arXiv subjects

Yida Wang

Publications and source records attributed to Yida Wang.

At least 37 records · Page 2Linked to original sources

arXiv2026

AccelOpt: A Self-Improving LLM Agentic System for AI Accelerator Kernel Optimization

We present AccelOpt, a self-improving large language model (LLM) agentic system that autonomously optimizes kernels for emerging AI acclerators, eliminating the need for expert-provided hardware-specific optimization knowledge. AccelOpt explores the kernel optimization space through iterative generation, informed by an optimization memory that curates experiences and insights from previously encountered slow-fast kernel pairs. We build NKIBench, a new benchmark suite of AWS Trainium accelerator kernels with varying complexity extracted from real-world LLM workloads to evaluate the effectiveness of AccelOpt. Our evaluation confirms that AccelOpt's capability improves over time, boosting the average percentage of peak throughput from $49\%$ to $61\%$ on Trainium 1 and from $45\%$ to $59\%$ on Trainium 2 for NKIBench kernels. Moreover, AccelOpt is highly cost-effective: using open-source models, it matches the kernel improvements of Claude Sonnet 4 while being $26\times$ cheaper. The code is open-sourced at https://github.com/zhang677/AccelOpt.

cs.LG↗

arXiv2026

StreetForward: Perceiving Dynamic Street with Feedforward Causal Attention

Feedforward reconstruction is crucial for autonomous driving applications, where rapid scene reconstruction enables efficient utilization of large-scale driving datasets in closed-loop simulation and other downstream tasks, eliminating the need for time-consuming per-scene optimization. We present StreetForward, a pose-free and tracker-free feedforward framework for dynamic street reconstruction. Building upon the alternating attention mechanism from Visual Geometry Grounded Transformer (VGGT), we propose a simple yet effective temporal mask attention module that captures dynamic motion information from image sequences and produces motion-aware latent representations. Static content and dynamic instances are represented uniformly with 3D Gaussian Splatting, and are optimized jointly by cross-frame rendering with spatio-temporal consistency, allowing the model to infer per-pixel velocities and produce high-fidelity novel views at new poses and times. We train and evaluate our model on the Waymo Open Dataset, demonstrating superior performance on novel view synthesis and depth estimation compared to existing methods. Furthermore, zero-shot inference on CARLA and other datasets validates the generalization capability of our approach. More visualizations are available on our project page: https://streetforward.github.io.

cs.CV↗

arXiv2026

Mobile-GS: Real-time Gaussian Splatting for Mobile Devices

3D Gaussian Splatting (3DGS) has emerged as a powerful representation for high-quality rendering across a wide range of applications.However, its high computational demands and large storage costs pose significant challenges for deployment on mobile devices. In this work, we propose a mobile-tailored real-time Gaussian Splatting method, dubbed Mobile-GS, enabling efficient inference of Gaussian Splatting on edge devices. Specifically, we first identify alpha blending as the primary computational bottleneck, since it relies on the time-consuming Gaussian depth sorting process. To solve this issue, we propose a depth-aware order-independent rendering scheme that eliminates the need for sorting, thereby substantially accelerating rendering. Although this order-independent rendering improves rendering speed, it may introduce transparency artifacts in regions with overlapping geometry due to the scarcity of rendering order. To address this problem, we propose a neural view-dependent enhancement strategy, enabling more accurate modeling of view-dependent effects conditioned on viewing direction, 3D Gaussian geometry, and appearance attributes. In this way, Mobile-GS can achieve both high-quality and real-time rendering. Furthermore, to facilitate deployment on memory-constrained mobile platforms, we also introduce first-order spherical harmonics distillation, a neural vector quantization technique, and a contribution-based pruning strategy to reduce the number of Gaussian primitives and compress the 3D Gaussian representation with the assistance of neural networks. Extensive experiments demonstrate that our proposed Mobile-GS achieves real-time rendering and compact model size while preserving high visual quality, making it well-suited for mobile applications.

