Beyond Pass or Fail: Multi-Dimensional Benchmarking of Foundation Models for Goal-based Mobile UI Navigation

• URL: http://arxiv.org/abs/2501.02863v2
• Date: Tue, 11 Feb 2025 13:34:12 GMT
• Title: Beyond Pass or Fail: Multi-Dimensional Benchmarking of Foundation Models for Goal-based Mobile UI Navigation
• Authors: Dezhi Ran, Mengzhou Wu, Hao Yu, Yuetong Li, Jun Ren, Yuan Cao, Xia Zeng, Haochuan Lu, Zexin Xu, Mengqian Xu, Ting Su, Liangchao Yao, Ting Xiong, Wei Yang, Yuetang Deng, Assaf Marron, David Harel, Tao Xie,
• Abstract summary: We propose Sphinx, a benchmark for evaluation of foundation models (FMs) in industrial settings of user interface ( UI) navigation.<n>We evaluate 8 FMs with 20 different configurations using both Google Play applications and WeChat's internal UI test cases.<n>Our results show that existing FMs universally struggle with goal-based testing tasks, primarily due to insufficient UI-specific capabilities.
• Score: 15.80796682874844
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent advances of foundation models (FMs) have made navigating mobile applications (apps) based on high-level goal instructions within reach, with significant industrial applications such as UI testing. While existing benchmarks evaluate FM-based UI navigation using the binary pass/fail metric, they have two major limitations: they cannot reflect the complex nature of mobile UI navigation where FMs may fail for various reasons (e.g., misunderstanding instructions and failed planning), and they lack industrial relevance due to oversimplified tasks that poorly represent real-world scenarios. To address the preceding limitations, we propose Sphinx, a comprehensive benchmark for multi-dimensional evaluation of FMs in industrial settings of UI navigation. Sphinx introduces a specialized toolkit that evaluates five essential FM capabilities, providing detailed insights into failure modes such as insufficient app knowledge or planning issues. Using both popular Google Play applications and WeChat's internal UI test cases, we evaluate 8 FMs with 20 different configurations. Our results show that existing FMs universally struggle with goal-based testing tasks, primarily due to insufficient UI-specific capabilities. We summarize seven lessons learned from benchmarking FMs with Sphinx, providing clear directions for improving FM-based mobile UI navigation.

