Skip to main content
Search

# Long Short-Term Memory Network

3 related articles

Subscribe to updates
  1. arXiv

    Continually updated models leak when data was used: TMIA reads confidence trajectories to time sample entry and exit, with timestamp F1 above 83%

    Continually updated models keep a sequence of queryable snapshots, so an attacker can infer not only whether a sample was used in training but when it entered and left. TMIA treats a sample's confidence across snapshots as a time series, labels entry and exit points with a sliding-window bidirectional LSTM, and pairs them chronologically into valid membership intervals. Across three language models and three image classifiers, TMIA reaches timestamp-level F1 above 83% in every language-model setting and improves interval-level F1 over both baselines by at least 21.82%.

  2. Coastal buoys get a three-level wave-height quality flag: one temporal model transfers across stations at a mean binary F1 of 0.856

    Significant wave-height records from a Brazilian coastal buoy network can now receive an hourly three-level quality flag, GOOD, SUSPECT or BAD, instead of relying only on fixed physical-limit and rate-of-change thresholds. The model decides one-sidedly from the current value and the preceding 168 hours, reaching a binary F1 of 0.853 on a pooled controlled-failure test set across six buoys and a mean binary F1 of 0.856 when transferred to unseen stations. In hourly streaming simulation, abrupt, multiplicative and sentinel failures were detected at onset, while drift and stuck-sensor episodes required accumulated evidence.

  3. BiLSTM model predicts distance-education students' academic failure risk by Week 6 from LMS weekly interaction logs, reporting 91.4% accuracy and 92.6% recall

    This study develops a Bidirectional Long Short-Term Memory (BiLSTM) deep learning model that predicts student academic performance and failure risk from weekly interaction patterns in a Learning Management System (LMS), using a dataset of interaction logs from 1,250 distance education students over one semester with features such as material access frequency, forum participation, and assignment submission timing, and reports 91.4% accuracy, 89.2% precision, and 92.6% recall as early as Week 6 of the course, enabling educators to implement timely pedagogical interventions to reduce dropout rates in digital and distance learning environments.

You’re all caught up