cs.CV↗

arXiv2026

TritonRL: Training LLMs to Think and Code Triton Without Cheating

The rapid evolution of Large Language Models (LLMs) has driven a growing demand for automated, high-performance system kernels to accelerate machine learning workloads. We introduce TritonRL, a domain-specialized 8B-scale LLM for Triton programming, trained via a novel reinforcement learning (RL) framework. While Triton synthesis faces unique challenges, including data scarcity and a high susceptibility to reward hacking, our approach enables robust kernel generation through two primary innovations. First, we implement a multi-layered verification system that provides high-fidelity reward signals, ensuring that generated kernels are both syntactically and functionally valid. Second, we propose Hierarchical Reward Decomposition (HRD), which decouples reinforcement for high-level reasoning and low-level implementation to resolve the credit assignment problem in long-sequence generation. Comprehensive evaluations on KernelBench demonstrate that TritonRL achieves state-of-the-art correctness and runtime speedup, outperforming concurrent Triton-specific models and matching the performance of frontier models with over 100B parameters. Our results highlight the effectiveness of hardware-aware RL paradigms in specialized domain adaptation.

cs.SE↗

arXiv2026

TTrace: Lightweight Error Checking and Diagnosis for Distributed Training

Distributed training is essential for scaling the training of large neural network models, such as large language models (LLMs), across thousands of GPUs. However, the complexity of distributed training programs makes them particularly prone to silent bugs, which do not produce explicit error signals but lead to incorrect training outcomes. Effectively detecting and localizing such silent bugs in distributed training is challenging. Common debugging practices based on monitoring training loss or gradient norm curves are indirect, inefficient, and provide no way to localize bugs. To address those challenges, we design and implement TTrace, the first systematic differential testing system for detecting and localizing silent bugs in distributed training. TTrace aligns intermediate tensors from distributed training with those from a trusted reference implementation. To properly compare the floating-point values in the corresponding tensors, we propose a novel mathematical analysis that provides a guideline for setting tolerances, enabling TTrace to distinguish bug-induced errors from numerical errors. Experimental results demonstrate that TTrace effectively detects 11 existing bugs and 3 new bugs in the widely used Megatron-LM framework, while requiring fewer than 10 lines of code changes. TTrace is effective in various training recipes, including low-precision recipes involving BF16 and FP8. Notably, a popular open-source training framework has already adopted the method proposed by TTrace in its development workflow.

cs.DC↗

arXiv2026

Block-Diagonal LoRA for Eliminating Communication Overhead in Tensor Parallel LoRA Serving

When serving a single base LLM with several different LoRA adapters simultaneously, the adapters cannot simply be merged with the base model's weights as the adapter swapping would create overhead and requests using different adapters could not be batched. Rather, the LoRA computations have to be separated from the base LLM computations, and in a multi-device setup the LoRA adapters can be sharded in a way that is well aligned with the base model's tensor parallel execution, as proposed in S-LoRA. However, the S-LoRA sharding strategy encounters some communication overhead, which may be small in theory, but can be large in practice. In this paper, we propose to constrain certain LoRA factors to be block-diagonal, which allows for an alternative way of sharding LoRA adapters that does not require any additional communication for the LoRA computations. We demonstrate in extensive experiments that our block-diagonal LoRA approach is similarly parameter efficient as standard LoRA (i.e., for a similar number of parameters it achieves similar downstream performance) and that it leads to significant end-to-end speed-up over S-LoRA. For example, when serving on eight A100 GPUs, we observe up to 1.79x (1.23x) end-to-end speed-up with 0.87x (1.74x) the number of adapter parameters for Llama-3.1-70B, and up to 1.63x (1.3x) end-to-end speed-up with 0.86x (1.73x) the number of adapter parameters for Llama-3.1-8B.

cs.LG↗

arXiv2026

InfiniDepth: Arbitrary-Resolution and Fine-Grained Depth Estimation with Neural Implicit Fields