Related papers

• VLM-Fuzz: Vision Language Model Assisted Recursive Depth-first Search Exploration for Effective UI Testing of Android Apps [6.122273281101832]
  Testing Android apps effectively requires a systematic exploration of the app's possible states. We propose a novel, automated fuzzing approach called VLM-Fuzz for effective UI testing of Android apps.
  arXiv  Detail & Related papers  (2025-04-16T00:19:31Z)
• UI-TARS: Pioneering Automated GUI Interaction with Native Agents [58.18100825673032]
  This paper introduces UI-TARS, a native GUI agent model that solely perceives the screenshots as input and performs human-like interactions. In the OSWorld benchmark, UI-TARS achieves scores of 24.6 with 50 steps and 22.7 with 15 steps, outperforming Claude (22.0 and 14.9 respectively)
  arXiv  Detail & Related papers  (2025-01-21T17:48:10Z)
• GUI Testing Arena: A Unified Benchmark for Advancing Autonomous GUI Testing Agent [24.97846085313314]
  We propose a formalized and comprehensive environment to evaluate the entire process of automated GUI Testing.<n>We divide the testing process into three key subtasks: test intention generation, test task execution, and GUI defect detection.<n>It evaluates the performance of different models using three data types: real mobile applications, mobile applications with artificially injected defects, and synthetic data.
  arXiv  Detail & Related papers  (2024-12-24T13:41:47Z)
• MMIE: Massive Multimodal Interleaved Comprehension Benchmark for Large Vision-Language Models [71.36392373876505]
  We introduce MMIE, a large-scale benchmark for evaluating interleaved multimodal comprehension and generation in Large Vision-Language Models (LVLMs) MMIE comprises 20K meticulously curated multimodal queries, spanning 3 categories, 12 fields, and 102 subfields, including mathematics, coding, physics, literature, health, and arts. It supports both interleaved inputs and outputs, offering a mix of multiple-choice and open-ended question formats to evaluate diverse competencies.
  arXiv  Detail & Related papers  (2024-10-14T04:15:00Z)
• Software Engineering and Foundation Models: Insights from Industry Blogs Using a Jury of Foundation Models [11.993910471523073]
  We analyze 155 FM4SE and 997 SE4FM blog posts from leading technology companies. We observed that while code generation is the most prominent FM4SE task, FMs are leveraged for many other SE activities. Although the emphasis is on cloud deployments, there is a growing interest in compressing FMs and deploying them on smaller devices.
  arXiv  Detail & Related papers  (2024-10-11T17:27:04Z)
• MobileVLM: A Vision-Language Model for Better Intra- and Inter-UI Understanding [37.15649883702765]
  We propose MobileVLM, which includes two additional pre-training stages to enhance both intra- and inter-UI understanding. To address the lack of mobile pre-training data, we built a large Chinese mobile dataset Mobile3M from scratch, which contains 3 million UI pages. Experimental results show MobileVLM excels on both our test set and public mobile benchmarks, outperforming existing VLMs.
  arXiv  Detail & Related papers  (2024-09-23T08:47:54Z)
• MMAU: A Holistic Benchmark of Agent Capabilities Across Diverse Domains [54.117238759317004]
  Massive Multitask Agent Understanding (MMAU) benchmark features comprehensive offline tasks that eliminate the need for complex environment setups. It evaluates models across five domains, including Tool-use, Directed Acyclic Graph (DAG) QA, Data Science and Machine Learning coding, Contest-level programming and Mathematics. With a total of 20 meticulously designed tasks encompassing over 3K distinct prompts, MMAU provides a comprehensive framework for evaluating the strengths and limitations of LLM agents.
  arXiv  Detail & Related papers  (2024-07-18T00:58:41Z)
• GUI Odyssey: A Comprehensive Dataset for Cross-App GUI Navigation on Mobile Devices [61.48043339441149]
  GUI Odyssey consists of 7,735 episodes from 6 mobile devices, spanning 6 types of cross-app tasks, 201 apps, and 1.4K app combos. We developed OdysseyAgent, a multimodal cross-app navigation agent by fine-tuning the Qwen-VL model with a history resampling module.
  arXiv  Detail & Related papers  (2024-06-12T17:44:26Z)
• SEED-Bench: Benchmarking Multimodal LLMs with Generative Comprehension [27.53415400454066]
  We introduce a benchmark named SEED-Bench to assess generative models. SEED-Bench consists of 19K multiple choice questions with accurate human annotations. We evaluate the performance of 18 models across all 12 dimensions, covering both the spatial and temporal understanding.
  arXiv  Detail & Related papers  (2023-07-30T04:25:16Z)
• MMBench: Is Your Multi-modal Model an All-around Player? [114.45702807380415]
  We propose MMBench, a benchmark for assessing the multi-modal capabilities of vision-language models. MMBench is meticulously curated with well-designed quality control schemes. MMBench incorporates multiple-choice questions in both English and Chinese versions.
  arXiv  Detail & Related papers  (2023-07-12T16:23:09Z)
• VideoGLUE: Video General Understanding Evaluation of Foundation Models [89.07145427268948]
  We evaluate video understanding capabilities of foundation models (FMs) using a carefully designed experiment protocol. We jointly profile FMs' hallmark and efficacy efficiency when adapting to general video understanding tasks.
  arXiv  Detail & Related papers  (2023-07-06T17:47:52Z)
• Mobile-Env: Building Qualified Evaluation Benchmarks for LLM-GUI Interaction [28.53259866617677]
  We introduce Mobile-Env, a comprehensive toolkit tailored for creating GUI benchmarks in the Android mobile environment. We collect an open-world task set across various real-world apps and a fixed world set, WikiHow, which captures a significant amount of dynamic online contents. Our findings reveal that even advanced models struggle with tasks that are relatively simple for humans.
  arXiv  Detail & Related papers  (2023-05-14T12:31:03Z)
• FETA: Towards Specializing Foundation Models for Expert Task Applications [49.57393504125937]
  Foundation Models (FMs) have demonstrated unprecedented capabilities including zero-shot learning, high fidelity data synthesis, and out of domain generalization. We show in this paper that FMs still have poor out-of-the-box performance on expert tasks. We propose a first of its kind FETA benchmark built around the task of teaching FMs to understand technical documentation.
  arXiv  Detail & Related papers  (2022-09-08T08:47:57Z)
• ELEVATER: A Benchmark and Toolkit for Evaluating Language-Augmented Visual Models [102.63817106363597]
  We build ELEVATER, the first benchmark to compare and evaluate pre-trained language-augmented visual models. It consists of 20 image classification datasets and 35 object detection datasets, each of which is augmented with external knowledge. We will release our toolkit and evaluation platforms for the research community.
  arXiv  Detail & Related papers  (2022-04-19T10:23:42Z)
• Explore before Moving: A Feasible Path Estimation and Memory Recalling Framework for Embodied Navigation [117.26891277593205]
  We focus on the navigation and solve the problem of existing navigation algorithms lacking experience and common sense. Inspired by the human ability to think twice before moving and conceive several feasible paths to seek a goal in unfamiliar scenes, we present a route planning method named Path Estimation and Memory Recalling framework. We show strong experimental results of PEMR on the EmbodiedQA navigation task.
  arXiv  Detail & Related papers  (2021-10-16T13:30:55Z)

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.