arXiv subjects

Zitong Zhou

Publications and source records attributed to Zitong Zhou.

7 recordsLinked to original sources

arXiv2026

WorldExam: Benchmarking World Models from Apparent Appearance to Inherent Reactivity

Controllable video generation models are increasingly being developed as world models. Accordingly, evaluating them in this role extends beyond the apparent appearance of generated videos to the inherent reactivity of the worlds they depict: the ability to infer from the scene state how the world should react and to generate plausible consequences not explicitly described in the input. Yet existing benchmarks mainly assess visual quality or explicit instruction fulfillment by checking whether requested actions and interaction outcomes are realized, leaving inherent reactivity underexamined. We introduce WorldExam, a hierarchical diagnostic benchmark spanning four levels: Visual Quality, Control Adherence, Spatial Consistency, and World Reactivity. It comprises 1,474 cases across eight dedicated tasks and supports unified evaluation of camera-, action-, and language-driven model paradigms. The World Reactivity level evaluates scene-conditioned reactions and goal-directed behaviors beyond what is explicitly specified in the input. Evaluation of 20 representative models reveals a clear capability split. Camera-driven models excel at camera control, but their interfaces do not support dynamic interaction; action-driven models control subjects more precisely but often leave the world unresponsive; and language-driven models perform better on interaction but follow complex controls less faithfully. No model combines broad task coverage with consistently strong performance, showing that high visual quality and explicit instruction fulfillment do not guarantee inherent reactivity.

cs.CV↗

arXiv2026

MedRealMM: A Real-World Multimodal Benchmark for Chinese Online Medical Consultation

Large language models (LLMs) are increasingly deployed in online medical consultation, yet existing benchmarks remain poorly aligned with real clinical practice. Many rely on synthetic conversations or patient simulators, omit patient-uploaded medical images, or evaluate open-ended clinical responses using multiple-choice or lexical-overlap metrics that poorly reflect clinical quality. We introduce \textbf{MedRealMM}, a large-scale benchmark for multimodal online medical consultation built from de-identified patient-doctor interactions collected from a nationwide Chinese internet hospital. MedRealMM uses a Multimodal Clinical Challenge Point (MCCP) extraction framework to identify clinically demanding moments in authentic consultation trajectories and converts each into a standardized next-response generation task while preserving the preceding text-image context. Each instance is paired with a case-specific rubric refined by physicians that rewards clinically desirable behaviors and penalizes unsafe, unsupported, or contradictory responses. The current release contains 5,620 real-world multimodal cases spanning 64 clinical departments. We evaluate 19 general-purpose and medical-specialized LLMs, including text-only and multimodal systems. Our results show that image information is critical for reliable clinical performance and that current frontier models remain below the online physician response. Although some frontier models satisfy as many or more positive clinical criteria than physicians, they trigger more negative criteria, indicating that safety-sensitive error avoidance remains a central bottleneck. MedRealMM offers a realistic and reproducible benchmark for evaluating multimodal medical reasoning in real-world online consultation. The dataset will be publicly available on Hugging Face at https://huggingface.co/datasets/jdh-algo/MedRealMM.

cs.AI↗

arXiv2026

Change And Cover: Last-Mile, Pull Request-Based Regression Test Augmentation

Software is in constant evolution, with developers frequently submitting pull requests (PRs) to introduce new features or fix bugs. Testing PRs is critical to maintaining software quality. Yet, even in projects with extensive test suites, some PR-modified lines remain untested, leaving a "last-mile" regression test gap. Existing test generators typically aim to improve overall coverage, but do not specifically target the uncovered lines in PRs. We present Change And Cover (ChaCo), an LLM-based test augmentation technique that addresses this gap. It makes three contributions: (i) ChaCo considers the PR-specific patch coverage, offering developers augmented tests for code just when it is on the developers' mind. (ii) We identify providing suitable test context as a crucial challenge for an LLM to generate useful tests, and present two techniques to extract relevant test content, such as existing test functions, fixtures, and data generators. (iii) To make augmented tests acceptable for developers, ChaCo carefully integrates them into the existing test suite, e.g., by matching the test's structure and style with the existing tests, and generates a summary of the test addition for developer review. We evaluate ChaCo on 145 PRs from three popular and complex open-source projects - SciPy, Qiskit, and Pandas. The approach successfully helps 30% of PRs achieve full patch coverage, at the cost of $0.11, showing its effectiveness and practicality. Human reviewers find the tests to be worth adding (4.53/5.0), well integrated (4.2/5.0), and relevant to the PR (4.7/5.0). Ablations show test context is crucial for context-aware test generation, leading to 2x coverage. We submitted 12 tests, of which 8 have already been merged, and two previously unknown bugs were exposed and fixed. We envision our approach to be integrated into CI workflows, automating the last mile of regression test augmentation.

