Searcharxiv⌕ Search

arXiv subjects

Zhitao Wang

Publications and source records attributed to Zhitao Wang.

At least 19 recordsLinked to original sources

Streaming Generative Video Compression with Embedding Alignment and In-Context Propagation

Video compression at ultra-low bitrates presents substantial challenges in preserving perceptual quality. Although pretrained video diffusion transformers (DiTs) provide powerful generative priors, existing generative video compression (GVC) methods built on them typically rely on computationally expensive VAE decoders to reconstruct video frames. Moreover, jointly refining multiple frames to exploit the inherent spatiotemporal modeling capabilities of video DiTs introduces additional buffering delay. To address these challenges, we propose SGVC, a streaming generative video compression framework that adapts a pretrained latent video DiT into a causal pixel-space refiner. Specifically, SGVC projects decoded features from a contextual pixel codec directly into the DiT input embedding space. To bridge the representation gap between codec-derived embeddings and pretrained DiT embeddings, we introduce Embedding Alignment, which aligns the projected embeddings with input embeddings extracted from the original frames during training. The refined DiT hidden states are then mapped directly to the pixel domain through a lightweight reconstruction path, bypassing the pretrained VAE decoder. To further exploit the spatiotemporal priors of the video DiT while maintaining causal decoding, we introduce In-Context Propagation, which reuses layer-wise K/V states across frames to exploit historical generative context, improving temporal consistency without requiring clip-wise buffering. Extensive experiments demonstrate that SGVC achieves competitive rate-perception performance, with average BD-rate savings of 71.97% over S$^2$VC in terms of DISTS. Project Page: https://chigland.github.io/SGVC/

eess.IV↗

From Ripple to Boundary: Deterministic Impact Boundaries for LLM-Based Software Evolution

The landscape of software evolution is undergoing a fundamental paradigm shift in the era of Large Language Models. As industrial software engineering increasingly transitions toward design-driven code generation, the bottleneck of evolution has shifted from writing code to maintaining deterministic consistency between multi-source design assets and the generated code. When a design asset changes, LLM agents currently lack a deterministic impact boundary to identify which code must be evolved, leading to two failure modes: hallucinating missing changes or breaking unrelated code. Probabilistic approaches, such as semantic retrieval or text-based diffing, fail to capture the structural ripple effects across highly interconnected design dependencies. We propose a paradigm shift from probabilistic semantic retrieval to deterministic structural constraints. We formalize the design assets as a Knowledge Graph and organize them into a content-addressed Merkle Directed Acyclic Graph. Our key insight reveals a topological isomorphism between the logical ripple effect of a design change and the content hash propagation within the DAG. By leveraging this, we transform ambiguous impact analysis into verifiable structural operations. Through a two-phase diffing process, we extract a precise causal chain that defines the exact modification boundary. This boundary enables deterministic context injection, providing the LLM agent with strictly scoped code and design updates, thereby fundamentally guaranteeing evolution safety. As a by-product, the KG serves as an agent-queryable knowledge base for natural language design retrieval. Early deployment on a large-scale industrial payment platform demonstrates that our mechanism successfully computes the exact impact boundary and generates incremental code modifications that meet the expected design changes.

cs.SE↗

Training an MoE LLM in MXFP4: How We Closed the Gap to BF16

Training large language models (LLMs) in sub-8-bit arithmetic promises substantial gains in memory bandwidth and compute throughput, yet FP4 formats introduce quantization error severe enough to destabilize optimization if left unaddressed. We present ALTO, an open-source training recipe for GPT-OSS-20B, a 20-billion-parameter mixture-of-experts (MoE) model, on the MLPerf Small MoE Training benchmark under the MXFP4 microscaling format. The recipe combines established techniques from Abecassis, Felix et al. [arXiv:2509.25149] (hybrid 1D/2D block quantization, randomized Hadamard transforms, and stochastic rounding) with a simplified weight de-oscillation scheme adapted from TetraJet-v2 [arXiv:2510.27527]. On C4 validation at global batch size 16, MXFP4 + RHT + SR + de-oscillation reaches a validation loss of 3.3350 at 16,128 steps, only 0.007 above the BF16 baseline (3.3283), closing roughly 50% of the gap left by MXFP4 without de-oscillation (3.3418). Measured by steps to reach the MLPerf quality target (validation loss 3.34), de-oscillation cuts the FP4 convergence overhead relative to BF16 from +20.0% to +5.0%. We also report negative end-to-end results from differential gradient estimation, outlier clipping, and macro-block scaling, despite each showing operator-level SNR gains. All experiments use simulated MXFP4 kernels on AMD MI300X hardware and wall-clock speed has not been optimized.

