arXiv subjects

Zixuan Zhang

Publications and source records attributed to Zixuan Zhang.

At least 19 recordsLinked to original sources

arXiv2026

Noise-Aware and Dynamically Adaptive Federated Defense Framework for SAR Image Target Recognition

As a critical application of computational intelligence in remote sensing, deep learning-based synthetic aperture radar (SAR) image target recognition facilitates intelligent perception but typically relies on centralized training, where multi-source SAR data are uploaded to a single server, raising privacy and security concerns. Federated learning (FL) provides an emerging computational intelligence paradigm for SAR image target recognition, enabling cross-site collaboration while preserving local data privacy. However, FL confronts critical security risks, where malicious clients can exploit SAR's multiplicative speckle noise to conceal backdoor triggers, severely challenging the robustness of the computational intelligence model. To address this challenge, we propose NADAFD, a noise-aware and dynamically adaptive federated defense framework that integrates frequency-domain, spatial-domain, and client-behavior analyses to counter SAR-specific backdoor threats. Specifically, we introduce a frequency-domain collaborative inversion mechanism to expose cross-client spectral inconsistencies indicative of hidden backdoor triggers. We further design a noise-aware adversarial training strategy that embeds $Γ$-distributed speckle characteristics into mask-guided adversarial sample generation to enhance robustness against both backdoor attacks and SAR speckle noise. In addition, we present a dynamic health assessment module that tracks client update behaviors across training rounds and adaptively adjusts aggregation weights to mitigate evolving malicious contributions. Experiments on MSTAR and OpenSARShip datasets demonstrate that NADAFD achieves higher accuracy on clean test samples and a lower backdoor attack success rate on triggered inputs than existing federated backdoor defenses for SAR target recognition.

cs.CR↗

arXiv2026

QUBRIC: Co-Designing Queries and Rubrics for RL Beyond Verifiable Rewards

Rubric-based RL is a promising route for extending reinforcement learning beyond verifiable rewards, yet existing methods optimize rubrics while treating the query distribution as fixed. We identify a structural bottleneck: rubric quality is constrained by query structure. Open-ended queries yield vague rubrics; naively narrowing them introduces fabricated references that no model can verify, so all responses fail and training receives no reward signal. We present QUBRIC, a framework that co-designs queries and rubrics. Teacher-derived key points ground the rewriting of open-ended queries into scenario-based, evaluable questions. Contrastive rubric generation then turns teacher-policy gaps into query-level criteria, and learnability filtering retains only informative query-rubric pairs for GRPO training. QUBRIC achieves a +5.5 point gain on ArenaHard over the SFT baseline. Trained only on instruction-following data, it further transfers to three held-out benchmarks spanning legal, moral, and narrative reasoning (+6.3 points on average), with improvements concentrated in reasoning-related dimensions. These results provide evidence that co-designing queries and rubrics can make rubric-based RL a practical complement to RLVR beyond strictly verifiable tasks.

cs.CL↗

arXiv2026

Enhancing Group Recommendation with Memory-Augmented Reasoning in LLM Agent

The core challenge in group recommendation lies in modeling the dynamic evolution of user preferences and explain?ing the consensus formation process. Existing Large Language Model (LLM)-based methods, despite improved interpretability, treat interaction history as fixed text, ignoring the natural evolution of group/user preferences over time, and lacking explicit modeling of the complex group decision-making process. To address these issues, we propose AGR, a LLM-based agent, which consists of a Memory Module and a Reasoning Module. The Memory Module employs a token-based hash table to dynamically manage the historical interactions of groups and users. This design supports fundamental operations including insertion, updating, retrieval, forgetting of irrelevant records, and summarization of evolving group and user profiles for efficiently tracking. Based on these retrieved dynamic profiles, the Reason?ing Module then performs a multi-step reasoning process includ?ing Group Interests Collection, Group Consensus Refinement, Multi-dimensional Evaluation and Explainable Recommendation Generation, thereby moving beyond black-box inference to de?liver fully interpretable recommendations. In practice, we adopt the Reinforcement Fine-Tuning (RFT) paradigm, where we first use Supervised Fine-Tuning (SFT) to equip the model with basic capabilities for invoking the Memory and Reasoning modules, and then employ Group Relative Policy Optimization (GRPO) to enhance its autonomous ability to coordinate these modules. Experiments on LastFM and Douban datasets demonstrate that AGR significantly outperforms existing state-of-the-art methods in both recommendation accuracy and explainability. Our model is open-sourced at https://huggingface.co/niuqimeng/AGR.

