arXiv subjects

Yanling Wang

Publications and source records attributed to Yanling Wang.

At least 19 recordsLinked to original sources

arXiv2026

HG-Bench: A Benchmark for Multi-Page Handwritten Answer-Region Grounding in Automated Homework Assessment

Automated homework assessment depends not only on recognizing student answers, but also on accurately locating where each answer and each intermediate reasoning step appears in noisy, multi-page handwritten work. This paper addresses the missing evaluation setting of page-aware, two-level answer-region grounding: given a sequence of homework page images, a model must localize complete answer regions and their ordered step-level subregions. We introduce HG-Bench, a benchmark of 500 human-annotated K-12 homework samples curated from a 1,489,278-image source pool, with question-level and step-level boxes linked by a hierarchical containment constraint. HG-Bench is paired with a page-aware evaluation protocol that separately measures complete-answer localization (FA) and step-level decomposition (FSm), revealing whether models truly ground the spatial structure of student reasoning rather than merely parse visible text. Across frontier closed-source APIs and competitive open-weight VLMs, no zero-shot system exceeds 55.22% on FA or 48.22% on FSm, while a GLM-4.6V 9B reference model fine-tuned on ~10k in-domain examples reaches 74.97/72.26. These results identify step-level handwritten grounding as a concrete capability gap and provide a reproducible benchmark, evaluation protocol, and trained reference point for future work on automated homework assessment.

cs.CV↗

arXiv2026

GLM-5V-Turbo: Toward a Native Foundation Model for Multimodal Agents

We present GLM-5V-Turbo, a step toward native foundation models for multimodal agents. As foundation models are increasingly deployed in real environments, agentic capability depends not only on language reasoning, but also on the ability to perceive, interpret, and act over heterogeneous contexts such as images, videos, webpages, documents, GUIs. GLM-5V-Turbo is built around this objective: multimodal perception is integrated as a core component of reasoning, planning, tool use, and execution, rather than as an auxiliary interface to a language model. This report summarizes the main improvements behind GLM-5V-Turbo across model design, multimodal training, reinforcement learning, toolchain expansion, and integration with agent frameworks. These developments lead to strong performance in multimodal coding, visual tool use, and framework-based agentic tasks, while preserving competitive text-only coding capability. More importantly, our development process offers practical insights for building multimodal agents, highlighting the central role of multimodal perception, hierarchical optimization, and reliable end-to-end verification.

cs.CV↗

arXiv2025

A Hybrid Approach for EMF Code Generation:Code Templates Meet Large Language Models

Template-based and LLM-based code generation are both key enablers of automated software development. The former provides correctness guarantees but are rigid for complex requirements, whereas LLMs offer high flexibility at the risk of producing faulty code.This paper proposes iEcoreGen, a hybrid approach that integrates Eclipse Modeling Framework (EMF) and LLMs. In EMF, an Ecore model defines a system structure and acts as a blueprint for code-generation.iEcoreGen decomposes requirements to derive operation specifications, uses EMF's template-based generator to produce initial Java code, and serializes specifications into docstrings. LLMs are then invoked to complete and fix unimplemented methods. We assessed iEcoreGen on twenty code-generation tasks across five LLMs. It surpasses LLM-only baselines on pass@k and performs on par with them on compilation@k. An ablation study clarified the contribution of each component of iEcoreGen. Overall, the findings indicate that LLM-enhanced model-driven development is a promising path toward more efficient software automation.

cs.SE↗

arXiv2025

GeoVLMath: Enhancing Geometry Reasoning in Vision-Language Models via Cross-Modal Reward for Auxiliary Line Creation