cs.AI↗

Harnessing Agentic Evolution

Agentic evolution has emerged as a powerful paradigm for improving programs, workflows, and scientific solutions by iteratively generating candidates, evaluating them, and using feedback to guide future search. However, existing methods are typically instantiated either as fixed hand-designed procedures that are modular but rigid, or as general-purpose agents that flexibly integrate feedback but can drift in long-horizon evolution. Both forms accumulate rich evidence over time, including candidates, feedback, traces, and failures, yet lack a stable interface for organizing this evidence and revising the mechanism that drives future evolution. We address this limitation by formulating agentic evolution as an interactive environment, where the accumulated evolution context serves as a process-level state. We introduce AEvo, a harnessed meta-editing framework in which a meta-agent observes this state and acts not by directly proposing the next candidate, but by editing the procedure or agent context that controls future evolution. This unified interface enables AEvo to steer both procedure-based and agent-based evolution, making accumulated evidence actionable for long-horizon search. Empirical evaluations on agentic and reasoning benchmarks show that AEvo outperforms five evolution baselines, achieving a 26 relative improvement over the strongest baseline. Across three open-ended optimization tasks, AEvo further outperforms four evolution baselines and achieves state-of-the-art performance under the same iteration budget.

cs.AI↗

NTIRE 2026 Challenge on Bitstream-Corrupted Video Restoration: Methods and Results

This paper reports on the NTIRE 2026 Challenge on Bitstream-Corrupted Video Restoration (BSCVR). The challenge aims to advance research on recovering visually coherent videos from corrupted bitstreams, whose decoding often produces severe spatial-temporal artifacts and content distortion. Built upon recent progress in bitstream-corrupted video recovery, the challenge provides a common benchmark for evaluating restoration methods under realistic corruption settings. We describe the dataset, evaluation protocol, and participating methods, and summarize the final results and main technical trends. The challenge highlights the difficulty of this emerging task and provides useful insights for future research on robust video restoration under practical bitstream corruption.

cs.CV↗

Enhancing End-to-End Autonomous Driving with Risk Semantic Distillaion from VLM

The autonomous driving (AD) system has exhibited remarkable performance in complex driving scenarios. However, generalization is still a key limitation for the current system, which refers to the ability to handle unseen scenarios or unfamiliar sensor configurations.Related works have explored the use of Vision-Language Models (VLMs) to address few-shot or zero-shot tasks. While promising, these methods introduce a new challenge: the emergence of a hybrid AD system, where two distinct systems are used to plan a trajectory, leading to potential inconsistencies. Alternative research directions have explored Vision-Language-Action (VLA) frameworks that generate control actions from VLM directly. However, these end-to-end solutions demonstrate prohibitive computational demands. To overcome these challenges, we introduce Risk Semantic Distillation (RSD), a novel framework that leverages VLMs to enhance the training of End-to-End (E2E) AD backbones. By providing risk attention for key objects, RSD addresses the issue of generalization. Specifically, we introduce RiskHead, a plug-in module that distills causal risk estimates from Vision-Language Models into Bird's-Eye-View (BEV) features, yielding interpretable risk-attention maps.This approach allows BEV features to learn richer and more nuanced risk attention representations, which directly enhance the model's ability to handle spatial boundaries and risky objects.By focusing on risk attention, RSD aligns better with human-like driving behavior, which is essential to navigate in complex and dynamic environments. Our experiments on the Bench2Drive benchmark demonstrate the effectiveness of RSD in managing complex and unpredictable driving conditions. Due to the enhanced BEV representations enabled by RSD, we observed a significant improvement in both perception and planning capabilities.

cs.CV↗

T-GVC: Trajectory-Guided Generative Video Coding at Ultra-Low Bitrates

Recent advances in video generation techniques have given rise to an emerging paradigm of generative video coding for Ultra-Low Bitrate (ULB) scenarios by leveraging powerful generative priors. However, most existing methods are limited by domain specificity (e.g., facial or human videos) or excessive dependence on high-level text guidance, which tend to inadequately capture fine-grained motion details, leading to unrealistic or incoherent reconstructions. To address these challenges, we propose Trajectory-Guided Generative Video Coding (dubbed T-GVC), a novel framework that bridges low-level motion tracking with high-level semantic understanding. T-GVC features a semantic-aware sparse motion sampling pipeline that extracts pixel-wise motion as sparse trajectory points based on their semantic importance, significantly reducing the bitrate while preserving critical temporal semantic information. In addition, by integrating trajectory-aligned loss constraints into diffusion processes, we introduce a training-free guidance mechanism in latent space to ensure physically plausible motion patterns without sacrificing the inherent capabilities of generative models. Experimental results demonstrate that T-GVC outperforms both traditional and neural video codecs under ULB conditions. Furthermore, additional experiments confirm that our framework achieves more precise motion control than existing text-guided methods, paving the way for a novel direction of generative video coding guided by geometric motion modeling.

