Pith. sign in

REVIEW 3 major objections 5 minor 46 references

No Free Lunch in Active Learning: LLM Embedding Quality Dictates Query Strategy Success

T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read No single active-learning query strategy wins across embeddings

desk verdict A genuinely useful frozen-embedding AL benchmark whose main qualitative claims hold, but the title's causal 'quality dictates' framing outruns the MTEB-based evidence and one key figure caption is inconsistent with the protocol. read the letter →

arxiv 2506.01992 v1 pith:TX6FKPUE submitted 2025-05-18 cs.CL cs.AIcs.IR

Lukas Rauch , Moritz Wirth , Denis Huseljic , Marek Herde , Bernhard Sick , Matthias A{ss}enmacher This is my paper · ORCID
classification cs.CLcs.AIcs.IR
keywords activelearningLLMembeddingsquerystrategiesinitialpoolselectionMTEBtextclassificationfrozennofreelunch
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper argues that in deep active learning with frozen LLM embeddings, the best query strategy is not universal: it depends on the quality of the embedding model and on the task. The authors benchmark seven embedding models—five top-ranked on the MTEB leaderboard plus BERT and ModernBERT—across ten text classification tasks and eight query strategies over twenty AL cycles. They find that a diversity-based initial pool (TypiClust) gives a real early-iteration head start when combined with high-quality embeddings, that Margin sampling produces strong spikes on easier tasks, and that Entropy and Badge are the most robust across tasks and models. If these results hold, active-learning practitioners should treat strategy choice as context-dependent and re-evaluate it whenever they change the embedding backbone.

What carries the argument

The machinery is a frozen-embedding active-learning loop: a pre-trained LLM maps each text to a single vector once; a logistic-regression head is retrained from scratch on the growing labeled pool each cycle; a query strategy picks the next batch; and accuracy over 20 cycles is the score. This isolates the embedding's contribution from fine-tuning dynamics. The named machinery includes the query strategies themselves—uncertainty methods (Margin, Entropy), diversity methods (CoreSet, ProbCover, TypiClust), and hybrids (Badge, DropQuery)—plus the initial-pool-selection choice (random vs. TypiClust). MTEB leaderboard rank is used as the operational measure of embedding quality, so the benchmark's comparisons are organized around that ranking.

What would settle it

Take the same ten datasets and query strategies, but order the embedding models by a different quality measure—for example, linear-probe accuracy on an unrelated held-out task or a cluster-separability score—and check whether the strategy rankings still sort by quality. A single low-MTEB model that matches high-MTEB models in Badge's robustness, or a high-MTEB model that shows Margin-style task spikes, would break the claimed link.

Watch

Extended reading notes

Core claim

The central claim is that the effectiveness of an active-learning query strategy is conditioned on the representation it sits on. On the paper's benchmark, no strategy is universally superior: Margin wins most often on datasets with simple, separable classes (AGNews, DBPedia, TREC-6), while Badge and Entropy win on complex multi-class tasks such as Banks77, Yelp-5, and MNLI. The paper also claims that embedding quality modulates initialization: TypiClust, which selects diverse and representative points, outperforms random initial pools mainly when the pool is small (under roughly 300 instances) and the embeddings are high-quality, especially on fine-grained tasks. Higher MTEB-ranked embeddings generally yield faster convergence and larger TypiClust advantages inside the AL loop. The conclusion is that AL strategy choice is a contextual decision, and that a static leaderboard like MTEB is a helpful but incomplete predictor of AL usefulness.

Load-bearing premise

The paper's central conclusion rests on treating MTEB leaderboard rank as a faithful measure of embedding quality for active learning; if MTEB scores do not track the representation properties that query strategies actually exploit, then differences in strategy success cannot be attributed to 'embedding quality'.