Auxiliary lines are essential for solving complex geometric problems but remain challenging for large vision-language models (LVLMs). Recent attempts construct auxiliary lines via code-driven rendering, a strategy that relies on accurate and executable code generation to produce visual renderings of the auxiliary lines for subsequent reasoning. However, in complex solid geometry settings, such a strong dependence on precise specifications substantially restricts the robustness of this strategy. Alternatively, we turn to a simpler and more stable solution, representing auxiliary-line constructions as structured textual descriptions. To bridge the gap between textual descriptions and spatial structure, we propose a reinforcement learning framework that enhances diagram-text alignment. The core is a cross-modal reward model that evaluates how well the generated auxiliary-line description matches the ground-truth auxiliary-line diagram. The reward signal drives a GRPO-based RL stage to yield informative auxiliary-line descriptions for the reasoning. To support the training and evaluation, we develop a scalable data pipeline and construct AuxSolidMath, a dataset of 3,018 real-exam geometry problems with paired diagrams and aligned textual fields. Based on this framework, we derive GeoVLMath, an LVLM for solving complex solid geometry.

cs.CV↗

arXiv2025

GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning

We present GLM-4.1V-Thinking, GLM-4.5V, and GLM-4.6V, a family of vision-language models (VLMs) designed to advance general-purpose multimodal understanding and reasoning. In this report, we share our key findings in the development of the reasoning-centric training framework. We first develop a capable vision foundation model with significant potential through large-scale pre-training, which arguably sets the upper bound for the final performance. We then propose Reinforcement Learning with Curriculum Sampling (RLCS) to unlock the full potential of the model, leading to comprehensive capability enhancement across a diverse range of tasks, including STEM problem solving, video understanding, content recognition, coding, grounding, GUI-based agents, and long document interpretation. In a comprehensive evaluation across 42 public benchmarks, GLM-4.5V achieves state-of-the-art performance on nearly all tasks among open-source models of similar size, and demonstrates competitive or even superior results compared to closed-source models such as Gemini-2.5-Flash on challenging tasks including Coding and GUI Agents. Meanwhile, the smaller GLM-4.1V-9B-Thinking remains highly competitive-achieving superior results to the much larger Qwen2.5-VL-72B on 29 benchmarks. We open-source both GLM-4.1V-9B-Thinking and GLM-4.5V. We further introduce the GLM-4.6V series, open-source multimodal models with native tool use and a 128K context window. A brief overview is available at https://z.ai/blog/glm-4.6v. Code, models and more information are released at https://github.com/zai-org/GLM-V.

cs.CV↗

arXiv2025

VisualSimpleQA: A Benchmark for Decoupled Evaluation of Large Vision-Language Models in Fact-Seeking Question Answering

Large vision-language models (LVLMs) have demonstrated remarkable achievements, yet the generation of non-factual responses remains prevalent in fact-seeking question answering (QA). Current multimodal fact-seeking benchmarks primarily focus on comparing model outputs to ground truth answers, providing limited insights into the performance of modality-specific modules. To bridge this gap, we introduce VisualSimpleQA, a multimodal fact-seeking benchmark with two key features. First, it enables streamlined and decoupled evaluation of LVLMs in visual and linguistic modalities. Second, it incorporates well-defined difficulty criteria to guide human annotation and facilitates the extraction of a challenging subset, VisualSimpleQA-hard. Experiments on 15 LVLMs show that even state-of-the-art models such as GPT-4o achieve merely 60%+ correctness in multimodal fact-seeking QA on VisualSimpleQA and 30%+ on VisualSimpleQA-hard. Furthermore, the decoupled evaluation across these models highlights substantial opportunities for improvement in both visual and linguistic modules. The dataset is available at https://huggingface.co/datasets/WYLing/VisualSimpleQA.

cs.CL↗

arXiv2024

P$^2$ Law: Scaling Law for Post-Training After Model Pruning

Pruning has become a widely adopted technique for reducing the hardware requirements of large language models (LLMs). To recover model performance after pruning, post-training is commonly employed to mitigate the resulting performance degradation. While post-training benefits from larger datasets, once the dataset size is already substantial, increasing the training data provides only limited performance gains. To balance post-training cost and model performance, it is necessary to explore the optimal amount of post-training data.Through extensive experiments on the Llama-3 and Qwen-2.5 series models, pruned using various common pruning methods, we uncover the scaling \textbf{Law} for \textbf{P}ost-training after model \textbf{P}runing, referred to as the P$^2$ Law.This law identifies four key factors for predicting the pruned model's post-training loss: the model size before pruning, the number of post-training tokens, the pruning rate, and the model's loss before pruning. Moreover, P$^2$ Law can generalize to larger dataset sizes, larger model sizes, and higher pruning rates, offering valuable insights for the post-training of pruned LLMs.