cs.CV↗

Data Dependency-Aware Code Generation from Enhanced UML Sequence Diagrams

Large language models (LLMs) excel at generating code from natural language (NL) descriptions. However, the plain textual descriptions are inherently ambiguous and often fail to capture complex requirements like intricate system behaviors, conditional logic, and architectural constraints; implicit data dependencies in service-oriented architectures are difficult to infer and handle correctly. To bridge this gap, we propose a novel step-by-step code generation framework named UML2Dep by leveraging unambiguous formal specifications of complex requirements. First, we introduce an enhanced Unified Modeling Language (UML) sequence diagram tailored for service-oriented architectures. This diagram extends traditional visual syntax by integrating decision tables and API specifications, explicitly formalizing structural relationships and business logic flows in service interactions to rigorously eliminate linguistic ambiguity. Second, recognizing the critical role of data flow, we introduce a dedicated data dependency inference (DDI) task. DDI systematically constructs an explicit data dependency graph prior to actual code synthesis. To ensure reliability, we formalize DDI as a constrained mathematical reasoning task through novel prompting strategies, aligning with LLMs' excellent mathematical strengths. Additional static parsing and dependency pruning further reduce context complexity and cognitive load associated with intricate specifications, thereby enhancing reasoning accuracy and efficiency.

cs.AI↗

ReSeFlow: Rectifying SE(3)-Equivariant Policy Learning Flows

Robotic manipulation in unstructured environments requires the generation of robust and long-horizon trajectory-level policy with conditions of perceptual observations and benefits from the advantages of SE(3)-equivariant diffusion models that are data-efficient. However, these models suffer from the inference time costs. Inspired by the inference efficiency of rectified flows, we introduce the rectification to the SE(3)-diffusion models and propose the ReSeFlow, i.e., Rectifying SE(3)-Equivariant Policy Learning Flows, providing fast, geodesic-consistent, least-computational policy generation. Crucially, both components employ SE(3)-equivariant networks to preserve rotational and translational symmetry, enabling robust generalization under rigid-body motions. With the verification on the simulated benchmarks, we find that the proposed ReSeFlow with only one inference step can achieve better performance with lower geodesic distance than the baseline methods, achieving up to a 48.5% error reduction on the painting task and a 21.9% reduction on the rotating triangle task compared to the baseline's 100-step inference. This method takes advantages of both SE(3) equivariance and rectified flow and puts it forward for the real-world application of generative policy learning models with the data and inference efficiency.

cs.RO↗

Hybrid Perception and Equivariant Diffusion for Robust Multi-Node Rebar Tying

Rebar tying is a repetitive but critical task in reinforced concrete construction, typically performed manually at considerable ergonomic risk. Recent advances in robotic manipulation hold the potential to automate the tying process, yet face challenges in accurately estimating tying poses in congested rebar nodes. In this paper, we introduce a hybrid perception and motion planning approach that integrates geometry-based perception with Equivariant Denoising Diffusion on SE(3) (Diffusion-EDFs) to enable robust multi-node rebar tying with minimal training data. Our perception module utilizes density-based clustering (DBSCAN), geometry-based node feature extraction, and principal component analysis (PCA) to segment rebar bars, identify rebar nodes, and estimate orientation vectors for sequential ranking, even in complex, unstructured environments. The motion planner, based on Diffusion-EDFs, is trained on as few as 5-10 demonstrations to generate sequential end-effector poses that optimize collision avoidance and tying efficiency. The proposed system is validated on various rebar meshes, including single-layer, multi-layer, and cluttered configurations, demonstrating high success rates in node detection and accurate sequential tying. Compared with conventional approaches that rely on large datasets or extensive manual parameter tuning, our method achieves robust, efficient, and adaptable multi-node tying while significantly reducing data requirements. This result underscores the potential of hybrid perception and diffusion-driven planning to enhance automation in on-site construction tasks, improving both safety and labor efficiency.