Existing depth estimation methods are fundamentally limited to predicting depth on discrete image grids. Such representations restrict their scalability to arbitrary output resolutions and hinder the geometric detail recovery. This paper introduces InfiniDepth, which represents depth as neural implicit fields. Through a simple yet effective local implicit decoder, we can query depth at continuous 2D coordinates, enabling arbitrary-resolution and fine-grained depth estimation. To better assess our method's capabilities, we curate a high-quality 4K synthetic benchmark from five different games, spanning diverse scenes with rich geometric and appearance details. Extensive experiments demonstrate that InfiniDepth achieves state-of-the-art performance on both synthetic and real-world benchmarks across relative and metric depth estimation tasks, particularly excelling in fine-detail regions. It also benefits the task of novel view synthesis under large viewpoint shifts, producing high-quality results with fewer holes and artifacts.

cs.CV↗

arXiv2025

Generative Head-Mounted Camera Captures for Photorealistic Avatars

Enabling photorealistic avatar animations in virtual and augmented reality (VR/AR) has been challenging because of the difficulty of obtaining ground truth state of faces. It is physically impossible to obtain synchronized images from head-mounted cameras (HMC) sensing input, which has partial observations in infrared (IR), and an array of outside-in dome cameras, which have full observations that match avatars' appearance. Prior works relying on analysis-by-synthesis methods could generate accurate ground truth, but suffer from imperfect disentanglement between expression and style in their personalized training. The reliance of extensive paired captures (HMC and dome) for the same subject makes it operationally expensive to collect large-scale datasets, which cannot be reused for different HMC viewpoints and lighting. In this work, we propose a novel generative approach, Generative HMC (GenHMC), that leverages large unpaired HMC captures, which are much easier to collect, to directly generate high-quality synthetic HMC images given any conditioning avatar state from dome captures. We show that our method is able to properly disentangle the input conditioning signal that specifies facial expression and viewpoint, from facial appearance, leading to more accurate ground truth. Furthermore, our method can generalize to unseen identities, removing the reliance on the paired captures. We demonstrate these breakthroughs by both evaluating synthetic HMC images and universal face encoders trained from these new HMC-avatar correspondences, which achieve better data efficiency and state-of-the-art accuracy.

cs.CV↗

arXiv2025

DCP: Addressing Input Dynamism In Long-Context Training via Dynamic Context Parallelism

Context parallelism has emerged as a key technique to support long-context training, a growing trend in generative AI for modern large models. However, existing context parallel methods rely on static parallelization configurations that overlook the dynamic nature of training data, specifically, the variability in sequence lengths and token relationships (i.e., attention patterns) across samples. As a result, these methods often suffer from unnecessary communication overhead and imbalanced computation. In this paper, we present DCP, a dynamic context parallel training framework that introduces fine-grained blockwise partitioning of both data and computation. By enabling flexible mapping of data and computation blocks to devices, DCP can adapt to varying sequence characteristics, effectively reducing communication and improving memory and computation balance. Micro-benchmarks demonstrate that DCP accelerates attention by 1.19x~2.45x under causal masks and 2.15x~3.77x under sparse attention patterns. Additionally, we observe up to 0.94x~1.16x end-to-end training speed-up for causal masks, and 1.00x~1.46x for sparse masks.

cs.DC↗

arXiv2025

TASP: Topology-aware Sequence Parallelism

Long-context large language models (LLMs) face constraints due to the quadratic complexity of the self-attention mechanism. The mainstream sequence parallelism (SP) method, Ring Attention, attempts to solve this by distributing the query into multiple query chunks across accelerators and enable each Q tensor to access all KV tensors from other accelerators via the Ring AllGather communication primitive. However, it exhibits low communication efficiency, restricting its practical applicability. This inefficiency stems from the mismatch between the Ring AllGather communication primitive it adopts and the AlltoAll topology of modern accelerators. A Ring AllGather primitive is composed of iterations of ring-styled data transfer, which can only utilize a very limited fraction of an AlltoAll topology. Inspired by the Hamiltonian decomposition of complete directed graphs, we identify that modern accelerator topology can be decomposed into multiple orthogonal ring datapaths which can concurrently transfer data without interference. Based on this, we further observe that the Ring AllGather primitive can also be decomposed into the same number of concurrent ring-styled data transfer at every iteration. Based on these insights, we propose TASP, a topology-aware SP method for long-context LLMs that fully utilizes the communication capacity of modern accelerators via topology decomposition and primitive decomposition. Experimental results on both single-node and multi-node NVIDIA H100 systems and a single-node AMD MI300X system demonstrate that TASP achieves higher communication efficiency than Ring Attention on these modern accelerator topologies and achieves up to 3.58 speedup than Ring Attention and its variant Zigzag-Ring Attention. The code is available at https://github.com/infinigence/HamiltonAttention.