Editorial extensions

If this is right

  • On a small labeling budget, starting with a diversity-based pool like TypiClust is likely to pay off when the embedding model is strong; with weaker embeddings or larger initial pools, random selection gets competitive.
  • Margin sampling is a cheap, effective first choice on simple classification tasks, but for complex multi-class tasks Badge or Entropy should be preferred.
  • An active-learning evaluation that reports only a single aggregated winner is misleading; results should be broken down by embedding model and task.
  • Higher-quality embeddings do not just lift final accuracy—they make the AL process converge faster, so their added compute cost can be offset by fewer labeling rounds.
  • Static embedding benchmarks such as MTEB are useful for ranking models but should not be treated as a substitute for task-specific AL evaluation.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • As embedding models continue to improve, the balance among strategies may shift further toward hybrid methods like Badge, which exploit richer representations, and away from pure diversity sampling.
  • A practical rule of thumb can be tested: compute a cheap proxy for embedding structure (cluster purity or per-class separability) and use it to pre-select between Margin, Entropy, and Badge before running any AL loop.
  • The frozen-embedding setting is likely to generalize to other modalities that use fixed pretrained features, so similar quality-dependent strategy rankings may appear in vision or audio active learning, but this would need a separate benchmark to confirm.
  • If MTEB rank is a valid proxy, then one can predict relative AL gains without running AL by comparing embedding models on retrieval and classification benchmarks—this is a testable shortcut the paper does not itself evaluate.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. This paper presents a benchmark study of active learning (AL) on text classification using frozen LLM embeddings. The authors evaluate seven embedding models (five MTEB top-ranked LLMs and two baselines) on ten tasks from the ActiveGLAE benchmark, with seven query strategies and five seeds. They investigate two stages: informed initial pool selection (IPS) via diversity-based methods, and the full AL cycle. The main claims are that (1) TypiClust-based IPS provides early advantages with high-quality embeddings, (2) no single query strategy is universally superior, and (3) the optimal strategy depends on both the embedding model and the task, with Margin performing well on simpler tasks while Badge and Entropy are more robust across settings, and Badge benefits from higher-quality embeddings. The paper also releases an extensible framework built on scikit-activeml.

Significance. If the claims are established, this would be a useful reference benchmark for practitioners using frozen LLM embeddings in AL, and it would strengthen the already-known message that AL strategy choice is context-dependent. The study's strengths include a reasonably broad evaluation (10 datasets, 7 models, 5 seeds), paired seed-level analyses, clear description of the logistic-regression training protocol, and the release of code. The negative result for CoreSet as an IPS strategy and the demonstration that no universal winner exists are valuable empirical findings. However, the central causal claim that "embedding quality dictates query strategy success" rests on an internal inconsistency in the IPS protocol and on an operationalization of quality (MTEB rank) that is confounded with model architecture and scale.

major comments (3)
  1. [Section 4.3, Figure 6] The text in §4.3 states "We conduct all experiments with TypiClust as our best-performing IPS strategy," but Figure 6, which is cited for the finding that higher-quality embeddings lead to higher overall accuracy, is captioned "with random IPS." This is a load-bearing inconsistency: the quality-attribution analysis and the claimed synergy between TypiClust embeddings and AL performance depend on which IPS was used. Please clarify which IPS was used for Figures 5–7 and the associated text. If Figure 6 was produced with random IPS, then it does not by itself support the TypiClust-specific synergy claim; if the caption is wrong, the reporting is unreliable at exactly the point where the paper's central claim is made.
  2. [Section 4.3, Table 3, Limitations] The paper equates "embedding quality" with MTEB leaderboard rank (Table 3) and bases its central claim on this operationalization, yet the Limitations concede that MTEB is "primarily a dynamic retrieval-focused benchmark" that may not generalize to AL. With only seven models, differences in strategy performance are confounded with parameter count (110M–7.8B), embedding dimension (768–4096), pooling strategy, and model family (BERT-derived vs. LLM-based). The observed variation in strategy rankings across models is therefore not uniquely attributable to "quality" in the AL-relevant sense. To support the causal wording in the title and abstract, the authors would need controlled comparisons (e.g., checkpoints of the same architecture with different MTEB scores, or analyses controlling for model size) or should soften the claims from "quality dictates" to "the embedding model, along with its MTEB score, is associated with strategy performance."
  3. [Section 3 (AL cycle budget)] The total labeling budget B is "empirically determined" per dataset by observing the convergence point of the BERT embedding with random sampling (Table 2). This exogenous choice may compress the AL curves of higher-quality embeddings, especially in later cycles, making differences among query strategies harder to detect for those models and potentially biasing the strategy rankings reported in Figures 5 and 7. Please justify this budget-selection rule with respect to fairness across embedding models, or include a sensitivity analysis with alternative budgets (e.g., a fixed budget across models or a budget scaled to each model's convergence).
minor comments (5)
  1. [Footnote 2 and Table 4] There are inconsistent model-name spellings: footnote 2 refers to "GWEN 2.5" and Table 4 to "QWEN 2.7-7B," while the main text and Table 3 use "QWEN2.5-7B"; please standardize the spelling.
  2. [Figure 2 caption] The caption refers to "Banks77" while the dataset is named "Banking77" in Table 2; please use one consistent name.
  3. [Table 5] Some runtime standard deviations are implausibly large relative to the reported means (e.g., TypiClust on QNLI: 01:37 ± 06:44; TypiClust on Yelp5: 03:50 ± 36:56); please verify the formatting or the values, as these look like unit or rounding errors.
  4. [Section 4.2] The phrase "differences exceeding 0.05-0.10 %" should likely read "0.05–0.10 percentage points" (or an absolute accuracy difference ), since the figure reports accuracy differences, not relative percentages.
  5. [Section 4.3, first paragraph] The phrase "Figure 6 illustrates this at the example of the Banks77 and MNLI datasets" should be reworded to "using the example of" or "on the Banks77 and MNLI datasets."

