Searcharxiv⌕ Search

arXiv subjects

Ziqi Chen

Publications and source records attributed to Ziqi Chen.

At least 19 recordsLinked to original sources

CascadeJEPA: Enhancing 3D Geometric Representations via Spatially Coupled Cascaded Prediction

Joint-Embedding Predictive Architectures (JEPAs) have emerged as a promising paradigm for predictive visual representation learning. However, our analysis of I-JEPA reveals a growing discrepancy between semantic and geometric learning: improved semantic recognition during pretraining is accompanied by degraded 3D geometric representations. We hypothesize that this partly stems from predicting highly abstract latent representations with incomplete spatial context, favoring semantic regularities over fine-grained geometric information. To address this issue, we introduce CascadeJEPA, a spatially coupled cascaded prediction framework that promotes richer patch-level interactions through full-grid context encoding and cross-attention between prediction stages, while largely preserving the original I-JEPA architecture. Extensive experiments demonstrate substantial improvements over I-JEPA across multiple 3D geometric tasks while maintaining strong semantic recognition performance. Furthermore, its advantages extend to future feature forecasting and world model planning, highlighting the potential of enhanced geometric representations to support predictive world modeling.

cs.CV↗

Enhancing Mixed Oil Length Prediction through Graph Representation Learning

During the pipeline transportation of refined oil products, the inevitable formation of mixed oil segments poses considerable economic challenges to pipeline networks. Therefore, it is important to accurately predict the length of these mixed oil segments. Although previous methods have achieved relatively accurate predictions by integrating machine and deep learning algorithms, they ignore the statistical dependencies that remain among real-world mixed oil samples, which degrades model accuracy. To address this issue, we propose a novel Sample Dependencies Incorporated Graph Representation Learning (SDI-GRL) model. By representing samples and pipeline features as nodes of a graph, SDI-GRL connects dependent samples through their shared features, allowing message passing to propagate information across related samples. This graph-based framework allows our model to capture the statistical dependencies among samples absent in prior works. A case study on a mixed oil dataset demonstrates that our SDI-GRL model surpasses the strongest existing approach, reducing both the RMSE and the MAE by more than 10%.

cs.LG↗

A Simple yet Effective Framework for Pre-treatment and Post-treatment Glioma Segmentation Using Multimodal MRI

Accurate glioma segmentation across pre- and post-treatment MRI remains challenging because treatment alters both image appearance and the target regions. In particular, the resection cavity (RC) is defined only for post-treatment cases, complicating joint learning from pooled data. We present RC-aware nnU-Netv2, a simple yet effective glioma segmentation framework that combines lesion- and boundary-aware single-cohort training, an RC-aware joint learning objective, and treatment-status-guided model routing at inference. During pooled training, the joint objective applies both standard and boundary-weighted binary cross-entropy to all four region channels, while employing Dice supervision for channels with non-empty targets and a controlled false-positive penalty for channels with empty targets across the mini-batch. This design explicitly accommodates the absence of RC in pre-treatment cases. Our approach uses only the standard nnU-Netv2 data augmentation and a single specialized model per case, without additional synthetic tumor generation or multi-model ensembling. RC-aware nnU-Netv2 ranked second in the BraTS 2025 Lighthouse Challenge Task 1 (BraTS-GLI 2025), with segmentation performance comparable to that of the first-place method, which relied on on-the-fly GliGAN-based synthetic tumor augmentation and multi-model ensembling. On the official blind test set, our method achieved mean lesion-wise Dice scores of 0.7878, 0.8709, 0.7923, and 0.8715 and mean NSD@1.0 scores of 0.8309, 0.8712, 0.7980, and 0.8336 for enhancing tumor, RC, tumor core, and whole tumor, respectively. In paired post-treatment analysis, the RC-aware joint model improved mean lesion-wise Dice by 0.042 (95% CI: [0.028, 0.057]) and mean NSD@1.0 by 0.043 (95% CI: [0.028, 0.059]) compared with the joint baseline.

