arXiv subjects

Xun Zhou

Publications and source records attributed to Xun Zhou.

At least 37 records · Page 2Linked to original sources

arXiv2025

Virtual Width Networks

We introduce Virtual Width Networks (VWN), a framework that delivers the benefits of wider representations without incurring the quadratic cost of increasing the hidden size. VWN decouples representational width from backbone width, expanding the embedding space while keeping backbone compute nearly constant. In our large-scale experiment, an 8-times expansion accelerates optimization by over 2 times for next-token and 3 times for next-2-token prediction. The advantage amplifies over training as both the loss gap grows and the convergence-speedup ratio increases, showing that VWN is not only token-efficient but also increasingly effective with scale. Moreover, we identify an approximately log-linear scaling relation between virtual width and loss reduction, offering an initial empirical basis and motivation for exploring virtual-width scaling as a new dimension of large-model efficiency.

cs.LG↗

arXiv2025

Parallel Loop Transformer for Efficient Test-Time Computation Scaling

Large Language Models (LLMs) are powerful but often too slow and costly for real-world use during inference. Looped transformers save on parameters by reusing the same weights for multiple computational steps, or "loops." However, this approach has a major flaw: the loops run one after another, causing inference latency and memory requirements to increase with each added loop. This makes them impractical for fast applications. To solve this problem, we introduce the Parallel Loop Transformer (PLT). PLT is a new architecture that delivers the performance benefits of a deep, looped model but with the low latency of a standard, non-looped model. PLT works using two key techniques. First, Cross-Loop Parallelism (CLP) breaks the sequential dependency by computing different loops for different tokens at the same time, all within a single pass. Second, to prevent memory costs from growing, we use an Efficient Representation Enhancement strategy. This method shares the memory (KV cache) from the first loop with all other loops. It then uses a Gated Sliding-Window Attention (G-SWA) to combine this shared global information with local information, maintaining high accuracy. Our experiments show that PLT achieves the high accuracy of a traditional looped model but with almost no extra latency or memory cost compared to a standard transformer.

cs.CL↗

arXiv2025

GatePro: Parameter-Free Expert Selection Optimization for Mixture-of-Experts Models

Modern large language models leverage Mixture-of-Experts (MoE) architectures for efficient scaling, but face a critical challenge: functionally similar experts are often selected simultaneously, creating redundant computation and limiting effective model capacity. Existing auxiliary balance loss methods improve token distribution but fail to address the underlying expert diversity problem. We introduce GatePro, a novel parameter-free method that directly promotes expert selection diversity. GatePro identifies the most similar expert pairs and introduces localized competition mechanisms, preventing redundant expert co-activation while maintaining natural expert specialization. Our comprehensive evaluation demonstrates GatePro's effectiveness across model scales and benchmarks. Analysis demonstrates GatePro's ability to achieve enhanced expert diversity, where experts develop more distinct and complementary capabilities, avoiding functional redundancy. This approach can be deployed hot-swappable during any training phase without additional learnable parameters, offering a practical solution for improving MoE effectiveness.

cs.CL↗

arXiv2025

Paradoxes of the public sector productivity measurement

This paper critically investigates standard total factor productivity (TFP) measurement in the public sector, where output information is often incomplete or distorted. The analysis reveals fundamental paradoxes under three common output measurement conventions. When cost-based value added is used as the aggregate output, measured TFP may paradoxically decline as a result of genuine productivity-enhancing changes such as technical progress and improved allocative and scale efficiencies, as well as reductions in real input prices. We show that the same problems carry over to the situation where the aggregate output is constructed as the cost-share weighted index of outputs. In the case of distorted output prices, measured TFP may move independently of any productivity changes and instead reflect shifts in pricing mechanisms. Using empirical illustrations from the United Kingdom and Finland, we demonstrate that such distortions are not merely theoretical but are embedded in widely used public productivity statistics. We argue that public sector TFP measurement requires a shift away from cost-based aggregation of outputs and toward non-market valuation methods grounded in economic theory.

