EditCLEVR: A Paired-Scene Intervention Benchmark for Compositional Faithfulness of Object-Centric Representations

• URL: http://arxiv.org/abs/2607.22705v1
• Date: Sun, 19 Jul 2026 16:11:43 GMT
• Title: EditCLEVR: A Paired-Scene Intervention Benchmark for Compositional Faithfulness of Object-Centric Representations
• Abstract summary: We introduce EditCLEVR, a paired-scene intervention benchmark for object-centric learning.<n>Each example contains a before/after pair of renders with the same object indices and scene layout.<n>It includes either exactly one known change attribute on one known object or a no-edit re-render for drift measurement.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Object-centric learning aims to represent scenes as objects whose properties can be reused in new combinations. Existing evaluations usually score segmentation, single-image factor prediction, or downstream accuracy, but these tests do not directly ask whether a per-object representation behaves correctly under a controlled semantic edit. We introduce EditCLEVR, a paired-scene intervention benchmark in which each example contains a before/after pair of CLEVR-style renders with the same object indices and scene layout, and either exactly one known attribute change on one known object or a no-edit re-render for drift measurement. The protocol includes probe-free diagnostics for representation-change localization and stability, together with probe-decoded semantic faithfulness metrics that test whether the predicted scene change matches the intended intervention across in-distribution and compositional out-of-distribution (OOD) suites, allowing code-space movement and decoded object-attribute correctness to be evaluated separately. We introduce the semantic metric Scene-Graph Intervention Accuracy (SGIA), which requires the full after-scene prediction to be correct and the only predicted before-to-after semantic change to be the intended object-factor edit. We also establish Delta-SGIA as a companion diagnostic that checks the single-site change pattern without requiring the full after-scene graph to be correct. Baseline evaluations on ground-truth-mask backbones, learned-slot models, SAM 2 + frozen-ViT models, and one mask-feature hybrid indicate that CoGenT-OOD-core degradation can persist under ground-truth instance masks, that mask source accounts for part but not all of native performance, and that locality or stability alone can overstate semantic faithfulness. Code is available at https://github.com/torux-bughunter/EditCLEVR.

Related papers

• Sparse MLLM Anchors, Dense Adaptation: Breaking the Self-Referential Loop in Wild Test-Time Adaptation [17.055807559866093]
  We introduce MASA (Multimodal-LLM-Anchored Semantic Adaptation), which complements model-internal evidence with structured semantic descriptions from a frozen large language model (MLLM)<n> MASA encodes descriptions, propagates them to neighboring test samples, and stores the resulting visual-semantic information in an online prototype memory.<n>We evaluate MASA on the WTTA ImageNet-C benchmark under limited-batch, mixed-domain, and imbalanced-label-shift settings with ResNet and ViT backbones.
  arXiv  Detail & Related papers  (2026-09-15T11:48:30Z)
• DPM++: Dynamic Masked Metric Learning for Occluded Person Re-identification [32.9985650634494]
  DPM++ is a Dynamic Masked Metric Learning framework for occluded person re-identification.<n>It learns an input-adaptive masked metric that dynamically selects reliable identity subspaces for each occluded instance.<n>It consistently outperforms previous state-of-the-art methods in both holistic and occlusion scenarios.
  arXiv  Detail & Related papers  (2026-05-07T17:47:23Z)
• IoUCert: Robustness Verification for Anchor-based Object Detectors [58.35703549470485]
  We introduce IoUCert, a novel formal verification framework designed specifically to overcome these bottlenecks in anchor-based object detection architectures.<n>We show that our method enables the robustness verification of realistic, anchor-based models including SSD, YOLOv2, and YOLOv3 variants against various input perturbations.
  arXiv  Detail & Related papers  (2026-03-03T14:36:46Z)
• RealDrag: The First Dragging Benchmark with Real Target Image [9.439854281295803]
  textbfRealDrag is the first comprehensive benchmark for point based image editing that includes paired ground truth target images.<n>Our dataset contains over 400 human annotated samples from diverse video sources.<n>We also propose four novel, task specific metrics: Semantical Distance (SeD), Outer Mask Preserving Score (OMPS), Inner Patch Preserving Score (IPPS), and Directional Similarity (DiS)
  arXiv  Detail & Related papers  (2025-12-13T11:14:03Z)
• ObjectAlign: Neuro-Symbolic Object Consistency Verification and Correction [25.651102245332766]
  We introduce ObjectAlign, a framework that blends perceptual metrics with symbolic reasoning to detect, verify, and correct inconsistencies in edited video sequences.<n>We show up to 1.4 point improvement in CLIP Score and up to 6.1 point improvement in warp error compared to SOTA baselines on the DAVIS and Pexels video datasets.
  arXiv  Detail & Related papers  (2025-11-24T02:50:01Z)
• 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)
• Semi-DETR: Semi-Supervised Object Detection with Detection Transformers [105.45018934087076]
  We analyze the DETR-based framework on semi-supervised object detection (SSOD) We present Semi-DETR, the first transformer-based end-to-end semi-supervised object detector. Our method outperforms all state-of-the-art methods by clear margins.
  arXiv  Detail & Related papers  (2023-07-16T16:32:14Z)
• Auditing AI models for Verified Deployment under Semantic Specifications [65.12401653917838]
  AuditAI bridges the gap between interpretable formal verification and scalability. We show how AuditAI allows us to obtain controlled variations for verification and certified training while addressing the limitations of verifying using only pixel-space perturbations.
  arXiv  Detail & Related papers  (2021-09-25T22:53:24Z)
• Synthesizing the Unseen for Zero-shot Object Detection [72.38031440014463]
  We propose to synthesize visual features for unseen classes, so that the model learns both seen and unseen objects in the visual domain. We use a novel generative model that uses class-semantics to not only generate the features but also to discriminatively separate them.
  arXiv  Detail & Related papers  (2020-10-19T12:36:11Z)
• End-to-End Object Detection with Transformers [88.06357745922716]
  We present a new method that views object detection as a direct set prediction problem. Our approach streamlines the detection pipeline, effectively removing the need for many hand-designed components. The main ingredients of the new framework, called DEtection TRansformer or DETR, are a set-based global loss.
  arXiv  Detail & Related papers  (2020-05-26T17:06:38Z)

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.