cs.IR↗

arXiv2026

RoamFlow: Reinforcement-Aligned One-Step Action MeanFlow Policy for Image-Goal Navigation

Image-goal navigation is a key challenge in embodied robotics, where an agent must reach a target specified solely by a goal image. While existing reinforcement learning approaches map perceptual observations directly to actions, they struggle to model long-horizon dependencies, often leading to suboptimal trajectories. To address this limitation, we propose RoamFlow, a generative navigation framework that leverages MeanFlow to predict the average velocity field for trajectory synthesis, enabling efficient few-step generation and reducing inference latency. We further adopt a two-stage training strategy that combines expert imitation for stable initialization with reinforcement learning for task-specific policy refinement. Extensive experiments in both Habitat simulation and real-world robotic platforms demonstrate that RoamFlow achieves efficient inference while maintaining strong navigation performance under real-time constraints.

cs.RO↗

arXiv2026

Instant Personalized Large Language Model Adaptation via Hypernetwork

Personalized large language models (LLMs) tailor content to individual preferences using user profiles or histories. However, existing parameter-efficient fine-tuning (PEFT) methods, such as the ``One-PEFT-Per-User'' (OPPU) paradigm, require training a separate adapter for each user, making them computationally expensive and impractical for real-time updates. We introduce Profile-to-PEFT, a scalable framework that employs a hypernetwork, trained end-to-end, to map a user's encoded profile directly to a full set of adapter parameters (e.g., LoRA), eliminating per-user training at deployment. This design enables instant adaptation, generalization to unseen users, and privacy-preserving local deployment. Experimental results demonstrate that our method outperforms both prompt-based personalization and OPPU while using substantially fewer computational resources at deployment. The framework exhibits strong generalization to out-of-distribution users and maintains robustness across varying user activity levels and different embedding backbones. The proposed Profile-to-PEFT framework enables efficient, scalable, and adaptive LLM personalization suitable for large-scale applications.

cs.CL↗

arXiv2026

How Accurately Can a Gaussian Approximate Stochastic Approximation Iterates?

Stochastic approximation (SA) is a method for finding the root of an operator perturbed by noise. The focus of this paper is studying the distribution of SA iterates in finite time. In general, it is not possible to characterize the exact distribution, and therefore our goal is to find an approximation which can yield useful tail bounds. Inspired by the rich literature on the asymptotic normality of rescaled SA iterates, we approximate the pre-limit distributions by a sequence of Gaussians whose covariance is recursively defined. In particular, we establish explicit bounds on the Wasserstein-1 distance between the rescaled iterate at time $k$ and the aforementioned Gaussian for various choices of step-sizes. Since these covariances converge to the classical asymptotic limit, our analysis also provides a convergence rate for asymptotic normality as a by-product. As an immediate consequence of our bounds, we obtain tail bounds on the error of SA iterates at any time. Finally, we establish the sharpness of our rates by providing matching lower bounds and validate our findings through simulations. We obtain the sharp rates by first studying the convergence rate of the discrete Ornstein-Uhlenbeck (O-U) process driven by general noise, whose stationary distribution is identical to the limiting Gaussian distribution of the rescaled SA iterates. We believe that this is of independent interest, given its connection to sampling literature. The analysis involves adapting Stein's method for Gaussian approximation to handle the matrix weighted sum of i.i.d. random variables. The desired finite-time bounds for SA are obtained by characterizing the error dynamics between the rescaled SA iterate and the discrete time O-U process and combining it with the convergence rate of the latter process.

stat.ML↗

arXiv2026

Optimistic ε-Greedy Exploration for Cooperative Multi-Agent Reinforcement Learning

The Centralized Training with Decentralized Execution (CTDE) paradigm is widely used in cooperative multi-agent reinforcement learning. However, conventional methods based on CTDE can suffer from value underestimation and converge to suboptimal solutions. While such underestimation is typically attributed to the representational limitations of monotonic structures, we provide a novel perspective by demonstrating that the insufficient sampling of optimal joint actions during exploration is also a critical factor. To address this problem, we propose Optimistic $ε$-Greedy Exploration. Our method introduces optimistic action-value networks that serve as decoupled exploration indicators, which we theoretically prove to converge in probability to the maximum achievable returns. By sampling actions from these distributions with a probability of $ε$, we effectively increase the selection frequency of high-return joint actions. Experimental results in various environments reveal that our strategy effectively prevents the algorithm from falling into suboptimal solutions and significantly improves final returns, win rates, and convergence speeds compared to other enhanced algorithms. Our code has been open-sourced at https://github.com/qxqxtxdy/OptimisticExploration.