eess.IV↗

A Wireless World Model for AI-Native 6G Networks

Integrating AI into the physical layer is a cornerstone of 6G networks. However, current data-driven approaches struggle to generalize across dynamic environments because they lack an intrinsic understanding of electromagnetic wave propagation. We introduce the Wireless World Model (WWM), a multi-modal foundation framework predicting the spatiotemporal evolution of wireless channels by internalizing the causal relationship between 3D geometry and signal dynamics. Pre-trained on a massive ray-traced multi-modal dataset, WWM overcomes the data authenticity gap, further validated under real-world measurement data. Using a joint-embedding predictive architecture with a multi-modal mixture-of-experts Transformer, WWM fuses channel state information, 3D point clouds, and user trajectories into a unified representation. Across the five key downstream tasks supported by WWM, it achieves remarkable performance in seen environments, unseen generalization scenarios, and real-world measurements, consistently outperforming SOTA uni-modal foundation models and task-specific models. This paves the way for physics-aware 6G intelligence that adapts to the physical world.

cs.NI↗

PACEvolve++: Improving Test-time Learning for Evolutionary Search Agents

Large language models have become drivers of evolutionary search, but most systems rely on a fixed, prompt-elicited policy to sample next candidates. This limits adaptation in practical engineering and research tasks, where evaluations are expensive, and progress depends on learning task-specific search dynamics. We introduce PACEvolve++, an advisor-model reinforcement learning framework for test-time policy adaptation in evolutionary search agents. PACEvolve++ decouples strategic search decisions from implementation: a trainable advisor generates, assesses, and selects hypotheses, while a stronger frontier model translates selected hypotheses into executable candidates. To train the advisor under non-stationary feedback, we propose a phase-adaptive approach that adapts its optimization strategy to different phases of the evolutionary process. Early in evolution, it uses group-relative feedback to learn broad search preferences; later, as reward gaps compress, it emphasizes best-of-$k$ frontier contribution to support stable refinement. Across expert-parallel load balancing, sequential recommendation, and protein fitness extrapolation, PACEvolve++ outperforms the state-of-the-art evolutionary search framework with frontier models, achieving faster convergence and stabilizing test-time training during evolutionary search.

cs.LG↗

Domain-Adapted Diffusion Models for Conditional Independence Testing

Conditional independence (CI) is a fundamental concept in statistics and machine learning. Recent advances in conditional generative modeling provide flexible tools for generative-model-based CI tests, which rely on an estimated conditional distribution to generate randomized samples. However, errors in estimating this distribution accumulate in existing Type I error bounds, and consistency of the generative estimator alone does not guarantee asymptotic Type I error control. To address this limitation, we formulate conditional generative modeling as a domain adaptation problem and leverage auxiliary data from multiple source domains to improve estimation in the target CI testing domain. We propose Domain-Adapted Diffusion (DA-Diff), a multi-source domain adaptation framework for conditional diffusion models based on weighted empirical risk minimization over both target and source domains. We establish the convergence rate of DA-Diff and show how transferable source data can improve target-domain estimation through an increased effective sample size while controlling transfer bias. Building on DA-Diff, we further propose Domain-Adapted Conditional Independence Testing (DA-CIT) and show that its Type I error satisfies $P(p \leq α) \leq α+ o(1)$. Experiments demonstrate that DA-Diff improved conditional generation quality compared with transfer-learning diffusion baselines, while DA-CIT provides strong Type I error control and competitive power.

stat.ML↗

PACEvolve: Enabling Progress-Aware Consistent Evolution

Self-evolving agents powered by Large Language Models (LLMs) have emerged as a promising direction across diverse domains, including code optimization and scientific discovery, yet their core failure modes remain underexplored. Through a comprehensive empirical study, we identify that the model's reasoning becomes anchored to the local context of current hypotheses, overemphasizing low-level details while neglecting the broader search landscape. As a result, such agents become prone to context pollution and mode collapse, repeatedly revisiting flawed hypotheses and converging on suboptimal solutions. To address this challenge, we propose Progress-Aware Consistent Evolution (PACEvolve), a systematic framework for governing agent memory and search dynamics. PACEvolve overcomes these limitations through three key techniques: (1) Hierarchical Context Management (HCM), which structures historical trajectories while dynamically pruning branches to preserve a high-signal memory state; (2) Momentum-Based Backtracking (MBB), which monitors optimization progress to escape local minima; and (3) a self-adaptive Collaborative Evolution policy (CE) that balances intra-trajectory refinement with inter-trajectory knowledge transfer. By decoupling high-level idea generation from low-level code evaluation, PACEvolve maintains a global view of search momentum and achieves state-of-the-art results across complex evolutionary benchmarks.

