arXiv subjects

Zhaohai Li

Publications and source records attributed to Zhaohai Li.

13 recordsLinked to original sources

arXiv2026

Qwen-Drive-1.0: An Initial Step towards a Vision-Language Foundation Model for Autonomous Driving

We present Qwen-Drive-1.0, an initial step towards a vision-language foundation model for autonomous driving. Qwen-Drive-1.0 retains the architecture of the pretrained vision-language model (VLM) and integrates 3D perception, visual question answering, and motion planning within a unified framework. An external bird's-eye-view (BEV) perception head jointly performs 3D object detection, semantic occupancy prediction, and BEV map segmentation. It serves as a probe of the 3D information accessible from the shared representations and provides an explicit, inspectable interface to 3D scene structure. A Planning Expert conditions on shared VLM representations to generate future ego trajectories. A staged training recipe combines driving supervision with general-purpose vision-language data to acquire driving-specific competence while helping preserve broad visual understanding and instruction-following capabilities. Experiments demonstrate strong 3D perception and driving scene understanding while largely preserving general vision-language capability. Comprehensive evaluations across open-loop, pseudo-closed-loop, and closed-loop settings further show highly competitive motion-planning performance.

cs.CV↗

arXiv2026

Science Edge Evaluation: SEE the Missing Step Toward Real Scientific Discovery

Large language models (LLMs) are increasingly involved in scientific discovery, yet it remains unclear whether they can support complex real laboratory science. Here we introduce Science Edge Evaluation (SEE), a multimodal benchmark of expert-curated questions grounded in peer-reviewed literature and experimental practice in chemistry, biology, and materials science. Evaluation of 19 multimodal large language models (MLLMs) shows that even the best-performing model reaches only 48.7% accuracy. Moreover, general-purpose models outperform science-specialized models on average. In the visual-agent evaluation, the use of tools increases the best accuracy to 52.7%. Tool use can expand the information available to models, but more information does not necessarily lead to reliable scientific reasoning. The key challenge is whether models can manage tool-derived information within the boundaries of the original experimental evidence. Together, these findings reveal that current MLLMs still cannot reliably make justified and evidence-bounded inferences from experimental results, which is an essential capability in real scientific discovery. Bridging this gap requires MLLMs to transition from explaining established scientific concepts to deriving novel and evidence-based insights from experimental data.

cs.AI↗

arXiv2026

Qwen-VLA: Unifying Vision-Language-Action Modeling across Tasks, Environments, and Robot Embodiments

Embodied intelligence is often studied through specialized models for individual tasks such as manipulation or navigation, resulting in fragmented capabilities and limited generalization across tasks, environments, and robot embodiments. In this work, we study whether heterogeneous embodied decision-making problems can be unified within a single vision-language-action model. We present Qwen-VLA, a unified embodied foundation model that extends Qwen's vision-language modeling stack from perception, understanding, and reasoning to continuous action and trajectory generation through a DiT-based action decoder. Qwen-VLA is trained with a large-scale joint pretraining recipe over diverse data sources, including robotics manipulation trajectories, human egocentric demonstrations, synthetic simulation data, vision-and-language navigation data, trajectory-centric supervision, and auxiliary vision-language data. To support multiple robot platforms, we introduce embodiment-aware prompt conditioning, where robot-specific textual descriptions specify the current embodiment and control convention. We further cast manipulation, navigation, and trajectory prediction into a unified action-and-trajectory prediction framework, enabling transferable visual grounding, spatial reasoning, and continuous action generation across robot morphologies, task families, and environments. Experiments on manipulation, navigation, and trajectory-centric benchmarks show consistent multi-task performance and out-of-distribution generalization under variations in scene layout, background, lighting, object configuration, and robot embodiment. Qwen-VLA-Instruct achieves 97.9% on LIBERO, 73.7% on Simpler-WidowX, 86.1%/87.2% on RoboTwin-Easy/Hard, 69.0% OSR on R2R, 59.6% SR on RxR, 76.9% average OOD success in real-world ALOHA experiments, and 26.6% zero-shot success on DOMINO dynamic manipulation.