cs.RO↗

Text-to-SQL for Enterprise Data Analytics

The introduction of large language models has brought rapid progress on Text-to-SQL benchmarks, but it is not yet easy to build a working enterprise solution. In this paper, we present insights from building an internal chatbot that enables LinkedIn's product managers, engineers, and operations teams to self-serve data insights from a large, dynamic data lake. Our approach features three components. First, we construct a knowledge graph that captures up-to-date semantics by indexing database metadata, historical query logs, wikis, and code. We apply clustering to identify relevant tables for each team or product area. Second, we build a Text-to-SQL agent that retrieves and ranks context from the knowledge graph, writes a query, and automatically corrects hallucinations and syntax errors. Third, we build an interactive chatbot that supports various user intents, from data discovery to query writing to debugging, and displays responses in rich UI elements to encourage follow-up chats. Our chatbot has over 300 weekly users. Expert review shows that 53% of its responses are correct or close to correct on an internal benchmark set. Through ablation studies, we identify the most important knowledge graph and modeling components, offering a practical path for developing enterprise Text-to-SQL solutions.

cs.CL↗

RL-OGM-Parking: Lidar OGM-Based Hybrid Reinforcement Learning Planner for Autonomous Parking

Autonomous parking has become a critical application in automatic driving research and development. Parking operations often suffer from limited space and complex environments, requiring accurate perception and precise maneuvering. Traditional rule-based parking algorithms struggle to adapt to diverse and unpredictable conditions, while learning-based algorithms lack consistent and stable performance in various scenarios. Therefore, a hybrid approach is necessary that combines the stability of rule-based methods and the generalizability of learning-based methods. Recently, reinforcement learning (RL) based policy has shown robust capability in planning tasks. However, the simulation-to-reality (sim-to-real) transfer gap seriously blocks the real-world deployment. To address these problems, we employ a hybrid policy, consisting of a rule-based Reeds-Shepp (RS) planner and a learning-based reinforcement learning (RL) planner. A real-time LiDAR-based Occupancy Grid Map (OGM) representation is adopted to bridge the sim-to-real gap, leading the hybrid policy can be applied to real-world systems seamlessly. We conducted extensive experiments both in the simulation environment and real-world scenarios, and the result demonstrates that the proposed method outperforms pure rule-based and learning-based methods. The real-world experiment further validates the feasibility and efficiency of the proposed method.

cs.RO↗

A Field Calibration Approach for Triaxial MEMS Gyroscopes Based on Gravity and Rotation Consistency

This paper developed an efficient method for calibrating triaxial MEMS gyroscopes, which can be effectively utilized in the field environment. The core strategy is to utilize the criterion that the dot product of the measured gravity and the rotation speed in a fixed frame remains constant. To eliminate the impact of external acceleration, the calibration process involves separate procedures for measuring local gravity and rotation speed. Moreover, unlike existing approaches for auto calibration of triaxial sensors that often result in nonlinear optimization problems, the proposed method simplifies the estimation of the gyroscope scale factor by employing a linear least squares algorithm. Extensive numerical simulations have been conducted to analyze the proposed method's performance in calibrating the six-parameter triaxial gyroscope model, taking into consideration measurements corrupted by simulated noise. Experimental validation was also carried out using two commercially available MEMS inertial measurement units (LSM9DS1) and a servo motor. The experimental results effectively demonstrate the efficacy of the proposed calibration approach.

cs.RO↗

On-site scale factor linearity calibration of MEMS triaxial gyroscopes

The calibration of MEMS triaxial gyroscopes is crucial for achieving precise attitude estimation for various wearable health monitoring applications. However, gyroscope calibration poses greater challenges compared to accelerometers and magnetometers. This paper introduces an efficient method for calibrating MEMS triaxial gyroscopes via only a servo motor, making it well-suited for field environments. The core strategy of the method involves utilizing the fact that the dot product of the measured gravity and the rotational speed in a fixed frame remains constant. To eliminate the influence of rotating centrifugal force on the accelerometer, the accelerometer data is measured while stationary. The proposed calibration experiment scheme, which allows gyroscopic measurements when operating each axis at a specific rotation speed, making it easier to evaluate the linearity across a related speed range constituted by a series of rotation speeds. Moreover, solely the classical least squares algorithm proves adequate for estimating the scale factor, notably streamlining the analysis of the calibration process. Extensive numerical simulations were conducted to analyze the proposed method's performance in calibrating a triaxial gyroscope model. Experimental validation was also carried out using a commercially available MEMS inertial measurement unit (LSM9DS1 from Arduino nano 33 BLE SENSE) and a servo motor capable of controlling precise speed. The experimental results effectively demonstrate the efficacy of the proposed calibration approach.

cs.RO↗

XUAT-Copilot: Multi-Agent Collaborative System for Automated User Acceptance Testing with Large Language Model

