STAP: A Shuffle-Tokenized App Predictor with Ultra Long Context for Vocabulary-Free Mobile App Prediction

• URL: http://arxiv.org/abs/2605.29863v1
• Date: Thu, 28 May 2026 12:44:54 GMT
• Title: STAP: A Shuffle-Tokenized App Predictor with Ultra Long Context for Vocabulary-Free Mobile App Prediction
• Abstract summary: Existing models rely on fixed app vocabularies, which prevents them from generalizing across different app ecosystems.<n>We propose STAP, a Transformer-based model that eliminates the need for a fixed vocabulary.<n>Experiments on two datasets from different continents demonstrate that STAP achieves strong cross-dataset zero-shot prediction accuracy.
• Score: 3.5809657956404135
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Predicting the next mobile application a user will launch is essential for intelligent device resource management and proactive assistance. Existing models rely on fixed app vocabularies, which prevents them from generalizing across different app ecosystems. Many also depend on user-specific knowledge, which complicates deployment in cold start scenarios. We propose STAP, a Transformer-based model that eliminates the need for a fixed vocabulary. STAP replaces true app identities with randomly reassigned virtual indices via a shuffle mechanism, and compensates for discarded semantic information by processing behavioral sequences with an ultra-long context design. A theoretical analysis shows that, given a sufficiently long context, the predicted distribution converges to the correct one despite the anonymity of the mapping. Experiments on two datasets from different continents demonstrate that STAP achieves strong cross-dataset zero-shot prediction accuracy -- a setting where all existing fixed-vocabulary methods are inherently inapplicable -- while its cold start performance within each dataset remains competitive with leading models. Furthermore, we introduce a deployment strategy that enables the model to retain a sufficiently long context during continuous inference while keeping latency within acceptable bounds.