econ.GN↗

arXiv2025

Maximum Degree-Based Quasi-Clique Search via an Iterative Framework

Cohesive subgraph mining is a fundamental problem in graph theory with numerous real-world applications, such as social network analysis and protein-protein interaction modeling. Among various cohesive subgraphs, the $γ$-quasi-clique is widely studied for its flexibility in requiring each vertex to connect to at least a $γ$ proportion of other vertices in the subgraph. However, solving the maximum $γ$-quasi-clique problem is NP-hard and further complicated by the lack of the hereditary property, which makes designing efficient pruning strategies challenging. Existing algorithms, such as DDA and FastQC, either struggle with scalability or exhibit significant performance declines for small values of $γ$. In this paper, we propose a novel algorithm, IterQC, which reformulates the maximum $γ$-quasi-clique problem as a series of $k$-plex problems that possess the hereditary property. IterQC introduces a non-trivial iterative framework and incorporates two key optimization techniques: (1) the pseudo lower bound (pseudo LB) technique, which leverages information across iterations to improve the efficiency of branch-and-bound searches, and (2) the preprocessing technique that reduces problem size and unnecessary iterations. Extensive experiments demonstrate that IterQC achieves up to four orders of magnitude speedup and solves significantly more graph instances compared to state-of-the-art algorithms DDA and FastQC.

cs.SI↗

arXiv2025

Conda: Column-Normalized Adam for Training Large Language Models Faster

Large language models (LLMs) have demonstrated impressive generalization and emergent capabilities, yet their pre-training remains computationally expensive and sensitive to optimization dynamics. While Adam-based optimizers offer fast convergence by adapting learning rates coordinate-wise, recent studies reveal that their updates often suffer from poor spectral conditioning and low-rank structures, hindering efficiency. Muon addresses this issue via global spectral normalization but lacks the per-coordinate adaptivity of Adam. In this work, we propose Column-Normalized Adam (Conda), a novel optimizer that bridges the strengths of both approaches. Conda projects updates into an orthogonal subspace and applies column-wise second moment normalization based on the projected gradients, thereby achieving both improved spectral conditioning and maintaining coordinate-wise adaptivity. This design alleviates the spectral pathologies of Adam while preserving its fast convergence behavior. Extensive experiments on the LLaMA and GPT-2 series show that Conda consistently outperforms AdamW, Muon, and other baselines in pre-training. Remarkably, on the LLaMA series, Conda achieves 2-2.5 the convergence speed of AdamW, measured in both training steps and training time. Further ablations demonstrate its robustness under diverse training setups. These results collectively highlight Conda as an effective and broadly applicable optimizer for large-scale LLM training. The code is released on https://github.com/jie040109/Conda

cs.LG↗

arXiv2025

UniTraj: Learning a Universal Trajectory Foundation Model from Billion-Scale Worldwide Traces

Building a universal trajectory foundation model is a promising solution to address the limitations of existing trajectory modeling approaches, such as task specificity, regional dependency, and data sensitivity. Despite its potential, data preparation, pre-training strategy development, and architectural design present significant challenges in constructing this model. Therefore, we introduce UniTraj, a Universal Trajectory foundation model that aims to address these limitations through three key innovations. First, we construct WorldTrace, an unprecedented dataset of 2.45 million trajectories with billions of GPS points spanning 70 countries, providing the diverse geographic coverage essential for region-independent modeling. Second, we develop novel pre-training strategies--Adaptive Trajectory Resampling and Self-supervised Trajectory Masking--that enable robust learning from heterogeneous trajectory data with varying sampling rates and quality. Finally, we tailor a flexible model architecture to accommodate a variety of trajectory tasks, effectively capturing complex movement patterns to support broad applicability. Extensive experiments across multiple tasks and real-world datasets demonstrate that UniTraj consistently outperforms existing methods, exhibiting superior scalability, adaptability, and generalization, with WorldTrace serving as an ideal yet non-exclusive training resource.

cs.ET↗

arXiv2025

TranTac: Leveraging Transient Tactile Signals for Contact-Rich Robotic Manipulation

