Pith. sign in

REVIEW 7 cited by

Mapping Natural Language Instructions to Mobile UI Action Sequences

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2005.03776 v2 pith:BJJGV4P6 submitted 2020-05-07 cs.CL cs.LG

Mapping Natural Language Instructions to Mobile UI Action Sequences

classification cs.CL cs.LG
keywords actioninstructionslanguagemobileactionsnaturalcreatedescriptions
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

We present a new problem: grounding natural language instructions to mobile user interface actions, and create three new datasets for it. For full task evaluation, we create PIXELHELP, a corpus that pairs English instructions with actions performed by people on a mobile UI emulator. To scale training, we decouple the language and action data by (a) annotating action phrase spans in HowTo instructions and (b) synthesizing grounded descriptions of actions for mobile user interfaces. We use a Transformer to extract action phrase tuples from long-range natural language instructions. A grounding Transformer then contextually represents UI objects using both their content and screen position and connects them to object descriptions. Given a starting screen and instruction, our model achieves 70.59% accuracy on predicting complete ground-truth action sequences in PIXELHELP.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 7 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments

    cs.AI 2024-04 accept novelty 8.0

    OSWorld provides the first unified real-computer benchmark for open-ended multimodal agent tasks, exposing large performance gaps between humans and state-of-the-art LLM/VLM agents.

  2. ScaleWoB: Guiding GUI Agents with Coding Agents via Large-Scale Environmental Synthesis

    cs.AI 2026-05 unverdicted novelty 7.0

    ScaleWoB generates 100+ synthetic interactive GUI environments and 1000+ verifiable tasks as web pages, releasing a 120-task mobile benchmark where state-of-the-art agents achieve 27.92% success (17.82% on long-horizo...

  3. Grounding Computer Use Agents on Human Demonstrations

    cs.LG 2025-11 conditional novelty 7.0

    GroundCUA, a 3.56M-element human-annotated desktop grounding dataset, and GroundNext models achieve strong UI grounding with less than one-tenth the SFT data of prior work.

  4. GUI-R1 : A Generalist R1-Style Vision-Language Action Model For GUI Agents

    cs.CV 2025-04 unverdicted novelty 7.0

    GUI-R1 uses reinforcement fine-tuning with GRPO on a small curated dataset to create a generalist vision-language action model that outperforms prior GUI agent methods across mobile, desktop, and web benchmarks using ...

  5. LogiDroid: Individual Functional Test Generation via Business Logic Extraction and Adaptation

    cs.SE 2026-02 conditional novelty 6.0

    LogiDroid generates functional Android test cases with verification assertions by retrieving similar test cases, fusing their business logic, and adapting it to the target app's real-time GUI state.

  6. Mobile GUI Agents under Real-world Threats: Are We There Yet?

    cs.CR 2025-07 conditional novelty 6.0

    Introduces an app-content instrumentation framework and benchmark showing that examined GUI agents suffer 42.0% and 36.1% average misleading rates from third-party content in dynamic and static tests respectively.

  7. SeeClick: Harnessing GUI Grounding for Advanced Visual GUI Agents

    cs.HC 2024-01 unverdicted novelty 6.0

    SeeClick improves visual GUI agents via GUI grounding pre-training on automatically curated data and introduces the ScreenSpot benchmark, with results indicating that stronger grounding boosts downstream task performance.