Related papers

• Task-State Adaptation with Prototype Memory for Multi-Task Dense Prediction [65.90697836568579]
  We propose a multi-task dense prediction framework that estimates a compact task state from global visual context.<n>MemMTL refines it through a learnable task-state prototype memory.<n>A separate task-agnostic residual bank provides a common adaptation path.
  arXiv  Detail & Related papers  (2026-08-28T08:48:22Z)
• Multi-Task Bayesian In-Context Learning [41.16196712546687]
  We introduce a multi-task in-context learning framework for amortized hierarchical Bayesian predictive inference.<n>A transformer trained on sequences of prior and target tasks learns to adapt its predictions across families of priors.<n>Our method matches oracle predictors while being orders of magnitude faster than existing approaches.
  arXiv  Detail & Related papers  (2026-06-18T17:50:10Z)
• RAID: Semantic Graph Diffusion for True Cold-Start and Cross-Lingual Forecasting [1.8923689868452591]
  RAID (Retrieval-Augmented Iterative Diffusion) is a framework that replaces history-based correlation learning with semantic retrieval and graph-conditioned diffusion.<n> RAID maps textual metadata into a shared semantic space using a frozen multilingual embedding model and constructs an inductive retrieval graph that extends naturally to unseen items.<n>Under a strict cold-start protocol, RAID outperforms strong foundation models and competitive baselines on both forecasting accuracy and prediction interval coverage.
  arXiv  Detail & Related papers  (2026-06-15T16:27:14Z)
• TapSampling: Inference-Time Sampling with a Task-Progress-Understanding Verifier for Robotic Manipulation [61.35569005726248]
  Existing embodied control research demonstrates remarkable performance improvements by scaling training data and model size.<n>Non-deterministic generative models, such as diffusion and autoregressive models, have been widely adopted in the field of embodied control.<n>We propose textbfTapSampling, a plug-and-play framework for inference-time sampling.
  arXiv  Detail & Related papers  (2026-05-25T08:03:31Z)
• OVI-MAP:Open-Vocabulary Instance-Semantic Mapping [108.66131262110095]
  OVI-MAP is a class-agnostic 3D instance map that is incrementally constructed from RGB-D input.<n> semantic features are extracted only from a small set of automatically selected views using vision-language models.<n>Our system operates in real time and outperforms state-of-the-art open-vocabulary mapping baselines on standard benchmarks.
  arXiv  Detail & Related papers  (2026-03-27T15:50:59Z)
• MISApp: Multi-Hop Intent-Aware Session Graph Learning for Next App Prediction [4.997426047348826]
  We propose MISApp, a profile-free framework for next app prediction based on multi-hop session graph learning.<n> MISApp constructs multi-hop session graphs to capture transition dependencies at different structural ranges.<n>Experiments show that MISApp consistently outperforms competitive baselines under both standard and cold-start settings.
  arXiv  Detail & Related papers  (2026-03-23T07:30:45Z)
• Matching Features, Not Tokens: Energy-Based Fine-Tuning of Language Models [102.20309135516186]
  Cross-entropy (CE) training provides dense and scalable supervision for language models.<n>We introduce a feature-matching objective for language-model fine-tuning.<n>To optimize this objective efficiently, we propose energy-based fine-tuning.
  arXiv  Detail & Related papers  (2026-03-12T17:57:50Z)
• FitPro: A Zero-Shot Framework for Interactive Text-based Pedestrian Retrieval in Open World [13.089848592467675]
  Text-based Pedestrian Retrieval (TPR) deals with retrieving specific target pedestrians in visual scenes according to natural language descriptions.<n>We propose FitPro, an open-world interactive zero-shot TPR framework with enhanced semantic comprehension and cross-scene adaptability.
  arXiv  Detail & Related papers  (2025-09-20T12:55:18Z)
• MobText-SISA: Efficient Machine Unlearning for Mobility Logs with Spatio-Temporal and Natural-Language Data [6.013377534625831]
  We introduce MobText-SISA, a scalable machine-unlearning framework that extends Sharded, Isolated Slice, and AggregatedSIS (A) training to heterogeneous-temporal data.<n>We show that MobText-SISA consistently outperforms random sharding in both error and convergence speed.<n>These results establish MobText-SISA as a practical foundation for privacy-compliant analytics on mobility data at urban scale.
  arXiv  Detail & Related papers  (2025-08-27T03:55:16Z)
• PATFinger: Prompt-Adapted Transferable Fingerprinting against Unauthorized Multimodal Dataset Usage [19.031839603738057]
  multimodal datasets can be leveraged to pre-train vision-adapted models by providing cross-modal semantics.<n>We propose a novel prompt-language transferable fingerprinting scheme called PATFinger.<n>Our scheme utilizes inherent dataset attributes as fingerprints instead of compelling the model to learn triggers.
  arXiv  Detail & Related papers  (2025-04-15T09:53:02Z)
• TGT: A Temporal Gating Transformer for Smartphone App Usage Prediction [8.738528109590863]
  We propose a Transformer framework equipped with a temporal gating module that conditions hidden representations on the hour-of-day.<n>TGT significantly outperforms 15 competitive baselines, achieving notable gains in HR@1 and maintaining robustness under cold-start scenarios.
  arXiv  Detail & Related papers  (2025-02-24T08:31:51Z)
• PeFAD: A Parameter-Efficient Federated Framework for Time Series Anomaly Detection [51.20479454379662]
  We propose a. Federated Anomaly Detection framework named PeFAD with the increasing privacy concerns. We conduct extensive evaluations on four real datasets, where PeFAD outperforms existing state-of-the-art baselines by up to 28.74%.
  arXiv  Detail & Related papers  (2024-06-04T13:51:08Z)
• STOP: A dataset for Spoken Task Oriented Semantic Parsing [66.14615249745448]
  End-to-end spoken language understanding (SLU) predicts intent directly from audio using a single model. We release the Spoken Task-Oriented semantic Parsing (STOP) dataset, the largest and most complex SLU dataset to be publicly available. In addition to the human-recorded audio, we are releasing a TTS-generated version to benchmark the performance for low-resource domain adaptation of end-to-end SLU systems.
  arXiv  Detail & Related papers  (2022-06-29T00:36:34Z)

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.