cs.SE↗

arXiv2025

Targeted Testing of Compiler Optimizations via Grammar-Level Composition Styles

Ensuring the correctness of compiler optimizations is critical, but existing fuzzers struggle to test optimizations effectively. First, most fuzzers use optimization pipelines (heuristics-based, fixed sequences of passes) as their harness. The phase-ordering problem can enable or preempt transformations, so pipelines inevitably miss optimization interactions; moreover, many optimizations are not scheduled, even at aggressive levels. Second, optimizations typically fire only when inputs satisfy specific structural relationships, which existing generators and mutations struggle to produce. We propose targeted fuzzing of individual optimizations to complement pipeline-based testing. Our key idea is to exploit composition styles - structural relations over program constructs (adjacency, nesting, repetition, ordering) - that optimizations look for. We build a general-purpose, grammar-based mutational fuzzer, TargetFuzz, that (i) mines composition styles from an optimization-relevant corpus, then (ii) rebuilds them inside different contexts offered by a larger, generic corpus via synthesized mutations to test variations of optimization logic. TargetFuzz is adaptable to a new programming language by lightweight, grammar-based, construct annotations - and it automatically synthesizes mutators and crossovers to rebuild composition styles. No need for hand-coded generators or language-specific mutators, which is particularly useful for modular frameworks such as MLIR, whose dialect-based, rapidly evolving ecosystem makes optimizations difficult to fuzz. Our evaluation on LLVM and MLIR shows that TargetFuzz improves coverage by 8% and 11% and triggers optimizations 2.8$\times$ and 2.6$\times$, compared to baseline fuzzers under the targeted fuzzing mode. We show that targeted fuzzing is complementary: it effectively tests all 37 sampled LLVM optimizations, while pipeline-fuzzing missed 12.

cs.SE↗

arXiv2022

Deep Learning for Simultaneous Inference of Hydraulic and Transport Properties

Identifying the heterogeneous conductivity field and reconstructing the contaminant release history are key aspects of subsurface remediation. Achieving these two goals with limited and noisy hydraulic head and concentration measurements is challenging. The obstacles include solving an inverse problem for high-dimensional parameters, and the high-computational cost needed for the repeated forward modeling. We use a convolutional adversarial autoencoder (CAAE) for the parameterization of the heterogeneous non-Gaussian conductivity field with a low-dimensional latent representation. Additionally, we trained a three-dimensional dense convolutional encoder-decoder (DenseED) network to serve as the forward surrogate for the flow and transport processes. Combining the CAAE and DenseED forward surrogate models, the ensemble smoother with multiple data assimilation (ESMDA) algorithm is used to sample from the Bayesian posterior distribution of the unknown parameters, forming a CAAE-DenseED-ESMDA inversion framework. We applied this CAAE-DenseED-ESMDA inversion framework in a three-dimensional contaminant source and conductivity field identification problem. A comparison of the inversion results from CAAE-ESMDA with physical flow and transport simulator and CAAE-DenseED-ESMDA is provided, showing that accurate reconstruction results were achieved with a much higher computational efficiency.

cs.LG↗

arXiv2021

Thermal experiments for fractured rock characterization: theoretical analysis and inverse modeling

Field-scale properties of fractured rocks play crucial role in many subsurface applications, yet methodologies for identification of the statistical parameters of a discrete fracture network (DFN) are scarce. We present an inversion technique to infer two such parameters, fracture density and fractal dimension, from cross-borehole thermal experiments data. It is based on a particle-based heat-transfer model, whose evaluation is accelerated with a deep neural network (DNN) surrogate that is integrated into a grid search. The DNN is trained on a small number of heat-transfer model runs, and predicts the cumulative density function of the thermal field. The latter is used to compute fine posterior distributions of the (to-be-estimated) parameters. Our synthetic experiments reveal that fracture density is well constrained by data, while fractal dimension is harder to determine. Adding non-uniform prior information related to the DFN connectivity improves the inference of this parameter.

physics.geo-ph↗

arXiv2020

Markov Chain Monte Carlo with Neural Network Surrogates: Application to Contaminant Source Identification

Subsurface remediation often involves reconstruction of contaminant release history from sparse observations of solute concentration. Markov Chain Monte Carlo (MCMC), the most accurate and general method for this task, is rarely used in practice because of its high computational cost associated with multiple solves of contaminant transport equations. We propose an adaptive MCMC method, in which a transport model is replaced with a fast and accurate surrogate model in the form of a deep convolutional neural network (CNN). The CNN-based surrogate is trained on a small number of the transport model runs based on the prior knowledge of the unknown release history. Thus reduced computational cost allows one to reduce the sampling error associated with construction of the approximate likelihood function. As all MCMC strategies for source identification, our method has an added advantage of quantifying predictive uncertainty and accounting for measurement errors. Our numerical experiments demonstrate the accuracy comparable to that of MCMC with the forward transport model, which is obtained at a fraction of the computational cost of the latter.

cs.CE↗