cs.AI↗

arXiv2024

A Learn-Then-Reason Model Towards Generalization in Knowledge Base Question Answering

Large-scale knowledge bases (KBs) like Freebase and Wikidata house millions of structured knowledge. Knowledge Base Question Answering (KBQA) provides a user-friendly way to access these valuable KBs via asking natural language questions. In order to improve the generalization capabilities of KBQA models, extensive research has embraced a retrieve-then-reason framework to retrieve relevant evidence for logical expression generation. These multi-stage efforts prioritize acquiring external sources but overlook the incorporation of new knowledge into their model parameters. In effect, even advanced language models and retrievers have knowledge boundaries, thereby limiting the generalization capabilities of previous KBQA models. Therefore, this paper develops KBLLaMA, which follows a learn-then-reason framework to inject new KB knowledge into a large language model for flexible end-to-end KBQA. At the core of KBLLaMA, we study (1) how to organize new knowledge about KBQA and (2) how to facilitate the learning of the organized knowledge. Extensive experiments on various KBQA generalization tasks showcase the state-of-the-art performance of KBLLaMA. Especially on the general benchmark GrailQA and domain-specific benchmark Bio-chemical, KBLLaMA respectively derives a performance gain of up to 3.8% and 9.8% compared to the baselines.

cs.CL↗

arXiv2024

FacLens: Transferable Probe for Foreseeing Non-Factuality in Fact-Seeking Question Answering of Large Language Models

Despite advancements in large language models (LLMs), non-factual responses still persist in fact-seeking question answering. Unlike extensive studies on post-hoc detection of these responses, this work studies non-factuality prediction (NFP), predicting whether an LLM will generate a non-factual response prior to the response generation. Previous NFP methods have shown LLMs' awareness of their knowledge, but they face challenges in terms of efficiency and transferability. In this work, we propose a lightweight model named Factuality Lens (FacLens), which effectively probes hidden representations of fact-seeking questions for the NFP task. Moreover, we discover that hidden question representations sourced from different LLMs exhibit similar NFP patterns, enabling the transferability of FacLens across different LLMs to reduce development costs. Extensive experiments highlight FacLens's superiority in both effectiveness and efficiency.

cs.CL↗

arXiv2024

SGSH: Stimulate Large Language Models with Skeleton Heuristics for Knowledge Base Question Generation

Knowledge base question generation (KBQG) aims to generate natural language questions from a set of triplet facts extracted from KB. Existing methods have significantly boosted the performance of KBQG via pre-trained language models (PLMs) thanks to the richly endowed semantic knowledge. With the advance of pre-training techniques, large language models (LLMs) (e.g., GPT-3.5) undoubtedly possess much more semantic knowledge. Therefore, how to effectively organize and exploit the abundant knowledge for KBQG becomes the focus of our study. In this work, we propose SGSH--a simple and effective framework to Stimulate GPT-3.5 with Skeleton Heuristics to enhance KBQG. The framework incorporates "skeleton heuristics", which provides more fine-grained guidance associated with each input to stimulate LLMs to generate optimal questions, encompassing essential elements like the question phrase and the auxiliary verb.More specifically, we devise an automatic data construction strategy leveraging ChatGPT to construct a skeleton training dataset, based on which we employ a soft prompting approach to train a BART model dedicated to generating the skeleton associated with each input. Subsequently, skeleton heuristics are encoded into the prompt to incentivize GPT-3.5 to generate desired questions. Extensive experiments demonstrate that SGSH derives the new state-of-the-art performance on the KBQG tasks.