cs.MA↗

arXiv2026

Breaking the Rigid Prior: Towards Articulated 3D Anomaly Detection

Existing 3D anomaly detection methods are built on a rigid prior: normal geometry is pose-invariant and can be canonicalized through registration or alignment. This prior does not hold for articulated objects with hinge or sliding joints, where valid pose changes induce structured geometric variations that cannot be collapsed to a single canonical template, causing pose-induced deformations to be misidentified as anomalies while true structural defects are obscured. No existing benchmark addresses this challenge. We introduce ArtiAD, the first large-scale benchmark for articulated 3D anomaly detection, comprising 15,229 point clouds across 39 object categories with dense joint-angle variations and six structural anomaly types. Each sample is annotated with its joint configuration and part-level motion labels, enabling explicit disentanglement of pose-induced geometry from structural defects. ArtiAD also provides a seen/unseen articulation split to evaluate both interpolation and extrapolation to novel joint configurations. We propose Shape-Pose-Aware Signed Distance Field (SPA-SDF), a baseline that replaces the rigid prior with a continuous pose-conditioned implicit field, factorized into an articulation-independent structural prior and a Fourier-encoded joint embedding. At inference, the articulation state is recovered by minimizing reconstruction energy, and anomalies are identified as point-wise deviations from the learned manifold. SPA-SDF achieves 0.884 object-level AUROC on seen configurations and 0.874 on unseen configurations, substantially outperforming all rigid-based baselines. Our code and benchmark will be publicly released to facilitate future research.

cs.CV↗

arXiv2026

Diffusion Model for Manifold Data: Score Decomposition, Curvature, and Statistical Complexity

Diffusion models have become a leading framework in generative modeling, yet their theoretical understanding -- especially for high-dimensional data concentrated on low-dimensional structures -- remains incomplete. This paper investigates how diffusion models learn such structured data, focusing on two key aspects: statistical complexity and influence of data geometric properties. By modeling data as samples from a smooth Riemannian manifold, our analysis reveals crucial decompositions of score functions in diffusion models under different levels of injected noise. We also highlight the interplay of manifold curvature with the structures in the score function. These analyses enable an efficient neural network approximation to the score function, built upon which we further provide statistical rates for score estimation and distribution learning. Remarkably, the obtained statistical rates are governed by the intrinsic dimension of data and the manifold curvature. These results advance the statistical foundations of diffusion models, bridging theory and practice for generative modeling on manifolds.

cs.LG↗

arXiv2026

BackPlay: Head-Only Look-Back Self-Correction for Diffusion Language Models

Diffusion Language Models (DLMs) decode multiple tokens in parallel, but aggressive multi-token decoding amplifies cross-token dependency errors and can sharply degrade generation quality. We propose BackPlay, a frozen-backbone self-correction framework that trains only a lightweight correction head on a finetuned DLM without updating any backbone or adapter parameters. Because the head is trained on errors produced by the same frozen generator used at inference time, its training distribution aligns with the error patterns of the deployed model. We further introduce Look-back Correction, a training mechanism that injects predictions from earlier, more corrupted denoising states into later, richer contexts, enabling the head to leverage later context to detect mistakes made in earlier generation steps. During inference, BackPlay periodically revisits previously generated tokens through selective remasking and regeneration to limit error accumulation. Across mathematical reasoning and code generation benchmarks, BackPlay improves the speed--quality trade-off of the underlying DLM under multi-token decoding.

cs.LG↗

arXiv2026

Controllable and Verifiable Tool-Use Data Synthesis for Agentic Reinforcement Learning

Existing synthetic tool-use corpora are primarily designed for offline supervised fine-tuning, yet reinforcement learning (RL) requires executable environments that support reward-checkable online rollouts. We propose COVERT, a two-stage pipeline that first generates reliable base tool-use trajectories through self-evolving synthesis with multi-level validation, and then applies oracle-preserving augmentations that systematically increase environmental complexity. These augmentations introduce distractor tools, indirect or ambiguous user queries, and noisy, multi-format, or erroneous tool outputs, while strictly preserving oracle tool calls and final answers as ground truth. This design enables automatic reward computation via reference matching for standard cases and lightweight judge-assisted verification for special behaviors such as error detection, supporting RL optimization of tool-calling policies. On Qwen2.5-Instruct-14B, COVERT-RL improves overall accuracy on BFCL v3 from 56.5 to 59.9 and on ACEBench from 53.0 to 59.3, with minimal regressions on general-ability benchmarks; when stacked on SFT, it further reaches 62.1 and 61.8, confirming additive gains. These results suggest that oracle-preserving synthetic environments offer a practical RL refinement stage, complementary to SFT, for improving tool-use robustness under ambiguity and unreliable tool feedback.

