3D-DefectBench: A Controlled Factorial Study of Vision-Language Model Evaluation Pipelines for Fine-Grained 3D Generation Defects

• URL: http://arxiv.org/abs/2607.10826v1
• Date: Sun, 12 Jul 2026 16:41:24 GMT
• Title: 3D-DefectBench: A Controlled Factorial Study of Vision-Language Model Evaluation Pipelines for Fine-Grained 3D Generation Defects
• Abstract summary: We introduce 3D-DefectBench, a benchmark and framework for analysis of 3D defect detection pipelines.<n>Using a balanced factorial design, we vary four pipeline factors, VLM, camera protocol, visual input, and prompt schema, across 84 inference designs.<n>The best of 12 VLM judges still lag behind trained human labelers, while texture agreement drops sharply when expert-consensus labels are replaced by noisier silver labels.
• Score: 17.49934543489464
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Automated evaluation is essential for scaling generative 3D systems, where exhaustive human review is costly and slow. However, the reliability of an automated judge depends on the entire evaluation pipeline, not only the underlying vision-language model (VLM), but also how assets are rendered, what visual evidence is provided, how the task is specified, and how human reference labels are constructed. We introduce 3D-DefectBench, a benchmark and framework for systematic analysis of VLM-based 3D defect detection pipelines. It complements holistic ratings and pairwise preferences with nine fine-grained binary defects spanning geometry, texture, and prompt adherence, providing actionable diagnostics for generator development and judge evaluation. Using a balanced factorial design, we vary four pipeline factors, VLM, camera protocol, visual input, and prompt schema, across 84 inference designs and approximately 3.2 million scored defect decisions, followed by staged validation on a broader set of frontier models. Model choice is the largest determinant of agreement with human labels, but the remaining factors also affect performance, interact with model selection, and can change the best configuration. Within the evaluated design space, a compact six-view RGB protocol performs comparably to denser multi-view settings and inputs augmented with depth or surface normals, making it a strong cost-effective default. Under this standardized pipeline, the best of 12 VLM judges still lag behind trained human labelers, while texture agreement drops sharply when expert-consensus labels are replaced by noisier silver labels. These findings show that automated judges should be evaluated as complete pipelines and calibrated across human reference regimes, rather than benchmarked only as standalone models. We release labels, prompts, predictions, and Croissant metadata on Hugging Face.