cs.LG↗

arXiv2025

Tilus: A Tile-Level GPGPU Programming Language for Low-Precision Computation

Serving Large Language Models (LLMs) is critical for AI-powered applications, yet it demands substantial computational resources, particularly in memory bandwidth and computational throughput. Low-precision computation has emerged as a key technique to improve efficiency while reducing resource consumption. Existing approaches for generating low-precision kernels are limited to weight bit widths that are powers of two and suffer from suboptimal performance because of high-level GPU programming abstractions. These abstractions restrict critical optimizations, such as fine-grained register management and optimized memory access patterns, that are essential for efficient low-precision computations. In this paper, we introduce Tilus, a domain-specific language designed for General-Purpose GPU (GPGPU) computing that supports low-precision data types with arbitrary bit widths from 1 to 8 while maintaining GPU programmability. Tilus features a thread-block-level programming model, a hierarchical memory space, a novel algebraic layout system, and extensive support for diverse low-precision data types. Tilus programs are compiled into highly efficient GPU programs through automatic vectorization and instruction selection. Extensive experiments demonstrate that Tilus efficiently supports a full spectrum of low-precision data types, and outperforms state-of-the-art low-precision kernels. Compared to existing compilers such as Triton and Ladder, as well as hand-optimized kernels such as QuantLLM and Marlin, Tilus achieves performance improvements of: $1.75\times$, $2.61\times$, $1.29\times$ and $1.03\times$, respectively. We open-source Tilus at https://github.com/NVIDIA/tilus.

cs.LG↗

arXiv2025

StreetCrafter: Street View Synthesis with Controllable Video Diffusion Models

This paper aims to tackle the problem of photorealistic view synthesis from vehicle sensor data. Recent advancements in neural scene representation have achieved notable success in rendering high-quality autonomous driving scenes, but the performance significantly degrades as the viewpoint deviates from the training trajectory. To mitigate this problem, we introduce StreetCrafter, a novel controllable video diffusion model that utilizes LiDAR point cloud renderings as pixel-level conditions, which fully exploits the generative prior for novel view synthesis, while preserving precise camera control. Moreover, the utilization of pixel-level LiDAR conditions allows us to make accurate pixel-level edits to target scenes. In addition, the generative prior of StreetCrafter can be effectively incorporated into dynamic scene representations to achieve real-time rendering. Experiments on Waymo Open Dataset and PandaSet demonstrate that our model enables flexible control over viewpoint changes, enlarging the view synthesis regions for satisfying rendering, which outperforms existing methods.

cs.CV↗

arXiv2025

A Physics-Driven Neural Network with Parameter Embedding for Generating Quantitative MR Maps from Weighted Images

We propose a deep learning-based approach that integrates MRI sequence parameters to improve the accuracy and generalizability of quantitative image synthesis from clinical weighted MRI. Our physics-driven neural network embeds MRI sequence parameters -- repetition time (TR), echo time (TE), and inversion time (TI) -- directly into the model via parameter embedding, enabling the network to learn the underlying physical principles of MRI signal formation. The model takes conventional T1-weighted, T2-weighted, and T2-FLAIR images as input and synthesizes T1, T2, and proton density (PD) quantitative maps. Trained on healthy brain MR images, it was evaluated on both internal and external test datasets. The proposed method achieved high performance with PSNR values exceeding 34 dB and SSIM values above 0.92 for all synthesized parameter maps. It outperformed conventional deep learning models in accuracy and robustness, including data with previously unseen brain structures and lesions. Notably, our model accurately synthesized quantitative maps for these unseen pathological regions, highlighting its superior generalization capability. Incorporating MRI sequence parameters via parameter embedding allows the neural network to better learn the physical characteristics of MR signals, significantly enhancing the performance and reliability of quantitative MRI synthesis. This method shows great potential for accelerating qMRI and improving its clinical utility.