cs.RO↗

arXiv2025

Qwen3-VL Technical Report

We introduce Qwen3-VL, the most capable vision-language model in the Qwen series to date, achieving superior performance across a broad range of multimodal benchmarks. It natively supports interleaved contexts of up to 256K tokens, seamlessly integrating text, images, and video. The model family includes both dense (2B/4B/8B/32B) and mixture-of-experts (30B-A3B/235B-A22B) variants to accommodate diverse latency-quality trade-offs. Qwen3-VL delivers three core pillars: (i) markedly stronger pure-text understanding, surpassing comparable text-only backbones in several cases; (ii) robust long-context comprehension with a native 256K-token window for both text and interleaved multimodal inputs, enabling faithful retention, retrieval, and cross-referencing across long documents and videos; and (iii) advanced multimodal reasoning across single-image, multi-image, and video tasks, demonstrating leading performance on comprehensive evaluations such as MMMU and visual-math benchmarks (e.g., MathVista and MathVision). Architecturally, we introduce three key upgrades: (i) an enhanced interleaved-MRoPE for stronger spatial-temporal modeling across images and video; (ii) DeepStack integration, which effectively leverages multi-level ViT features to tighten vision-language alignment; and (iii) text-based time alignment for video, evolving from T-RoPE to explicit textual timestamp alignment for more precise temporal grounding. Under comparable token budgets and latency constraints, Qwen3-VL achieves superior performance in both dense and Mixture-of-Experts (MoE) architectures. We envision Qwen3-VL serving as a foundational engine for image-grounded reasoning, agentic decision-making, and multimodal code intelligence in real-world workflows.

cs.CV↗

arXiv2025

On MCMC mixing for predictive inference under unidentified transformation models

Reliable Bayesian predictive inference has long been an open problem under unidentified transformation models, since the Markov Chain Monte Carlo (MCMC) chains of posterior predictive distribution (PPD) values are generally poorly mixed. We address the poorly mixed PPD value chains under unidentified transformation models through an adaptive scheme for prior adjustment. Specifically, we originate a conception of sufficient informativeness, which explicitly quantifies the information level provided by nonparametric priors, and assesses MCMC mixing by comparison with the within-chain MCMC variance. We formulate the prior information level by a set of hyperparameters induced from the nonparametric prior elicitation with an analytic expression, which is guaranteed by asymptotic theory for the posterior variance under unidentified transformation models. The analytic prior information level consequently drives a hyperparameter tuning procedure to achieve MCMC mixing. The proposed method is general enough to cover various data domains through a multiplicative error working model. Comprehensive simulations and real-world data analysis demonstrate that our method successfully achieves MCMC mixing and outperforms state-of-the-art competitors in predictive capability.

stat.ME↗

arXiv2025

Qwen2.5-VL Technical Report

We introduce Qwen2.5-VL, the latest flagship model of Qwen vision-language series, which demonstrates significant advancements in both foundational capabilities and innovative functionalities. Qwen2.5-VL achieves a major leap forward in understanding and interacting with the world through enhanced visual recognition, precise object localization, robust document parsing, and long-video comprehension. A standout feature of Qwen2.5-VL is its ability to localize objects using bounding boxes or points accurately. It provides robust structured data extraction from invoices, forms, and tables, as well as detailed analysis of charts, diagrams, and layouts. To handle complex inputs, Qwen2.5-VL introduces dynamic resolution processing and absolute time encoding, enabling it to process images of varying sizes and videos of extended durations (up to hours) with second-level event localization. This allows the model to natively perceive spatial scales and temporal dynamics without relying on traditional normalization techniques. By training a native dynamic-resolution Vision Transformer (ViT) from scratch and incorporating Window Attention, we reduce computational overhead while maintaining native resolution. As a result, Qwen2.5-VL excels not only in static image and document understanding but also as an interactive visual agent capable of reasoning, tool usage, and task execution in real-world scenarios such as operating computers and mobile devices. Qwen2.5-VL is available in three sizes, addressing diverse use cases from edge AI to high-performance computing. The flagship Qwen2.5-VL-72B model matches state-of-the-art models like GPT-4o and Claude 3.5 Sonnet, particularly excelling in document and diagram understanding. Additionally, Qwen2.5-VL maintains robust linguistic performance, preserving the core language competencies of the Qwen2.5 LLM.

cs.CV↗

arXiv2024

CC-OCR: A Comprehensive and Challenging OCR Benchmark for Evaluating Large Multimodal Models in Literacy