cs.NE↗

CRT*: Conditional Randomization Testing with Heterogeneous External and Unlabeled Data

The conditional randomization test (CRT) provides a principled approach to conditional independence (CI) testing, guaranteeing exact type-I error control when the true conditional distribution is known. In practice, however, this distribution must be estimated, and estimation errors can inflate type-I errors, while high dimensionality and limited sample sizes can reduce power. Although external and unlabeled data offer the potential to improve CI testing, naive integration that ignores distributional heterogeneity can compromise type-I error control and fail to enhance power. We propose \textbf{CRT*}, a novel framework that robustly integrates external and unlabeled datasets to enhance CI testing in heterogeneous scenarios. CRT* employs smooth residual-bootstrap (SRB) with transfer learning for conditional distribution estimation, combined with adaptive data fusion via an optimal convex combination of test statistics. We theoretically establish that the SRB-based estimator converges to the true conditional distribution in expected total variation distance. Furthermore, even in high-dimensional regimes, CRT* maintains valid type-I error control and achieves strictly higher power than standard CRT without external data. Simulations and RNA-seq breast cancer data analyses demonstrate that CRT* substantially improves power while maintaining type-I error control in heterogeneous settings.

stat.ME↗

Generating Developable 3D Molecules via Pocket-Conditioned Diffusion and Property-Aware Optimization

Drug discovery and development is time-consuming and resource-intensive, motivating computational approaches such as diffusion models for de novo drug design. Many such models follow the structure-based drug design (SBDD) paradigm, generating molecules to fit a target binding pocket. However, existing diffusion-based SBDD methods typically couple pocket and ligand representation learning, model interactions only at the atom level, and prioritize binding affinity over other developability properties. Here, we introduce conDitar-dev, a conditional diffusion-based SBDD framework for generating ligands with strong binding affinities and favorable ADMET properties. It consists of three modules: msPRL, a pretrained multi-scale pocket representation learning module; conDitar, a pocket-conditioned diffusion model guided by msPRL representations; and paOPT, a generation-time method for optimizing ligand developability. On a newly curated benchmark of human disease targets, conDitar outperforms state-of-the-art SBDD baselines, achieving an average binding score of -8.85 kcal/mol. Across five ADMET properties, conDitar-dev improves performance by up to 73% over conDitar. To further validate the abilities of conDitar-dev to generate developable molecules, we have applied it to two validated druggable targets: programmed death-ligand 1 (PD-L1) and colony-stimulating factor 1 receptor (CSF1R) proteins. Top-ranked generatively designed molecules and their analogs have been experimentally synthesized and biologically tested. Two molecules generated directly by conDitar-dev for PD-L1 exhibited SPR-derived $K_D$ values of 3.49 and 3.75 $μ$M, respectively. Hit expansion based on conDitar-dev-designed molecules identified selective CSF1R inhibitors with IC$_{50}$ values as low as 200 nM, while also uncovering opportunities for drug repositioning.

cs.LG↗

DocMaster: A Hierarchical Structure-Aware System for Document Analysis

Leveraging large language models (LLMs) to analyze complex documents -- such as academic papers, technical manuals, and financial reports -- has emerged as a mainstream and critical task in both research and industry. In practice, users must first filter relevant documents from large collections and then conduct in-depth analysis (e.g. question answering) over the selected subset, yet existing systems flatten documents into plain-text chunks, discarding the rich hierarchical structures (sections, tables, figures, equations) and degrading downstream performance. We present DocMaster, a hierarchical structure-aware document analysis system. DocMaster parses documents into hierarchical document trees preserving original layouts and constructs a structure-aware semantic index that enables accurate document filtering and in-depth analysis. We demonstrate DocMaster through an interactive web interface that enables users to upload document collections, construct tree-based and multi-view semantic indices, filter relevant documents via natural-language conditions, and perform follow-up question answering over the filtered results. The source code, data, and demo are available at https://doc-master.github.io/.

cs.DB↗

Enhancing Multilingual LLM-based ASR with Mixture of Experts and Dynamic Downsampling