Robotic manipulation tasks such as inserting a key into a lock or plugging a USB device into a port can fail when visual perception is insufficient to detect misalignment. In these situations, touch sensing is crucial for the robot to monitor the task's states and make precise, timely adjustments. Current touch sensing solutions are either insensitive to detect subtle changes or demand excessive sensor data. Here, we introduce TranTac, a data-efficient and low-cost tactile sensing and control framework that integrates a single contact-sensitive 6-axis inertial measurement unit within the elastomeric tips of a robotic gripper for completing fine insertion tasks. Our customized sensing system can detect dynamic translational and torsional deformations at the micrometer scale, enabling the tracking of visually imperceptible pose changes of the grasped object. By leveraging transformer-based encoders and diffusion policy, TranTac can imitate human insertion behaviors using transient tactile cues detected at the gripper's tip during insertion processes. These cues enable the robot to dynamically control and correct the 6-DoF pose of the grasped object. When combined with vision, TranTac achieves an average success rate of 79% on object grasping and insertion tasks, outperforming both vision-only policy and the one augmented with end-effector 6D force/torque sensing. Contact localization performance is also validated through tactile-only misaligned insertion tasks, achieving an average success rate of 88%. We assess the generalizability by training TranTac on a single prism-slot pair and testing it on unseen data, including a USB plug and a metal key, and find that the insertion tasks can still be completed with an average success rate of nearly 70%. The proposed framework may inspire new robotic tactile sensing systems for delicate manipulation tasks.

cs.RO↗

arXiv2025

MARS: Unleashing the Power of Variance Reduction for Training Large Models

Training deep neural networks--and more recently, large models demands efficient and scalable optimizers. Adaptive gradient algorithms like Adam, AdamW, and their variants have been central to this task. Despite the development of numerous variance reduction algorithms in the past decade aimed at accelerating stochastic optimization in both convex and nonconvex settings, variance reduction has not found widespread success in training deep neural networks or large language models. Consequently, it has remained a less favored approach in modern AI. In this paper, to unleash the power of variance reduction for efficient training of large models, we propose a unified optimization framework, MARS (Make vAriance Reduction Shine), which reconciles preconditioned gradient methods with variance reduction via a scaled stochastic recursive momentum technique. Within our framework, we introduce three instances of MARS that leverage preconditioned gradient updates based on AdamW, Lion, and Shampoo, respectively. We also draw a connection between our algorithms and existing optimizers. Experimental results on training GPT-2 models indicate that MARS consistently outperforms AdamW by a large margin. The implementation of MARS is available at https://github.com/AGI-Arena/MARS.

cs.LG↗

arXiv2025

IndexTTS2: A Breakthrough in Emotionally Expressive and Duration-Controlled Auto-Regressive Zero-Shot Text-to-Speech

Existing autoregressive large-scale text-to-speech (TTS) models have advantages in speech naturalness, but their token-by-token generation mechanism makes it difficult to precisely control the duration of synthesized speech. This becomes a significant limitation in applications requiring strict audio-visual synchronization, such as video dubbing. This paper introduces IndexTTS2, which proposes a novel, general, and autoregressive model-friendly method for speech duration control. The method supports two generation modes: one explicitly specifies the number of generated tokens to precisely control speech duration; the other freely generates speech in an autoregressive manner without specifying the number of tokens, while faithfully reproducing the prosodic features of the input prompt. Furthermore, IndexTTS2 achieves disentanglement between emotional expression and speaker identity, enabling independent control over timbre and emotion. In the zero-shot setting, the model can accurately reconstruct the target timbre (from the timbre prompt) while perfectly reproducing the specified emotional tone (from the style prompt). To enhance speech clarity in highly emotional expressions, we incorporate GPT latent representations and design a novel three-stage training paradigm to improve the stability of the generated speech. Additionally, to lower the barrier for emotional control, we designed a soft instruction mechanism based on text descriptions by fine-tuning Qwen3, effectively guiding the generation of speech with the desired emotional orientation. Finally, experimental results on multiple datasets show that IndexTTS2 outperforms state-of-the-art zero-shot TTS models in terms of word error rate, speaker similarity, and emotional fidelity. Audio samples are available at: https://index-tts.github.io/index-tts2.github.io/

cs.CL↗

arXiv2025

Balanced Actor Initialization: Stable RLHF Training of Distillation-Based Reasoning Models

The development of alignment and reasoning capabilities in large language models has seen remarkable progress through two paradigms: instruction tuning and reinforcement learning from human feedback (RLHF) alignment paradigm, and distillation-based reasoning fine-tuning paradigm. While both approaches prove effective independently, the third paradigm of applying RLHF to distillation-trained models presents significant challenges. Our investigation reveals two critical phenomena that emerge in this paradigm: Sequence Length Collapse, where language generation dramatically reduces during early RLHF training, and the Reward Hockey Stick Curve, featuring severe reward score drops followed by gradual recovery. These instabilities fundamentally compromise the model's alignment and reasoning capabilities. To address these challenges, we propose Balanced Actor Initialization (BAI), a two-stage weighted model merging approach. BAI first merges instruction-following and distillation-based reasoning fine-tuned models, then further combines this intermediate model with the pretrained model to preserve foundational knowledge. Through comprehensive experiments across diverse benchmarks and detailed analysis of training experiments, we demonstrate that BAI resolves Sequence Length Collapse, mitigates the Reward Hockey Stick Curve, and enables continuous sequence length improvement during training. Additionally, our analysis reveals that balanced merging ratios achieve optimal trade-offs between training stability and reasoning capability preservation. Our work provides the effective solution for stable training in this third paradigm, enabling more capable reasoning models that combine distillation efficiency with RLHF alignment.