eess.IV↗

arXiv2025

A Large Language Model Powered Integrated Circuit Footprint Geometry Understanding

Printed-Circuit-board (PCB) footprint geometry labeling of integrated circuits (IC) is essential in defining the physical interface between components and the PCB layout, requiring exceptional visual perception proficiency. However, due to the unstructured footprint drawing and abstract diagram annotations, automated parsing and accurate footprint geometry modeling remain highly challenging. Despite its importance, no methods currently exist for automated package geometry labeling directly from IC mechanical drawings. In this paper, we first investigate the visual perception performance of Large Multimodal Models (LMMs) when solving IC footprint geometry understanding. Our findings reveal that current LMMs severely suffer from inaccurate geometric perception, which hinders their performance in solving the footprint geometry labeling problem. To address these limitations, we propose LLM4-IC8K, a novel framework that treats IC mechanical drawings as images and leverages LLMs for structured geometric interpretation. To mimic the step-by-step reasoning approach used by human engineers, LLM4-IC8K addresses three sub-tasks: perceiving the number of pins, computing the center coordinates of each pin, and estimating the dimensions of individual pins. We present a two-stage framework that first trains LMMs on synthetically generated IC footprint diagrams to learn fundamental geometric reasoning and then fine-tunes them on real-world datasheet drawings to enhance robustness and accuracy in practical scenarios. To support this, we introduce ICGeo8K, a multi-modal dataset with 8,608 labeled samples, including 4138 hand-crafted IC footprint samples and 4470 synthetically generated samples. Extensive experiments demonstrate that our model outperforms state-of-the-art LMMs on the proposed benchmark.

cs.CV↗

arXiv2025

KVFlow: Efficient Prefix Caching for Accelerating LLM-Based Multi-Agent Workflows

Large language model (LLM) based agentic workflows have become a popular paradigm for coordinating multiple specialized agents to solve complex tasks. To improve serving efficiency, existing LLM systems employ prefix caching to reuse key-value (KV) tensors corresponding to agents' fixed prompts, thereby avoiding redundant computation across repeated invocations. However, current systems typically evict KV caches using a Least Recently Used (LRU) policy, which fails to anticipate future agent usage and often discards KV caches shortly before their reuse. This leads to frequent cache misses and substantial recomputation or swapping overhead. We present KVFlow, a workflow-aware KV cache management framework tailored for agentic workloads. KVFlow abstracts the agent execution schedule as an Agent Step Graph and assigns each agent a steps-to-execution value that estimates its temporal proximity to future activation. These values guide a fine-grained eviction policy at the KV node level, allowing KVFlow to preserve entries likely to be reused and efficiently manage shared prefixes in tree-structured caches. Moreover, KVFlow introduces a fully overlapped KV prefetching mechanism, which proactively loads required tensors from CPU to GPU in background threads for agents scheduled in the next step, thereby avoiding cache miss stalls during generation. Compared to SGLang with hierarchical radix cache, KVFlow achieves up to 1.83$\times$ speedup for single workflows with large prompts, and up to 2.19$\times$ speedup for scenarios with many concurrent workflows.

cs.DC↗

arXiv2025

HiNeuS: High-fidelity Neural Surface Mitigating Low-texture and Reflective Ambiguity