cs.CL↗

arXiv2024

Streamlining Redundant Layers to Compress Large Language Models

This paper introduces LLM-Streamline, a pioneer work on layer pruning for large language models (LLMs). It is based on the observation that different layers have varying impacts on hidden states, enabling the identification of less important layers to be pruned.LLM-Streamline comprises two parts: layer pruning, which removes consecutive layers with the lowest importance based on target sparsity, and layer replacement, a novel module that trains a lightweight network to replace the pruned layers to mitigate performance loss. Additionally, a new metric called stability is proposed to address the limitations of the widely used accuracy metric in evaluating model compression. Experiments show that LLM-Streamline outperforms both previous and concurrent state-of-the-art pruning methods in terms of both performance and training efficiency.Our code is available at https://github.com/RUCKBReasoning/LLM-Streamline

cs.CL↗

arXiv2024

Open-World Semi-Supervised Learning for Node Classification

Open-world semi-supervised learning (Open-world SSL) for node classification, that classifies unlabeled nodes into seen classes or multiple novel classes, is a practical but under-explored problem in the graph community. As only seen classes have human labels, they are usually better learned than novel classes, and thus exhibit smaller intra-class variances within the embedding space (named as imbalance of intra-class variances between seen and novel classes). Based on empirical and theoretical analysis, we find the variance imbalance can negatively impact the model performance. Pre-trained feature encoders can alleviate this issue via producing compact representations for novel classes. However, creating general pre-trained encoders for various types of graph data has been proven to be challenging. As such, there is a demand for an effective method that does not rely on pre-trained graph encoders. In this paper, we propose an IMbalance-Aware method named OpenIMA for Open-world semi-supervised node classification, which trains the node classification model from scratch via contrastive learning with bias-reduced pseudo labels. Extensive experiments on seven popular graph benchmarks demonstrate the effectiveness of OpenIMA, and the source code has been available on GitHub.

cs.LG↗

arXiv2024

Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems

Large language models (LLMs) have strong capabilities in solving diverse natural language processing tasks. However, the safety and security issues of LLM systems have become the major obstacle to their widespread application. Many studies have extensively investigated risks in LLM systems and developed the corresponding mitigation strategies. Leading-edge enterprises such as OpenAI, Google, Meta, and Anthropic have also made lots of efforts on responsible LLMs. Therefore, there is a growing need to organize the existing studies and establish comprehensive taxonomies for the community. In this paper, we delve into four essential modules of an LLM system, including an input module for receiving prompts, a language model trained on extensive corpora, a toolchain module for development and deployment, and an output module for exporting LLM-generated content. Based on this, we propose a comprehensive taxonomy, which systematically analyzes potential risks associated with each module of an LLM system and discusses the corresponding mitigation strategies. Furthermore, we review prevalent benchmarks, aiming to facilitate the risk assessment of LLM systems. We hope that this paper can help LLM participants embrace a systematic perspective to build their responsible LLM systems.

cs.CL↗

arXiv2023

Towards Runtime Customizable Trusted Execution Environment on FPGA-SoC

Processing sensitive data and deploying well-designed Intellectual Property (IP) cores on remote Field Programmable Gate Array (FPGA) are prone to private data leakage and IP theft. One effective solution is constructing Trusted Execution Environment (TEE) on FPGA-SoCs (FPGA System on Chips). Researchers have integrated this type TEE with Trusted Platform Module (TPM)-based trusted boot, denoted as FPGA-SoC tbTEE. But there is no effort on secure and trusted runtime customization of FPGA-SoC TEE. This paper extends FPGA-SoC tbTEE to build Runtime Customizable TEE (RCTEE) on FPGA-SoC by additive three major components (our work): 1) CrloadIP, which can load an IP core at runtime such that RCTEE can be adjusted dynamically and securely; 2) CexecIP, which can not only execute an IP core without modifying the operating system of FPGA-SoC TEE, but also prevent insider attacks from executing IPs deployed in RCTEE; 3) CremoAT, which can provide the newly measured RCTEE state and establish a secure and trusted communication path between remote verifiers and RCTEE. We conduct a security analysis of RCTEE and its performance evaluation on Xilinx Zynq UltraScale+ XCZU15EG 2FFVB1156 MPSoC.

cs.CR↗

arXiv2023

FC-KBQA: A Fine-to-Coarse Composition Framework for Knowledge Base Question Answering