cs.CL↗

arXiv2025

UltraMemV2: Memory Networks Scaling to 120B Parameters with Superior Long-Context Learning

While Mixture of Experts (MoE) models achieve remarkable efficiency by activating only subsets of parameters, they suffer from high memory access costs during inference. Memory-layer architectures offer an appealing alternative with very few memory access, but previous attempts like UltraMem have only matched the performance of 2-expert MoE models, falling significantly short of state-of-the-art 8-expert configurations. We present UltraMemV2, a redesigned memory-layer architecture that closes this performance gap. Our approach introduces five key improvements: integrating memory layers into every transformer block, simplifying value expansion with single linear projections, adopting FFN-based value processing from PEER, implementing principled parameter initialization, and rebalancing memory-to-FFN computation ratios. Through extensive evaluation, we demonstrate that UltraMemV2 achieves performance parity with 8-expert MoE models under same computation and parameters but significantly low memory access. Notably, UltraMemV2 shows superior performance on memory-intensive tasks, with improvements of +1.6 points on long-context memorization, +6.2 points on multi-round memorization, and +7.9 points on in-context learning. We validate our approach at scale with models up to 2.5B activated parameters from 120B total parameters, and establish that activation density has greater impact on performance than total sparse parameter count. Our work brings memory-layer architectures to performance parity with state-of-the-art MoE models, presenting a compelling alternative for efficient sparse computation.

cs.LG↗

arXiv2025

Maximum Score Routing For Mixture-of-Experts

Routing networks in sparsely activated mixture-of-experts (MoE) dynamically allocate input tokens to top-k experts through differentiable sparse transformations, enabling scalable model capacity while preserving computational efficiency. Traditional MoE networks impose an expert capacity constraint to ensure GPU-friendly computation. However, this leads to token dropping when capacity is saturated and results in low hardware efficiency due to padding in underutilized experts. Removing the capacity constraint, in turn, compromises load balancing and computational efficiency. To address these issues, we propose Maximum Score Routing ($\mathbf{MaxScore}$), a novel MoE routing paradigm that models routing as a minimum-cost maximum-flow problem and integrates a SoftTopk operator. MaxScore resolves the fundamental limitations of iterative rerouting and optimal transport formulations, achieving lower training losses and higher evaluation scores at equivalent FLOPs compared to both constrained and unconstrained baselines. Implementation details and experimental configurations can be obtained from $\href{https://github.com/dongbw18/MaxScore.git}{MaxScore}$.

cs.LG↗

arXiv2025

UW-3DGS: Underwater 3D Reconstruction with Physics-Aware Gaussian Splatting

Underwater 3D scene reconstruction faces severe challenges from light absorption, scattering, and turbidity, which degrade geometry and color fidelity in traditional methods like Neural Radiance Fields (NeRF). While NeRF extensions such as SeaThru-NeRF incorporate physics-based models, their MLP reliance limits efficiency and spatial resolution in hazy environments. We introduce UW-3DGS, a novel framework adapting 3D Gaussian Splatting (3DGS) for robust underwater reconstruction. Key innovations include: (1) a plug-and-play learnable underwater image formation module using voxel-based regression for spatially varying attenuation and backscatter; and (2) a Physics-Aware Uncertainty Pruning (PAUP) branch that adaptively removes noisy floating Gaussians via uncertainty scoring, ensuring artifact-free geometry. The pipeline operates in training and rendering stages. During training, noisy Gaussians are optimized end-to-end with underwater parameters, guided by PAUP pruning and scattering modeling. In rendering, refined Gaussians produce clean Unattenuated Radiance Images (URIs) free from media effects, while learned physics enable realistic Underwater Images (UWIs) with accurate light transport. Experiments on SeaThru-NeRF and UWBundle datasets show superior performance, achieving PSNR of 27.604, SSIM of 0.868, and LPIPS of 0.104 on SeaThru-NeRF, with ~65% reduction in floating artifacts.