Related papers

• PluginEval: A Diagnostic Benchmark for Fine-Grained Error Attribution in Function Calling [11.1017602726296]
  Large Language Models increasingly operate as autonomous agents.<n>Current benchmarks face three structural limitations.<n>Eval is a benchmark constructed through a two-stage framework.
  arXiv  Detail & Related papers  (2026-08-09T13:25:54Z)
• UXBench: Measuring the Actionability of LLM-Generated UX Critiques [58.14408744110387]
  We introduce UXBench, a benchmark for evaluating large language models (LLMs) as interaction-grounded UX judges.<n> UXBench comprises local-first runnable web fixtures spanning ten product-surface families, paired with coverage-gated rubric exploration.<n>We evaluate eight frontier models under both an automated repair-lift protocol and a blind human validation study.
  arXiv  Detail & Related papers  (2026-06-15T06:08:39Z)
• $μ_0$: A Scalable 3D Interaction-Trace World Model [54.38434730157622]
  We present a scalable world model based on 3D traces.<n>Our TraceExtract system automatically extracts 3D supervision.<n>These results establish 3D traces as a scalable and transferable representation for cross-embodiment manipulation.
  arXiv  Detail & Related papers  (2026-06-11T17:59:56Z)
• DB-3DME: From Dataset to Benchmark for Human-aligned Automatic 3D Mesh Evaluation [16.60840648468348]
  We introduce DB-3DME, the dataset and benchmark for 3D mesh evaluation.<n>DB-3DME contains 2,619 synthetic 3D meshes paired with human ratings on Geometry and Prompt Adherence.<n>We fine-tune an open-weight VLM, Qwen-2.5-VL-7B, for 3D mesh evaluation by adapting the visual encoder while freezing the language model.
  arXiv  Detail & Related papers  (2026-06-08T20:17:49Z)
• DisPOSE: Projected Polystochastic Diffusion for Self-Supervised Multi-View 3D Human Pose Estimation [58.47973015036709]
  DisPOSE is a self-supervised framework that approximates the inherently discrete multi-view person-assignment problem.<n>By employing differentiable Sinkhorn projections during denoising, our model learns to guide solutions toward valid and feasible assignments.<n>The proposed approach outperforms current state-of-the-art self-supervised methods on standard datasets.
  arXiv  Detail & Related papers  (2026-06-05T16:14:39Z)
• Employing Vision-Language Models for Face Image Quality Assessment [0.023141219541346198]
  Face Image Quality Assessment (FIQA) is a crucial control step in biometric pipelines.<n>State-of-the-art FIQA methods achieve high utility but typically operate as "black boxes"<n>We investigate the potential of off-the-shelf Vision-Language Models (VLMs) to bridge this gap by performing FIQA in a zero-shot setting.
  arXiv  Detail & Related papers  (2026-05-17T14:57:52Z)
• Evaluating Agentic AI in the Wild: Failure Modes, Drift Patterns, and a Production Evaluation Framework [0.0]
  We present a taxonomy of seven failure modes unique to production agentic systems.<n>Standard metrics fail to detect four of the seven failure modes entirely.<n>We propose PAEF, a five-dimension evaluation framework with an open-source reference implementation.
  arXiv  Detail & Related papers  (2026-05-02T21:02:08Z)
• ReVSI: Rebuilding Visual Spatial Intelligence Evaluation for Accurate Assessment of VLM 3D Reasoning [59.558706734431276]
  Current evaluations of spatial intelligence can be systematically invalid under modern vision-language model (VLM) settings.<n>We introduce ReVSI, a benchmark and protocol that ensures each QA pair is answerable and correct under the model's actual inputs.
  arXiv  Detail & Related papers  (2026-04-27T10:45:51Z)
• Evian: Towards Explainable Visual Instruction-tuning Data Auditing [14.93566912726999]
  We introduce a novel "Decomposition-then-Evaluation" paradigm that breaks model responses into constituent cognitive components.<n>We instantiate this paradigm via EVIAN, an automated framework that evaluates these components along the axes of Image-Text Consistency, Logical Coherence, and Factual Accuracy.
  arXiv  Detail & Related papers  (2026-04-22T13:28:27Z)
• EdiVal-Agent: An Object-Centric Framework for Automated, Fine-Grained Evaluation of Multi-Turn Editing [170.71134330650796]
  EdiVal-Agent is an object-centric evaluation framework for instruction-based image editing.<n>It is designed to assess not only standard single-turn but also multi-turn instruction-based editing with precision.<n>We build EdiVal-Bench, a benchmark covering 9 instruction types and 13 state-of-the-art editing models spanning in-context, flow-matching, and diffusion paradigms.
  arXiv  Detail & Related papers  (2025-09-16T17:45:39Z)
• 3D Face Modeling via Weakly-supervised Disentanglement Network joint Identity-consistency Prior [62.80458034704989]
  Generative 3D face models featuring disentangled controlling factors hold immense potential for diverse applications in computer vision and computer graphics. Previous 3D face modeling methods face a challenge as they demand specific labels to effectively disentangle these factors. This paper introduces a Weakly-Supervised Disentanglement Framework, denoted as WSDF, to facilitate the training of controllable 3D face models without an overly stringent labeling requirement.
  arXiv  Detail & Related papers  (2024-04-25T11:50:47Z)
• VIBUS: Data-efficient 3D Scene Parsing with VIewpoint Bottleneck and Uncertainty-Spectrum Modeling [2.0624279915507047]
  Training 3D scene parsing models with sparse supervision is an intriguing alternative. We term this task as data-efficient 3D scene parsing. We propose an effective two-stage framework named VIBUS to resolve it.
  arXiv  Detail & Related papers  (2022-10-20T17:59:57Z)
• GLENet: Boosting 3D Object Detectors with Generative Label Uncertainty Estimation [70.75100533512021]
  In this paper, we formulate the label uncertainty problem as the diversity of potentially plausible bounding boxes of objects. We propose GLENet, a generative framework adapted from conditional variational autoencoders, to model the one-to-many relationship between a typical 3D object and its potential ground-truth bounding boxes with latent variables. The label uncertainty generated by GLENet is a plug-and-play module and can be conveniently integrated into existing deep 3D detectors.
  arXiv  Detail & Related papers  (2022-07-06T06:26:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.