The rapid progress of large language models (LLMs) has opened up a new frontier for automatic speech recognition (ASR), making their effective integration a critical and challenging research direction. To this end, this work proposes a projector-based LLM-ASR framework targeting the key challenges of multilingual generalization and modality alignment. Our approach incorporates a Mixture of Experts (MoE) architecture to improve cross-lingual adaptability, and a Continuous Integrate-and-Fire (CIF) mechanism for dynamic downsampling and modality alignment. Experimental results show that the combination of these components yields substantial performance improvements, surpassing strong baseline models. The proposed method represents a step toward building more accurate, robust, and generalizable LLM-based ASR systems.

cs.SD↗

MOSS-Audio Technical Report

MOSS-Audio is a unified audio-language model for speech, environmental sound, and music understanding, supporting audio captioning, time-aware question answering, timestamped transcription, and audio-grounded reasoning. MOSS-Audio couples a dedicated audio encoder with a modality adapter and a large language model: the encoder produces 12.5 Hz temporal representations, the adapter projects them into the decoder space, and the decoder generates autoregressive text outputs. Two design choices are central to the system: DeepStack cross-layer feature injection, which exposes the decoder to acoustic information from multiple encoder depths, and time markers, which provide explicit temporal cues by inserting timestamp markers into the audio-token stream. At the data level, we design an event-preserving audio annotation pipeline that segments raw audio at coherent event boundaries, applies branch-specific annotation to speech, music, and general audio, and merges the results into unified captions for pretraining. The intermediate branch-specific captions are further retained to support the construction of task-oriented SFT data. The model is pretrained on large-scale audio-language data, with time-aware objectives incorporated to support temporal grounding, and then undergoes multi-stage post-training to enhance instruction following and audio-grounded reasoning. We release 4B and 8B variants in both Instruct and Thinking configurations. MOSS-Audio achieves strong performance across general audio understanding, speech captioning, ASR, and timestamped ASR, positioning it as a promising understanding foundation for future voice agents.

cs.SD↗

Bayesian characterization of porous media using three-microphone tube method in extended frequency ranges

The characteristic impedance and the propagation coefficient are among the most important parameters for evaluating the acoustic performance of porous materials. This work investigates the influence of cylindrical modes in an impermeable tube and applies multiple microphones distributed along the tube circumference within the three-microphone framework to extend the valid frequency range of characteristic impedance measurement. During the extended broadband measurements, discontinuities or phase jumps are observed in the experimentally measured propagation coefficient of the porous material under test. A Bayesian inference is applied in a sequential manner to estimate the unwrapped propagation coefficient and characteristic impedance. The results demonstrate that the inferred parameters accurately capture the behavior of the transfer function, allowing accurate parameter estimation.

physics.class-ph↗

MOSS-TTS Technical Report

This technical report presents MOSS-TTS, a speech generation foundation model built on a scalable recipe: discrete audio tokens, autoregressive modeling, and large-scale pretraining. Built on MOSS-Audio-Tokenizer, a causal Transformer tokenizer that compresses 24 kHz audio to 12.5 fps with variable-bitrate RVQ and unified semantic-acoustic representations, we release two complementary generators: MOSS-TTS, which emphasizes structural simplicity, scalability, and long-context/control-oriented deployment, and MOSS-TTS-Local-Transformer, which introduces a frame-local autoregressive module for higher modeling efficiency, stronger speaker preservation, and a shorter time to first audio. Across multilingual and open-domain settings, MOSS-TTS supports zero-shot voice cloning, token-level duration control, phoneme-/pinyin-level pronunciation control, smooth code-switching, and stable long-form generation. This report summarizes the design, training recipe, and empirical characteristics of the released models.

cs.SD↗

AgenticTagger: Structured Item Representation for Recommendation with LLM Agents

High-quality representations are a core requirement for effective recommendation. In this work, we study the problem of LLM-based descriptor generation, i.e., keyphrase-like natural language item representation generation frameworks with minimal constraints on downstream applications. We propose AgenticTagger, a framework that queries LLMs for representing items with sequences of text descriptors. However, open-ended generation provides little control over the generation space, leading to high cardinality, low-performance descriptors that render downstream modeling challenging. To this end, AgenticTagger features two core stages: (1) a vocabulary-building stage in which a set of hierarchical, low-cardinality, and high-quality descriptors is identified, and (2) a vocabulary-assignment stage in which LLMs assign in-vocabulary descriptors to items. To effectively and efficiently ground vocabulary in the item corpus of interest, we design a multi-agent reflection mechanism in which an architect LLM iteratively refines the vocabulary guided by parallelized feedback from annotator LLMs that validate the vocabulary against item data. Experiments on public and private data show AgenticTagger brings consistent improvements across diverse recommendation scenarios, including generative and term-based retrieval, ranking, and controllability-oriented, critique-based recommendation.