cs.CV↗

arXiv2025

RankMixer: Scaling Up Ranking Models in Industrial Recommenders

Recent progress on large language models (LLMs) has spurred interest in scaling up recommendation systems, yet two practical obstacles remain. First, training and serving cost on industrial Recommenders must respect strict latency bounds and high QPS demands. Second, most human-designed feature-crossing modules in ranking models were inherited from the CPU era and fail to exploit modern GPUs, resulting in low Model Flops Utilization (MFU) and poor scalability. We introduce RankMixer, a hardware-aware model design tailored towards a unified and scalable feature-interaction architecture. RankMixer retains the transformer's high parallelism while replacing quadratic self-attention with multi-head token mixing module for higher efficiency. Besides, RankMixer maintains both the modeling for distinct feature subspaces and cross-feature-space interactions with Per-token FFNs. We further extend it to one billion parameters with a Sparse-MoE variant for higher ROI. A dynamic routing strategy is adapted to address the inadequacy and imbalance of experts training. Experiments show RankMixer's superior scaling abilities on a trillion-scale production dataset. By replacing previously diverse handcrafted low-MFU modules with RankMixer, we boost the model MFU from 4.5\% to 45\%, and scale our ranking model parameters by 100x while maintaining roughly the same inference latency. We verify RankMixer's universality with online A/B tests across two core application scenarios (Recommendation and Advertisement). Finally, we launch 1B Dense-Parameters RankMixer for full traffic serving without increasing the serving cost, which improves user active days by 0.3\% and total in-app usage duration by 1.08\%.

cs.IR↗

arXiv2025

Orthogonality conditions for convex regression

Econometric identification generally relies on orthogonality conditions, which usually state that the random error term is uncorrelated with the explanatory variables. In convex regression, the orthogonality conditions for identification are unknown. Applying Lagrangian duality theory, we establish the sample orthogonality conditions for convex regression, including additive and multiplicative formulations of the regression model, with and without monotonicity and homogeneity constraints. We then propose a hybrid instrumental variable control function approach to mitigate the impact of potential endogeneity in convex regression. The superiority of the proposed approach is shown in a Monte Carlo study and examined in an empirical application to Chilean manufacturing data.

stat.ME↗

arXiv2025

Expert Race: A Flexible Routing Strategy for Scaling Diffusion Transformer with Mixture of Experts

Diffusion models have emerged as mainstream framework in visual generation. Building upon this success, the integration of Mixture of Experts (MoE) methods has shown promise in enhancing model scalability and performance. In this paper, we introduce Race-DiT, a novel MoE model for diffusion transformers with a flexible routing strategy, Expert Race. By allowing tokens and experts to compete together and select the top candidates, the model learns to dynamically assign experts to critical tokens. Additionally, we propose per-layer regularization to address challenges in shallow layer learning, and router similarity loss to prevent mode collapse, ensuring better expert utilization. Extensive experiments on ImageNet validate the effectiveness of our approach, showcasing significant performance gains while promising scaling properties.

cs.CV↗

arXiv2025

CENTAUR: Bridging the Impossible Trinity of Privacy, Efficiency, and Performance in Privacy-Preserving Transformer Inference

With the growing deployment of pre-trained models like Transformers on cloud platforms, privacy concerns about model parameters and inference data are intensifying. Existing Privacy-Preserving Transformer Inference (PPTI) frameworks face the "impossible trinity" of balancing privacy, efficiency, and performance: Secure Multi-Party Computation (SMPC)-based approaches ensure strong privacy but suffer from high computational overhead and performance losses; Conversely, permutation-based methods achieve near-plaintext efficiency and accuracy but compromise privacy by exposing sensitive model parameters and intermediate results. Bridging this gap with a single approach presents substantial challenges, motivating the introduction of CENTAUR, a groundbreaking PPTI framework that seamlessly integrates random permutations and SMPC to address the "impossible trinity". By designing efficient PPTI algorithms tailored to the structural properties of Transformer models, CENTAUR achieves an unprecedented balance among privacy, efficiency, and performance. Our experiments demonstrate CENTAUR's ability to resist diverse data reconstruction attacks, achieve plaintext-level inference accuracy, and boost inference speed by 5.0-30.4 times, unlocking new possibilities for secure and efficient AI deployment.

cs.LG↗