to address the challenges posed by underwater environments and IC-UWSNs. To construct ICRL-JSA, we integrate deep Q-network into IC-UWSNs and propose an advanced training mechanism to deal with complex acoustic channels, limited energy supplies, and unexpected node malfunctions. Simulation results demonstrate the superiority of the proposed ICRL-JSA scheme with an advanced training mechanism compared to various benchmark algorithms.

cs.NI↗

arXiv2025

Discovery of Itinerant Magnetic Domain Wall and Quasiparticle Boundary State in Spin-Density-Waves

Conventional magnetic domain walls are characterized by reorientation of local spins. However, what occurs at the boundary of itinerant magnets is largely unknown. Here using spin-sensitive scanning tunneling microscopy, we investigated the microscopic domain wall structure of the spin-density-wave (SDW) state in a prototypical itinerant antiferromagnet - chromium (Cr). At the boundary of two incommensurate SDW domains, we found the spins undergo finite-scale decay rather than reorientation. This generates a double-Q SDW state, which is further evidenced by an accompanying second-order charge modulation. In the commensurate SDW domains, a clear SDW energy gap is observed. Interestingly, the screw dislocations induced half vortex and anti-vortex of SDW, paired by antiphase domain wall. The spin density vanished at such antiphase domain walls. Remarkably, for the first time we observed the SDW quasiparticle states at the boundary, resembling the Andreev bound states in superconductors. These unique SDW boundary structures can be viewed as consequences of local interference of two SDWs, either with different Q or reversed phases. Our findings thus reveal a new type of domain wall distinct to that of local moment magnetism, with a mechanism rooted in the itinerant nature of SDW.

cond-mat.str-el↗

arXiv2025

Multi-Modal Hypergraph Enhanced LLM Learning for Recommendation

The burgeoning presence of Large Language Models (LLM) is propelling the development of personalized recommender systems. Most existing LLM-based methods fail to sufficiently explore the multi-view graph structure correlations inherent in recommendation scenarios. To this end, we propose a novel framework, Hypergraph Enhanced LLM Learning for multimodal Recommendation (HeLLM), designed to equip LLMs with the capability to capture intricate higher-order semantic correlations by fusing graph-level contextual signals with sequence-level behavioral patterns. In the recommender pre-training phase, we design a user hypergraph to uncover shared interest preferences among users and an item hypergraph to capture correlations within multimodal similarities among items. The hypergraph convolution and synergistic contrastive learning mechanism are introduced to enhance the distinguishability of learned representations. In the LLM fine-tuning phase, we inject the learned graph-structured embeddings directly into the LLM's architecture and integrate sequential features capturing each user's chronological behavior. This process enables hypergraphs to leverage graph-structured information as global context, enhancing the LLM's ability to perceive complex relational patterns and integrate multimodal information, while also modeling local temporal dynamics. Extensive experiments demonstrate the superiority of our proposed method over state-of-the-art baselines, confirming the advantages of fusing hypergraph-based context with sequential user behavior in LLMs for recommendation.

cs.IR↗

arXiv2025

Thiol post-translational modifications modulate allosteric regulation of the OpcA-G6PDH complex through conformational gate control