cs.IR↗

Conditionally Whitened Generative Models for Probabilistic Time Series Forecasting

Probabilistic forecasting of multivariate time series is challenging due to non-stationarity, inter-variable dependencies, and distribution shifts. While recent diffusion and flow matching models have shown promise, they often ignore informative priors such as conditional means and covariances. In this work, we propose Conditionally Whitened Generative Models (CW-Gen), a framework that incorporates prior information through conditional whitening. Theoretically, we establish sufficient conditions under which replacing the traditional terminal distribution of diffusion models, namely the standard multivariate normal, with a multivariate normal distribution parameterized by estimators of the conditional mean and covariance improves sample quality. Guided by this analysis, we design a novel Joint Mean-Covariance Estimator (JMCE) that simultaneously learns the conditional mean and sliding-window covariance. Building on JMCE, we introduce Conditionally Whitened Diffusion Models (CW-Diff) and extend them to Conditionally Whitened Flow Matching (CW-Flow). Experiments on five real-world datasets with six state-of-the-art generative models demonstrate that CW-Gen consistently enhances predictive performance, capturing non-stationary dynamics and inter-variable correlations more effectively than prior-free approaches. Empirical results further demonstrate that CW-Gen can effectively mitigate the effects of distribution shift.

stat.ML↗

Learning Heterogeneous Ordinal Graphical Models via Bayesian Nonparametric Clustering

Graphical models are powerful tools for capturing conditional dependence structures in complex systems but remain underexplored in analyzing ordinal data, especially in sports analytics. Ordinal variables, such as team rankings, player performance ratings, and survey responses, are pervasive in sports data but present unique challenges, particularly when accounting for heterogeneous subgroups, such as teams with varying styles or players with distinct roles. Existing methods, including probit graphical models, struggle with modeling heterogeneity and selecting the number of subgroups effectively. We propose a novel nonparametric Bayesian framework using the Mixture of Finite Mixtures (MFM) approach to address these challenges. Our method allows for flexible subgroup discovery and models each subgroup with a probit graphical model, simultaneously estimating the number of clusters and their configurations. We develop an efficient Gibbs sampling algorithm for inference, enabling robust estimation of cluster-specific structures and parameters. This framework is particularly suited to sports analytics, uncovering latent patterns in player performance metrics. Our work bridges critical gaps in modeling ordinal data and provides a foundation for advanced decision-making in sports performance and strategy.

stat.ME↗

RIS-based Communication Enhancement and Location Privacy Protection in UAV Networks

With the explosive advancement of unmanned aerial vehicles (UAVs), the security of efficient UAV networks has become increasingly critical. Owing to the open nature of its communication environment, illegitimate malicious UAVs (MUs) can infer the position of the source UAV (SU) by analyzing received signals, thus compromising the SU location privacy. To protect the SU location privacy while ensuring efficient communication with legitimate receiving UAVs (RUs), we propose an Active Reconfigurable Intelligent Surface (ARIS)-assisted covert communication scheme based on virtual partitioning and artificial noise (AN). Specifically, we design a novel ARIS architecture integrated with an AN module. This architecture dynamically partitions its reflecting elements into multiple sub-regions: one subset is optimized to enhance the communication rate between the SU and RUs, while the other subset generates AN to interfere with the localization of the SU by MUs. We first derive the Cramér-Rao Lower Bound (CRLB) for the localization with received signal strength (RSS), based on which, we establish a joint optimization framework for communication enhancement and localization interference. Subsequently, we derive and validate the optimal ARIS partitioning and power allocation under average channel conditions. Finally, tailored optimization methods are proposed for the reflection precoding and AN design of the two partitions. Simulation results validate that, compared to baseline schemes, the proposed scheme significantly increases the localization error of the SU by MUs while maintaining efficient communication between the SU and RUs, thereby effectively protecting the SU location privacy.

eess.SY↗