The generalization problem on KBQA has drawn considerable attention. Existing research suffers from the generalization issue brought by the entanglement in the coarse-grained modeling of the logical expression, or inexecutability issues due to the fine-grained modeling of disconnected classes and relations in real KBs. We propose a Fine-to-Coarse Composition framework for KBQA (FC-KBQA) to both ensure the generalization ability and executability of the logical expression. The main idea of FC-KBQA is to extract relevant fine-grained knowledge components from KB and reformulate them into middle-grained knowledge pairs for generating the final logical expressions. FC-KBQA derives new state-of-the-art performance on GrailQA and WebQSP, and runs 4 times faster than the baseline.

cs.AI↗

arXiv2022

Continual Transfer Learning for Cross-Domain Click-Through Rate Prediction at Taobao

As one of the largest e-commerce platforms in the world, Taobao's recommendation systems (RSs) serve the demands of shopping for hundreds of millions of customers. Click-Through Rate (CTR) prediction is a core component of the RS. One of the biggest characteristics in CTR prediction at Taobao is that there exist multiple recommendation domains where the scales of different domains vary significantly. Therefore, it is crucial to perform cross-domain CTR prediction to transfer knowledge from large domains to small domains to alleviate the data sparsity issue. However, existing cross-domain CTR prediction methods are proposed for static knowledge transfer, ignoring that all domains in real-world RSs are continually time-evolving. In light of this, we present a necessary but novel task named Continual Transfer Learning (CTL), which transfers knowledge from a time-evolving source domain to a time-evolving target domain. In this work, we propose a simple and effective CTL model called CTNet to solve the problem of continual cross-domain CTR prediction at Taobao, and CTNet can be trained efficiently. Particularly, CTNet considers an important characteristic in the industry that models has been continually well-trained for a very long time. So CTNet aims to fully utilize all the well-trained model parameters in both source domain and target domain to avoid losing historically acquired knowledge, and only needs incremental target domain data for training to guarantee efficiency. Extensive offline experiments and online A/B testing at Taobao demonstrate the efficiency and effectiveness of CTNet. CTNet is now deployed online in the recommender systems of Taobao, serving the main traffic of hundreds of millions of active users.

cs.IR↗

arXiv2020

Deep Learning-Based Gait Recognition Using Smartphones in the Wild

Compared to other biometrics, gait is difficult to conceal and has the advantage of being unobtrusive. Inertial sensors, such as accelerometers and gyroscopes, are often used to capture gait dynamics. These inertial sensors are commonly integrated into smartphones and are widely used by the average person, which makes gait data convenient and inexpensive to collect. In this paper, we study gait recognition using smartphones in the wild. In contrast to traditional methods, which often require a person to walk along a specified road and/or at a normal walking speed, the proposed method collects inertial gait data under unconstrained conditions without knowing when, where, and how the user walks. To obtain good person identification and authentication performance, deep-learning techniques are presented to learn and model the gait biometrics based on walking data. Specifically, a hybrid deep neural network is proposed for robust gait feature representation, where features in the space and time domains are successively abstracted by a convolutional neural network and a recurrent neural network. In the experiments, two datasets collected by smartphones for a total of 118 subjects are used for evaluations. The experiments show that the proposed method achieves higher than 93.5\% and 93.7\% accuracies in person identification and authentication, respectively.

cs.LG↗

arXiv2012

Moderate deviations principle for empirical covariance from a unit root

In the present paper, we consider the linear autoregressive model in $\rr$, $$ X_{k,n}=θ_n X_{k,n-1}+ξ_k, k=0,1,...,n, n\ge 1$$ where $θ_n\in [0,1)$ is unknown, $(ξ_k)_{k\in\zz}$ is a sequence of centered i.i.d. r.v. valued in $\rr$ representing the noise. When $θ_n\to 1$, the moderate deviations principle for empirical covariance is discussed and as statistical applications we provide the moderate deviation estimates of the least square and the Yule-Walker estimators of the parameter $θ_n$.

math.PR↗