cs.AI↗

arXiv2026

Training LLMs for Multi-Step Tool Orchestration with Constrained Data Synthesis and Graduated Rewards

Multi-step tool orchestration remains challenging for LLMs, as state-of-the-art models frequently fail on full sequence execution due to parameter errors. Training for these workflows faces two obstacles: the lack of environments supporting complex real-world API dependencies, and sparse binary rewards that provide no signal for partial correctness. We propose a reinforcement learning framework addressing both challenges. First, we construct a deterministic environment backed by a large-scale cache of real API responses, enabling constrained synthesis of valid multi-step traces with controllable complexity. Second, we introduce a graduated reward that decomposes correctness into atomic validity (call-level correctness at increasing granularity) and orchestration consistency (correct sequencing with dependency respect). On ComplexFuncBench, our approach substantially improves turn accuracy, with ablations confirming both reward components are essential. Cross-benchmark evaluation on BFCL v4 shows that the learned orchestration skills transfer to entirely different API ecosystems (e.g., agentic web search and memory management), yielding consistent gains while maintaining stable single-step performance. Code is available at https://github.com/horizon-rl/ToolOrchestrationReward

cs.LG↗

arXiv2026

Learning to Optimize Multi-Objective Alignment Through Dynamic Reward Weighting

Prior work in multi-objective reinforcement learning typically uses linear reward scalarization with fixed weights, which provably fails to capture non-convex Pareto fronts and thus yields suboptimal results. This limitation becomes especially critical in online preference alignment for large language models. Here, stochastic trajectories generated by parameterized policies create highly non-linear and non-convex mappings from parameters to objectives that no single static weighting scheme can find optimal trade-offs. We address this limitation by introducing dynamic reward weighting, which adaptively adjusts reward weights during the online reinforcement learning process. Unlike existing approaches that rely on fixed-weight interpolation, our dynamic weighting continuously balances and prioritizes objectives in training, facilitating effective exploration of Pareto fronts in objective space. We introduce two approaches of increasing sophistication and generalizability: hypervolume-guided weight adaptation and gradient-based weight optimization, offering a versatile toolkit for online multi-objective alignment. Our extensive experiments demonstrate their compatibility with commonly used online reinforcement learning algorithms, effectiveness across multiple datasets, and applicability to different model families, consistently achieving Pareto dominant solutions with fewer training steps than fixed-weight linear scalarization baselines.

cs.LG↗

arXiv2026

Data-Efficient and Robust Trajectory Generation through Pathlet Dictionary Learning

Trajectory generation has recently drawn growing interest in privacy-preserving urban mobility studies and location-based service applications. Although many studies have used deep learning or generative AI methods to model trajectories and have achieved promising results, the robustness and interpretability of such models are largely unexplored. This limits the application of trajectory generation algorithms on noisy real-world data and their trustworthiness in downstream tasks. To address this issue, we exploit the regular structure in urban trajectories and propose a deep generative model based on the pathlet representation, which encode trajectories with binary vectors associated with a learned dictionary of trajectory segments. Specifically, we introduce a probabilistic graphical model to describe the trajectory generation process, which includes a Variational Autoencoder (VAE) component and a linear decoder component. During training, the model can simultaneously learn the latent embedding of pathlet representations and the pathlet dictionary that captures mobility patterns in the trajectory dataset. The conditional version of our model can also be used to generate customized trajectories based on temporal and spatial constraints. Our model can effectively learn data distribution even using noisy data, achieving relative improvements of $35.4\%$ and $26.3\%$ over strong baselines on two real-world trajectory datasets. Moreover, the generated trajectories can be conveniently utilized for multiple downstream tasks, including trajectory prediction and data denoising. Lastly, the framework design offers a significant efficiency advantage, saving $64.8\%$ of the time and $56.5\%$ of GPU memory compared to previous approaches.