Cyanobacteria require ultra-fast metabolic switching to maintain reducing power balance during environmental fluctuations. Glucose-6-phosphate dehydrogenase (G6PDH), catalyzing the rate-limiting step of the oxidative pentose phosphate pathway (OPPP), provides essential NADPH and metabolic intermediates for biosynthetic processes and redox homeostasis. In cyanobacteria, the unique redox-sensitive protein OpcA acts as a metabolic switch for G6PDH, enabling rapid adjustment of reducing power generation from glycogen catabolism and resulting in precise regulation of carbon flux between anabolic and catabolic pathways. While the redox-sensitive cysteine structures of OpcA are known to regulate G6PDH, the detailed mechanisms of how redox post-translational modifications (PTMs) influence OpcA's allosteric effects on G6PDH structures and function remain elusive. To investigate this mechanism, we utilized computational modeling combined with experimental redox proteomics using Synechococcus elongatus PCC 7942 as a model system. Redox proteomics captured modified cysteine residues under light/dark or circadian shifts. Computational simulation revealed that thiol PTMs near the OpcA-G6PDH interface are crucial to allosteric regulation of regions affecting the G6PDH activity, including a potential gate region for substrate ingress and product egress, as well as critical hydrogen bond networks within the active site. These PTMs promote rapid metabolic switching by enhancing G6PDH catalytic activity when OpcA is oxidized. This study provides evidence for novel molecular mechanisms that elucidate the importance of thiol PTMs of OpcA in modulating G6PDH structure and function in an allosteric manner, demonstrating how PTM-level regulation provides a critical control mechanism that enables cyanobacteria to rapidly adapt to environmental fluctuations through precise metabolic fine-tuning.

physics.bio-ph↗

arXiv2025

AlignDistil: Token-Level Language Model Alignment as Adaptive Policy Distillation

In modern large language models (LLMs), LLM alignment is of crucial importance and is typically achieved through methods such as reinforcement learning from human feedback (RLHF) and direct preference optimization (DPO). However, in most existing methods for LLM alignment, all tokens in the response are optimized using a sparse, response-level reward or preference annotation. The ignorance of token-level rewards may erroneously punish high-quality tokens or encourage low-quality tokens, resulting in suboptimal performance and slow convergence speed. To address this issue, we propose AlignDistil, an RLHF-equivalent distillation method for token-level reward optimization. Specifically, we introduce the reward learned by DPO into the RLHF objective and theoretically prove the equivalence between this objective and a token-level distillation process, where the teacher distribution linearly combines the logits from the DPO model and a reference model. On this basis, we further bridge the accuracy gap between the reward from the DPO model and the pure reward model, by building a contrastive DPO reward with a normal and a reverse DPO model. Moreover, to avoid under- and over-optimization on different tokens, we design a token adaptive logit extrapolation mechanism to construct an appropriate teacher distribution for each token. Experimental results demonstrate the superiority of our AlignDistil over existing methods and showcase fast convergence due to its token-level distributional reward optimization.

cs.CL↗

arXiv2025

Robust Convergency Indicator using MIMO-PI Controller in the presence of disturbances

The PID controller remains the most widely adopted control architecture, with groundbreaking success across extensive implications. However, optimal parameter tuning for PID controller remains a critical challenge. Existing theories predominantly focus on linear time-invariant systems and Single-Input Single-Output (SISO) scenarios, leaving a research gap in addressing complex PID control problems for Multi-Input Multi-Output (MIMO) nonlinear systems with disturbances. This study enhances controller robustness by leveraging insights into the velocity form of nonlinear systems. It establishes a quantitative metric to evaluate the robustness of MIMO-PI controller, clarifies key theories on how robustness influences exponential error stabilization. Guided by these theories, an optimal robust MIMO-PI controller is developed without oversimplifying assumptions. Experimental results demonstrate that the controller achieves effective exponential stabilization and exhibits exceptional robustness under the guidance of the proposed robust indicator. Notably, the robust convergence indicator can also effectively assess comprehensive performance.

eess.SY↗

arXiv2025

Optimizing Fluid Antenna Configurations for Constructive Interference Precoding

The fluid antenna system (FAS) has emerged as a new physical-layer concept to provide enhanced propagation conditions for multiuser multiple-input multiple-output (MIMO) communications over conventional fixed arrays. This work focuses on minimizing the maximum symbol error probability (SEP) under $M$-ary phase shift keying (MPSK) signaling in a multiuser downlink equipped with FAS, where each antenna moves within nonoverlapping intervals. This specific problem of joint SEP minimization with FAS and constructive interference (CI) precoding has not been previously addressed. The resulting problem turns out to be a nonconvex and nonsmooth optimization challenge. We transform the SEP minimization problem into a safety margin maximization problem in constructive interference precoding. Then, we customize a smoothing technique and a block coordinate descent (BCD) algorithm, with emphasis on low computational complexity. Simulation results show that our approach can reduce bit error rate (BER) compared to both the fixed arrays and FAS designed by existing particle swarm optimization (PSO). Also, our approach shows attractively low computational complexity compared to PSO benchmarks.