Large Multimodal Models (LMMs) have demonstrated impressive performance in recognizing document images with natural language instructions. However, it remains unclear to what extent capabilities in literacy with rich structure and fine-grained visual challenges. The current landscape lacks a comprehensive benchmark to effectively measure the literate capabilities of LMMs. Existing benchmarks are often limited by narrow scenarios and specified tasks. To this end, we introduce CC-OCR, a comprehensive benchmark that possesses a diverse range of scenarios, tasks, and challenges. CC-OCR comprises four OCR-centric tracks: multi-scene text reading, multilingual text reading, document parsing, and key information extraction. It includes 39 subsets with 7,058 full annotated images, of which 41% are sourced from real applications, and released for the first time. We evaluate nine prominent LMMs and reveal both the strengths and weaknesses of these models, particularly in text grounding, multi-orientation, and hallucination of repetition. CC-OCR aims to comprehensively evaluate the capabilities of LMMs on OCR-centered tasks, facilitating continued progress in this crucial area.

cs.CV↗

arXiv2024

VL-Reader: Vision and Language Reconstructor is an Effective Scene Text Recognizer

Text recognition is an inherent integration of vision and language, encompassing the visual texture in stroke patterns and the semantic context among the character sequences. Towards advanced text recognition, there are three key challenges: (1) an encoder capable of representing the visual and semantic distributions; (2) a decoder that ensures the alignment between vision and semantics; and (3) consistency in the framework during pre-training, if it exists, and fine-tuning. Inspired by masked autoencoding, a successful pre-training strategy in both vision and language, we propose an innovative scene text recognition approach, named VL-Reader. The novelty of the VL-Reader lies in the pervasive interplay between vision and language throughout the entire process. Concretely, we first introduce a Masked Visual-Linguistic Reconstruction (MVLR) objective, which aims at simultaneously modeling visual and linguistic information. Then, we design a Masked Visual-Linguistic Decoder (MVLD) to further leverage masked vision-language context and achieve bi-modal feature interaction. The architecture of VL-Reader maintains consistency from pre-training to fine-tuning. In the pre-training stage, VL-Reader reconstructs both masked visual and text tokens, while in the fine-tuning stage, the network degrades to reconstruct all characters from an image without any masked regions. VL-reader achieves an average accuracy of 97.1% on six typical datasets, surpassing the SOTA by 1.1%. The improvement was even more significant on challenging datasets. The results demonstrate that vision and language reconstructor can serve as an effective scene text recognizer.

cs.CV↗

arXiv2024

Platypus: A Generalized Specialist Model for Reading Text in Various Forms

Reading text from images (either natural scenes or documents) has been a long-standing research topic for decades, due to the high technical challenge and wide application range. Previously, individual specialist models are developed to tackle the sub-tasks of text reading (e.g., scene text recognition, handwritten text recognition and mathematical expression recognition). However, such specialist models usually cannot effectively generalize across different sub-tasks. Recently, generalist models (such as GPT-4V), trained on tremendous data in a unified way, have shown enormous potential in reading text in various scenarios, but with the drawbacks of limited accuracy and low efficiency. In this work, we propose Platypus, a generalized specialist model for text reading. Specifically, Platypus combines the best of both worlds: being able to recognize text of various forms with a single unified architecture, while achieving excellent accuracy and high efficiency. To better exploit the advantage of Platypus, we also construct a text reading dataset (called Worms), the images of which are curated from previous datasets and partially re-labeled. Experiments on standard benchmarks demonstrate the effectiveness and superiority of the proposed Platypus model. Model and data will be made publicly available at https://github.com/AlibabaResearch/AdvancedLiterateMachinery/tree/main/OCR/Platypus.

cs.CV↗

arXiv2023

Conditional Text Image Generation with Diffusion Models

Current text recognition systems, including those for handwritten scripts and scene text, have relied heavily on image synthesis and augmentation, since it is difficult to realize real-world complexity and diversity through collecting and annotating enough real text images. In this paper, we explore the problem of text image generation, by taking advantage of the powerful abilities of Diffusion Models in generating photo-realistic and diverse image samples with given conditions, and propose a method called Conditional Text Image Generation with Diffusion Models (CTIG-DM for short). To conform to the characteristics of text images, we devise three conditions: image condition, text condition, and style condition, which can be used to control the attributes, contents, and styles of the samples in the image generation process. Specifically, four text image generation modes, namely: (1) synthesis mode, (2) augmentation mode, (3) recovery mode, and (4) imitation mode, can be derived by combining and configuring these three conditions. Extensive experiments on both handwritten and scene text demonstrate that the proposed CTIG-DM is able to produce image samples that simulate real-world complexity and diversity, and thus can boost the performance of existing text recognizers. Besides, CTIG-DM shows its appealing potential in domain adaptation and generating images containing Out-Of-Vocabulary (OOV) words.