In past years, we have been dedicated to automating user acceptance testing (UAT) process of WeChat Pay, one of the most influential mobile payment applications in China. A system titled XUAT has been developed for this purpose. However, there is still a human-labor-intensive stage, i.e, test scripts generation, in the current system. Therefore, in this paper, we concentrate on methods of boosting the automation level of the current system, particularly the stage of test scripts generation. With recent notable successes, large language models (LLMs) demonstrate significant potential in attaining human-like intelligence and there has been a growing research area that employs LLMs as autonomous agents to obtain human-like decision-making capabilities. Inspired by these works, we propose an LLM-powered multi-agent collaborative system, named XUAT-Copilot, for automated UAT. The proposed system mainly consists of three LLM-based agents responsible for action planning, state checking and parameter selecting, respectively, and two additional modules for state sensing and case rewriting. The agents interact with testing device, make human-like decision and generate action command in a collaborative way. The proposed multi-agent system achieves a close effectiveness to human testers in our experimental studies and gains a significant improvement of Pass@1 accuracy compared with single-agent architecture. More importantly, the proposed system has launched in the formal testing environment of WeChat Pay mobile app, which saves a considerable amount of manpower in the daily development work.

cs.AI↗

RotoGBML: Towards Out-of-Distribution Generalization for Gradient-Based Meta-Learning

Gradient-based meta-learning (GBML) algorithms are able to fast adapt to new tasks by transferring the learned meta-knowledge, while assuming that all tasks come from the same distribution (in-distribution, ID). However, in the real world, they often suffer from an out-of-distribution (OOD) generalization problem, where tasks come from different distributions. OOD exacerbates inconsistencies in magnitudes and directions of task gradients, which brings challenges for GBML to optimize the meta-knowledge by minimizing the sum of task gradients in each minibatch. To address this problem, we propose RotoGBML, a novel approach to homogenize OOD task gradients. RotoGBML uses reweighted vectors to dynamically balance diverse magnitudes to a common scale and uses rotation matrixes to rotate conflicting directions close to each other. To reduce overhead, we homogenize gradients with the features rather than the network parameters. On this basis, to avoid the intervention of non-causal features (e.g., backgrounds), we also propose an invariant self-information (ISI) module to extract invariant causal features (e.g., the outlines of objects). Finally, task gradients are homogenized based on these invariant causal features. Experiments show that RotoGBML outperforms other state-of-the-art methods on various few-shot image classification benchmarks.

cs.CV↗

Pairwise Learning for Neural Link Prediction

In this paper, we aim at providing an effective Pairwise Learning Neural Link Prediction (PLNLP) framework. The framework treats link prediction as a pairwise learning to rank problem and consists of four main components, i.e., neighborhood encoder, link predictor, negative sampler and objective function. The framework is flexible that any generic graph neural convolution or link prediction specific neural architecture could be employed as neighborhood encoder. For link predictor, we design different scoring functions, which could be selected based on different types of graphs. In negative sampler, we provide several sampling strategies, which are problem specific. As for objective function, we propose to use an effective ranking loss, which approximately maximizes the standard ranking metric AUC. We evaluate the proposed PLNLP framework on 4 link property prediction datasets of Open Graph Benchmark, including ogbl-ddi, ogbl-collab, ogbl-ppa and ogbl-ciation2. PLNLP achieves top 1 performance on ogbl-ddi and ogbl-collab, and top 2 performance on ogbl-ciation2 only with basic neural architecture. The performance demonstrates the effectiveness of PLNLP.

cs.LG↗

Reinforcement Learning based Negotiation-aware Motion Planning of Autonomous Vehicles

For autonomous vehicles integrating onto roadways with human traffic participants, it requires understanding and adapting to the participants' intention and driving styles by responding in predictable ways without explicit communication. This paper proposes a reinforcement learning based negotiation-aware motion planning framework, which adopts RL to adjust the driving style of the planner by dynamically modifying the prediction horizon length of the motion planner in real time adaptively w.r.t the event of a change in environment, typically triggered by traffic participants' switch of intents with different driving styles. The framework models the interaction between the autonomous vehicle and other traffic participants as a Markov Decision Process. A temporal sequence of occupancy grid maps are taken as inputs for RL module to embed an implicit intention reasoning. Curriculum learning is employed to enhance the training efficiency and the robustness of the algorithm. We applied our method to narrow lane navigation in both simulation and real world to demonstrate that the proposed method outperforms the common alternative due to its advantage in alleviating the social dilemma problem with proper negotiation skills.

cs.RO↗