Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

A Good Initialization is All You Need for Faithful Visual Attribution

About

Faithful visual attribution identifies which image regions support a model prediction. Search-based perturbation methods lead the insertion--deletion faithfulness frontier by masking regions and measuring score changes, but they usually output a complete ordering of all regions. Many applications, especially MLLM attribution and repair, only need a compact top-\(k\) evidence mask. We study this mask-first attribution problem. An exactly \(k\)-region mask is combinatorial: useful evidence can depend on interactions among fine regions. Coarse grouping can stabilize early search but aggregates redundant content, whereas one-step scoring can miss high-value combinations. We introduce two forward-only methods. \textsc{CoPAIR} uses a PhaseWin--Greedy gap diagnosis to construct coarse singleton/pair candidates that warm-start full-ordering search. \textsc{TRACE} directly searches fixed-cardinality fine-region masks with cross-entropy sampling, elite retention, and distribution updates, with a finite-budget recovery analysis. The resulting evidence set can be returned as a compact attribution mask or used to initialize Greedy or PhaseWin when a complete ranking is required. Across ImageNet classification with CLIP ViT-L/14, CLIP RN101, and ResNet-101, our initialized search methods establish a new state-of-the-art frontier for faithful full-ordering attribution under inclusive forward-call accounting. On POPE and RePOPE with Qwen2.5-VL-3B-Instruct and LLaVA-v1.5-7B, \textsc{TRACE}+Greedy gives the strongest search-based MLLM attribution results. Direct \textsc{TRACE} masks further achieve single-point RePOPE repair rates of \(94.44\%\) and \(96.00\%\), showing that compact evidence masks can be actionable attribution outputs, not merely prefixes of full rankings.

Zihan Gu, Jiayu Wang, Hua Zhang, Yue Hu• 2026

Related benchmarks

TaskDatasetResultRank
Classification attributionImageNet correct samples
Insertion Score0.8679
19
Hallucination AttributionPOPE
Ins Score93.53
16
Hallucination AttributionRePOPE
Insertion Score80.45
16
Image Classification AttributionImageNet Cause
Insertion Score75.27
12
Image Classification AttributionImageNet (Repair)
Insertion (Ins)56.32
12
Visual Attribution-based Hallucination RepairRePOPE
Repair Rate96
8
Showing 6 of 6 rows

Other info

Follow for update