eess.SP↗

arXiv2025

LLM Agent for Hyper-Parameter Optimization

Hyper-parameters are essential and critical for the performance of communication algorithms. However, current hyper-parameters optimization approaches for Warm-Start Particles Swarm Optimization with Crossover and Mutation (WS-PSO-CM) algorithm, designed for radio map-enabled unmanned aerial vehicle (UAV) trajectory and communication, are primarily heuristic-based, exhibiting low levels of automation and improvable performance. In this paper, we design an Large Language Model (LLM) agent for automatic hyper-parameters-tuning, where an iterative framework and Model Context Protocol (MCP) are applied. In particular, the LLM agent is first set up via a profile, which specifies the boundary of hyper-parameters, task objective, terminal condition, conservative or aggressive strategy of optimizing hyper-parameters, and LLM configurations. Then, the LLM agent iteratively invokes WS-PSO-CM algorithm for exploration. Finally, the LLM agent exits the loop based on the terminal condition and returns an optimized set of hyperparameters. Our experiment results show that the minimal sum-rate achieved by hyper-parameters generated via our LLM agent is significantly higher than those by both human heuristics and random generation methods. This indicates that an LLM agent with PSO and WS-PSO-CM algorithm knowledge is useful in seeking high-performance hyper-parameters.

cs.IT↗

arXiv2025

Listwise Preference Alignment Optimization for Tail Item Recommendation

Preference alignment has achieved greater success on Large Language Models (LLMs) and drawn broad interest in recommendation research. Existing preference alignment methods for recommendation either require explicit reward modeling or only support pairwise preference comparison. The former directly increases substantial computational costs, while the latter hinders training efficiency on negative samples. Moreover, no existing effort has explored preference alignment solutions for tail-item recommendation. To bridge the above gaps, we propose LPO4Rec, which extends the Bradley-Terry model from pairwise comparison to listwise comparison, to improve the efficiency of model training. Specifically, we derive a closed form optimal policy to enable more efficient and effective training without explicit reward modeling. We also present an adaptive negative sampling and reweighting strategy to prioritize tail items during optimization and enhance performance in tail-item recommendations. Besides, we theoretically prove that optimizing the listwise preference optimization (LPO) loss is equivalent to maximizing the upper bound of the optimal reward. Our experiments on three public datasets show that our method outperforms 10 baselines by a large margin, achieving up to 50% performance improvement while reducing 17.9% GPU memory usage when compared with direct preference optimization (DPO) in tail-item recommendation. Our code is available at https://github.com/Yuhanleeee/LPO4Rec.

cs.IR↗

arXiv2025

Adapt Your Body: Mitigating Proprioception Shifts in Imitation Learning

Imitation learning models for robotic tasks typically rely on multi-modal inputs, such as RGB images, language, and proprioceptive states. While proprioception is intuitively important for decision-making and obstacle avoidance, simply incorporating all proprioceptive states leads to a surprising degradation in imitation learning performance. In this work, we identify the underlying issue as the proprioception shift problem, where the distributions of proprioceptive states diverge significantly between training and deployment. To address this challenge, we propose a domain adaptation framework that bridges the gap by utilizing rollout data collected during deployment. Using Wasserstein distance, we quantify the discrepancy between expert and rollout proprioceptive states and minimize this gap by adding noise to both sets of states, proportional to the Wasserstein distance. This strategy enhances robustness against proprioception shifts by aligning the training and deployment distributions. Experiments on robotic manipulation tasks demonstrate the efficacy of our method, enabling the imitation policy to leverage proprioception while mitigating its adverse effects. Our approach outperforms the naive solution which discards proprioception, and other baselines designed to address distributional shifts.

cs.RO↗