cs.CV↗

arXiv2023

Bayesian prediction via nonparametric transformation models

This article tackles the old problem of prediction via a nonparametric transformation model (NTM) in a new Bayesian way. Estimation of NTMs is known challenging due to model unidentifiability though appealing because of its robust prediction capability in survival analysis. Inspired by the uniqueness of the posterior predictive distribution, we achieve efficient prediction via the NTM aforementioned under the Bayesian paradigm. Our strategy is to assign weakly informative priors to nonparametric components rather than identify the model by adding complicated constraints in the existing literature. The Bayesian success pays tribute to i) a subtle cast of NTMs by an exponential transformation for the purpose of compressing spaces of infinite-dimensional parameters to positive quadrants considering non-negativity of the failure time; ii) a newly constructed weakly informative quantile-knots I-splines prior for the recast transformation function together with the Dirichlet process mixture model assigned to the error distribution. In addition, we provide a convenient and precise estimator for the identified parameter component subject to the general unit-norm restriction through posterior modification, enabling effective relative risks. Simulations and applications on real datasets reveal that our method is robust and outperforms the competing methods. An R package BuLTM is available to predict survival curves, estimate relative risks, and facilitate posterior checking.

stat.ME↗

arXiv2016

An efficient and doubly robust empirical likelihood approach for estimating equations with missing data

This paper considers an empirical likelihood inference for parameters defined by general estimating equations, when data are missing at random. The efficiency of existing estimators depends critically on correctly specifying the conditional expectation of the estimating function given the observed components of the random observations. When the conditional expectation is not correctly specified, the efficiency of estimation can be severely compromised even if the propensity function (of missingness) is correctly specified. We propose an efficient estimator which enjoys the double-robustness property and can achieve the semiparametric efficiency bound within the class of the estimating functions that are generated by the estimating function of estimating equations, if both the propensity model and the regression model (of the conditional expectation) are specified correctly. Moreover, if the propensity model is specified correctly but the regression model is misspecified, the proposed estimator still achieves a semiparametric efficiency lower bound within a more general class of estimating functions. Simulation results suggest that the proposed estimators are robust against misspecification of the propensity model or regression model and outperform many existing competitors in the sense of having smaller mean-square errors. Moreover, using our approach for statistical inference requires neither resampling nor kernel smoothing. A real data example is used to illustrate the proposed approach.

stat.ME↗

arXiv2015

A Conversation with Robert C. Elston

Robert C. Elston was born on February 4, 1932, in London, England. He went to Cambridge University to study natural science from 1952-1956 and obtained B.A., M.A. and Diploma in Agriculture (Dip Ag). He came to the US at age 24 to study animal breeding at Cornell University and received his Ph.D. in 1959. From 1959-1960, he was a post-doctoral fellow in biostatistics at University of North Carolina (UNC), Chapel Hill, where he studied mathematical statistics. He then rose through the academic ranks in the department of biostatistics at UNC, becoming a full professor in 1969. From 1979-1995, he was a professor and head of the Department of Biometry and Genetics at Louisiana State University Medical Center in New Orleans. In 1995, he moved to Case Western Reserve University where he is a professor of epidemiology and biostatistics and served as chairman from 2008 to 2014. Between 1966 and 2013, he directed 42 Ph.D. students and mentored over 40 post-doctoral fellows. If one regards him as a founder of a pedigree in research in genetic epidemiology, it was estimated in 2007 that there were more than 500 progeny. Among his many honors are a NIH Research Career Development Award (1966-1976), the Leadership Award from International Society of Human Genetics (1995), William Allan Award from American Society of Human Genetics (1996), NIH MERIT Award (1998) and the Marvin Zelen Leadership Award, Harvard University (2004). He is a Fellow of the American Statistical Association and the Institute of Mathematical Statistics as well as a Fellow of the Ohio Academy of Science. A leader in research in genetic epidemiology for over 40 years, he has published over 600 research articles in biostatistics, genetic epidemiology and applications. He has also coauthored and edited 9 books in biostatistics, population genetics and methods for the analysis of genetic data.

stat.OT↗