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

PGT: Procedurally Generated Tasks for improving visual grounding in MLLMs

About

Despite remarkable progress in Multimodal Large Language Models (MLLMs), these models still struggle with fine-grained understanding tasks. In this work, we propose Procedurally Generated Tasks (PGT), a simple data-driven framework that serves a dual purpose: inducing fine-grained visual understanding and acting as a low-cost diagnostic tool to identify the source of perception failures. By overlaying unambiguous geometric primitives on images, PGT generate additional dense supervision that disentangles visual grounding capability from semantic priors. Extensive experiments on relational, quantitative, and 3D/depth understanding benchmarks show that PGT yields remarkable gains across diverse architectures. Instruction tuning MLLMs on LLaVA-v1.5-Instruct augmented with PGT data results in improvements of up to +20% on the What'sUp benchmark and +13.3% on CV-Bench-2D, while maintaining general perception capabilities. Moreover, finetuning state-of-the-art MLLMs on PGT data leads to boosts of up to +5.5% on What'sUp and +8.3% on CV-Bench-2D. These findings demonstrate that PGT effectively address the bottleneck of fine-grained perception, revealing that many spatial reasoning deficits stem from inadequate supervision signals rather than inherent architectural or resolution limitations.

Rim Assouel, Amir Bar, Michal Drozdzal, Adriana Romero-Soriano• 2026

Related benchmarks

TaskDatasetResultRank
Object Hallucination EvaluationPOPE
Accuracy90.8
2019
Visual PerceptionBLINK
Accuracy74.8
241
Visual PerceptionMMVP
Accuracy62.7
118
CountingTallyQA
Accuracy77.4
67
Visual PerceptionAI2D
Accuracy84
47
3D Computer Vision BenchmarkingCVBench3D
Accuracy86
24
Relational ReasoningVSR
Accuracy85.7
16
Vision-Centric / 3DCV-Bench 2D
Accuracy82.5
16
Vision-Centric PerceptionRealworldQA
Accuracy69.3
16
General PerceptionMMStar
Accuracy68.5
16
Showing 10 of 13 rows

Other info

Follow for update