cs.LG↗

arXiv2025

COSMOS: A Hybrid Adaptive Optimizer for Memory-Efficient Training of LLMs

Large Language Models (LLMs) have demonstrated remarkable success across various domains, yet their optimization remains a significant challenge due to the complex and high-dimensional loss landscapes they inhabit. While adaptive optimizers such as AdamW are widely used, they suffer from critical limitations, including an inability to capture interdependencies between coordinates and high memory consumption. Subsequent research, exemplified by SOAP, attempts to better capture coordinate interdependence but incurs greater memory overhead, limiting scalability for massive LLMs. An alternative approach aims to reduce memory consumption through low-dimensional projection, but this leads to substantial approximation errors, resulting in less effective optimization (e.g., in terms of per-token efficiency). In this paper, we propose COSMOS, a novel hybrid optimizer that leverages the varying importance of eigensubspaces in the gradient matrix to achieve memory efficiency without compromising optimization performance. The design of COSMOS is motivated by our empirical insights and practical considerations. Specifically, COSMOS applies SOAP to the leading eigensubspace, which captures the primary optimization dynamics, and MUON to the remaining eigensubspace, which is less critical but computationally expensive to handle with SOAP. This hybrid strategy significantly reduces memory consumption while maintaining robust optimization performance, making it particularly suitable for massive LLMs. Numerical experiments on various datasets and transformer architectures are provided to demonstrate the effectiveness of COSMOS. Our code is available at https://github.com/lliu606/COSMOS.

cs.LG↗

arXiv2025

Machine Learning Powered Feasible Path Framework with Adaptive Sampling for Black-box Optimization

Black-box optimization (BBO) involves functions that are unknown, inexact and/or expensive-to-evaluate. Existing BBO algorithms face several challenges, including high computational cost from extensive evaluations, difficulty in handling complex constraints, lacking theoretical convergence guarantees and/or instability due to large solution quality variation. In this work, a machine learning-powered feasible path optimization framework (MLFP) is proposed for general BBO problems including complex constraints. An adaptive sampling strategy is first proposed to explore optimal regions and pre-filter potentially infeasible points to reduce evaluations. Machine learning algorithms are leveraged to develop surrogates of black-boxes. The feasible path algorithm is employed to accelerate theoretical convergence by updating independent variables rather than all. Computational studies demonstrate MLFP can rapidly and robustly converge around the KKT point, even training surrogates with small datasets. MLFP is superior to the state-of-the-art BBO algorithms, as it stably obtains the same or better solutions with fewer evaluations for benchmark examples.

math.OC↗

arXiv2025

Why Does New Knowledge Create Messy Ripple Effects in LLMs?

Extensive previous research has focused on post-training knowledge editing (KE) for language models (LMs) to ensure that knowledge remains accurate and up-to-date. One desired property and open question in KE is to let edited LMs correctly handle ripple effects, where LM is expected to answer its logically related knowledge accurately. In this paper, we answer the question of why most KE methods still create messy ripple effects. We conduct extensive analysis and identify a salient indicator, GradSim, that effectively reveals when and why updated knowledge ripples in LMs. GradSim is computed by the cosine similarity between gradients of the original fact and its related knowledge. We observe a strong positive correlation between ripple effect performance and GradSim across different LMs, KE methods, and evaluation metrics. Further investigations into three counter-intuitive failure cases (Negation, Over-Ripple, Multi-Lingual) of ripple effects demonstrate that these failures are often associated with very low GradSim. This finding validates that GradSim is an effective indicator of when knowledge ripples in LMs.

cs.CL↗

arXiv2025

Orbital chiral lasing in twisted bilayer metasurfaces

Chirality is a fundamental concept in physics that underpins various phenomena in nonlinear optics, quantum physics, and topological photonics. Although the spin of a photon naturally brings chirality, orbital angular momentum can also become chirally active in the structures with a broken mirror symmetry. Here, we observe orbital chiral lasing from a twisted bilayer photonic structure leveraging its inherent structural chirality. Specifically, we design and fabricate a Moire-type optical structure by bonding and rotating two separate semiconductor membrane metasurfaces. We achieve single-mode lasing over a broad spectral range of 250 nm by optically pumping the twisted structure. The lasing emission exhibits orbital chiral characteristics, arising from helical and non-Hermitian couplings between clockwise and counter-clockwise rotating collective guided resonances, confirmed by polarization-resolved imaging and self-interference patterns. Our results provide the first observation of orbital chiral lasing in twisted photonics, and they can contribute to diverse applications of chiral light in diagnostics, optical manipulation, and communication with light.

physics.optics↗