Pith. sign in

REVIEW 6 cited by

Active Learning by Acquiring Contrastive Examples

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

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 2109.03764 v1 pith:WXM3UFDF submitted 2021-09-08 cs.CL cs.AIcs.LG

Katerina Margatina , Giorgos Vernikos , Loïc Barrault , Nikolaos Aletras This is my paper · ORCID
classification cs.CLcs.AIcs.LG
keywords dataacquisitionactivecontrastivelearningbestbetterdatasets
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Common acquisition functions for active learning use either uncertainty or diversity sampling, aiming to select difficult and diverse data points from the pool of unlabeled data, respectively. In this work, leveraging the best of both worlds, we propose an acquisition function that opts for selecting \textit{contrastive examples}, i.e. data points that are similar in the model feature space and yet the model outputs maximally different predictive likelihoods. We compare our approach, CAL (Contrastive Active Learning), with a diverse set of acquisition functions in four natural language understanding tasks and seven datasets. Our experiments show that CAL performs consistently better or equal than the best performing baseline across all tasks, on both in-domain and out-of-domain data. We also conduct an extensive ablation study of our method and we further analyze all actively acquired datasets showing that CAL achieves a better trade-off between uncertainty and diversity compared to other strategies.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 6 Pith papers

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

  1. A computational fluid dynamics model for the simulation of flashboiling flow inside pressurized metered dose inhalers

    physics.flu-dyn 2025-08 unverdicted novelty 6.0 of 10

    The abstract claims a first-of-kind open-source CFD model, combining Volume-of-Fluid and cavitation modeling, that quantitatively predicts flashboiling flow in pressurized metered dose inhalers, but the supplied text ...

  2. Decouple before Align: Visual Disentanglement Enhances Prompt Tuning

    cs.CV 2025-08 conditional novelty 6.0 of 10

    Decoupling images into foreground and background before aligning them with text improves CLIP prompt tuning on few-shot and generalization benchmarks.

  3. Foundation Model Insights and a Multi-Model Approach for Superior Fine-Grained One-shot Subset Selection

    cs.CV 2025-06 conditional novelty 6.0 of 10

    RAM-APL combines distance rankings and pseudo-class label accuracy from two foundation models to select training subsets, outperforming twelve baselines on fine-grained image datasets.

  4. Info-Coevolution: An Efficient Framework for Data Model Coevolution

    cs.LG 2025-06 conditional novelty 6.0 of 10

    A data-model coevolution framework that fuses model and nearest-neighbor predictions to select labels, reaching ImageNet-1K accuracy with 68% of annotations and 50% under semi-supervised training.

  5. \emph{FoQuS}: A Forgetting-Quality Coreset Selection Framework for Automatic Modulation Recognition

    cs.LG 2025-09 reject novelty 5.0 of 10

    FoQuS uses three training-dynamics scores to select a 1% to 30% coreset for AMR and claims higher accuracy than ten existing coreset methods.

  6. NMS: Efficient Edge DNN Training via Near-Memory Sampling on Manifolds

    cs.LG 2025-08 reject novelty 5.0 of 10

    A t-SNE-based sampling algorithm with differential evolution and near-memory hardware is claimed to speed up edge DNN training and reduce memory energy.

Pith tools