Neural surface reconstruction faces persistent challenges in reconciling geometric fidelity with photometric consistency under complex scene conditions. We present HiNeuS, a unified framework that holistically addresses three core limitations in existing approaches: multi-view radiance inconsistency, missing keypoints in textureless regions, and structural degradation from over-enforced Eikonal constraints during joint optimization. To resolve these issues through a unified pipeline, we introduce: 1) Differential visibility verification through SDF-guided ray tracing, resolving reflection ambiguities via continuous occlusion modeling; 2) Planar-conformal regularization via ray-aligned geometry patches that enforce local surface coherence while preserving sharp edges through adaptive appearance weighting; and 3) Physically-grounded Eikonal relaxation that dynamically modulates geometric constraints based on local radiance gradients, enabling detail preservation without sacrificing global regularity. Unlike prior methods that handle these aspects through sequential optimizations or isolated modules, our approach achieves cohesive integration where appearance-geometry constraints evolve synergistically throughout training. Comprehensive evaluations across synthetic and real-world datasets demonstrate state-of-the-art performance, including a 21.4% reduction in Chamfer distance over reflection-aware baselines and 2.32 dB PSNR improvement against neural rendering counterparts. Qualitative analyses reveal superior capability in recovering specular instruments, urban layouts with centimeter-scale infrastructure, and low-textured surfaces without local patch collapse. The method's generalizability is further validated through successful application to inverse rendering tasks, including material decomposition and view-consistent relighting.

cs.CV↗

arXiv2025

3DRealCar: An In-the-wild RGB-D Car Dataset with 360-degree Views

3D cars are commonly used in self-driving systems, virtual/augmented reality, and games. However, existing 3D car datasets are either synthetic or low-quality, limiting their applications in practical scenarios and presenting a significant gap toward high-quality real-world 3D car datasets. In this paper, we propose the first large-scale 3D real car dataset, termed 3DRealCar, offering three distinctive features. (1) \textbf{High-Volume}: 2,500 cars are meticulously scanned by smartphones, obtaining car images and point clouds with real-world dimensions; (2) \textbf{High-Quality}: Each car is captured in an average of 200 dense, high-resolution 360-degree RGB-D views, enabling high-fidelity 3D reconstruction; (3) \textbf{High-Diversity}: The dataset contains various cars from over 100 brands, collected under three distinct lighting conditions, including reflective, standard, and dark. Additionally, we offer detailed car parsing maps for each instance to promote research in car parsing tasks. Moreover, we remove background point clouds and standardize the car orientation to a unified axis for the reconstruction only on cars and controllable rendering without background. We benchmark 3D reconstruction results with state-of-the-art methods across different lighting conditions in 3DRealCar. Extensive experiments demonstrate that the standard lighting condition part of 3DRealCar can be used to produce a large number of high-quality 3D cars, improving various 2D and 3D tasks related to cars. Notably, our dataset brings insight into the fact that recent 3D reconstruction methods face challenges in reconstructing high-quality 3D cars under reflective and dark lighting conditions. \textcolor{red}{\href{https://xiaobiaodu.github.io/3drealcar/}{Our dataset is here.}}

cs.CV↗

arXiv2025

Self-supervised Latent Space Optimization with Nebula Variational Coding

Deep learning approaches process data in a layer-by-layer way with intermediate (or latent) features. We aim at designing a general solution to optimize the latent manifolds to improve the performance on classification, segmentation, completion and/or reconstruction through probabilistic models. This paper proposes a variational inference model which leads to a clustered embedding. We introduce additional variables in the latent space, called \textbf{nebula anchors}, that guide the latent variables to form clusters during training. To prevent the anchors from clustering among themselves, we employ the variational constraint that enforces the latent features within an anchor to form a Gaussian distribution, resulting in a generative model we refer as Nebula Variational Coding (NVC). Since each latent feature can be labeled with the closest anchor, we also propose to apply metric learning in a self-supervised way to make the separation between clusters more explicit. As a consequence, the latent variables of our variational coder form clusters which adapt to the generated semantic of the training data, \textit{e.g.} the categorical labels of each sample. We demonstrate experimentally that it can be used within different architectures designed to solve different problems including text sequence, images, 3D point clouds and volumetric data, validating the advantage of our proposed method.

cs.LG↗