Circularity Check

0 steps flagged · score 0.0 of 10

Empirical benchmark with no derivation that reduces to its inputs; embedding quality is an external MTEB measure, and the central associations are measured rather than constructed.

full rationale

No circular step is present. The paper is an empirical benchmark: it measures accuracies and win rates across 10 datasets, 7 embedding models, 7 query strategies, and 5 seeds, and it makes no claim to derive AL outcomes from fitted parameters or from a self-citation chain. 'Embedding quality' is operationalized by the external MTEB leaderboard (Table 3), not defined by the paper's AL results, so the association between MTEB rank and AL performance is an empirical correlation rather than a tautology. The choice of TypiClust as the IPS is a protocol selection from earlier isolated IPS experiments, not a fitted parameter that forces the later reported strategy rankings. Self-citations such as ActiveGLAE [28] and scikit-activeml [15] supply datasets and infrastructure, but the underlying tasks are standard public NLP datasets and the central claims do not depend on unverified self-cited theorems. Two non-circular reporting/validity concerns should be noted but do not affect the circularity score: Section 4.3 states 'We conduct all experiments with TypiClust as our best-performing IPS strategy,' while Figure 6 is captioned 'with random IPS,' an internal inconsistency that should be resolved; and the Limitations concede that MTEB is 'primarily a dynamic retrieval-focused benchmark' whose generalization to AL is qualified. These concern correctness and interpretation, not circular construction. The benchmark is self-contained against the external MTEB standard, so the derivation chain is not circular.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central claims rest on the choice of MTEB quality proxy, the frozen-embedding linear-probing setup, and the per-dataset budgets chosen from BERT+random convergence. The budget choice is a potential confounder for cross-model strategy rankings. The MTEB proxy is explicitly qualified by the authors in the Limitations. No new entities are introduced.

free parameters (3)
  • Per-dataset AL budget B = 1000; 5000; 500; 3500; 4000; 4500; 500; 1000; 3000; 2500 (Table 2)
    Chosen by observing BERT+random baseline convergence; sets batch size b=(B-k0)/20 and conditions all strategy comparisons (Section 3).
  • Number of AL cycles T = 20
    Fixed by design; strategy robustness and convergence claims are measured over 20 cycles.
  • Embedding model set (7 models) = MTEB ranks 1-4, 6 plus BERT and ModernBERT
    Hand-selected top MTEB models; small N limits the strength of the quality-to-strategy correlation.
assumptions (4)
  • domain assumption MTEB leaderboard rank is a valid proxy for embedding quality in active learning
    Used in Table 3 and Section 4.3 to rank models as high- or low-quality; the paper itself notes MTEB is retrieval-focused and may not generalize to AL (Limitations).
  • domain assumption Frozen embeddings with linear (logistic regression) probing are representative of deep AL with LLMs
    The entire benchmark relies on this setting to isolate embedding quality; the paper acknowledges it does not cover full fine-tuning (Limitations).
  • ad hoc to paper Per-dataset budgets chosen from BERT+random convergence provide a fair comparison across embedding models and strategies
    Budgets in Table 2 are set by observing BERT+random baseline convergence; if budgets differed by model, strategy rankings and convergence claims could change (Section 3, AL cycle).
  • domain assumption Standard pool-based AL assumptions: clean oracle labels and a fixed unlabeled pool
    The AL cycle assumes labels are obtained from an oracle and pools are static; standard in AL literature but not tested here.

how reviews work

0 comments
Cite this review

Pith. "Pith review of No Free Lunch in Active Learning: LLM Embedding Quality Dictates Query Strategy Success." pith.science (2026). https://pith.science/paper/TX6FKPUE

@misc{pith2026250601992,
  author       = {Pith},
  title        = {Pith review of: No Free Lunch in Active Learning: LLM Embedding Quality Dictates Query Strategy Success},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TX6FKPUE}},
  note         = {Machine review of arXiv:2506.01992}
}
read the original abstract

The advent of large language models (LLMs) capable of producing general-purpose representations lets us revisit the practicality of deep active learning (AL): By leveraging frozen LLM embeddings, we can mitigate the computational costs of iteratively fine-tuning large backbones. This study establishes a benchmark and systematically investigates the influence of LLM embedding quality on query strategies in deep AL. We employ five top-performing models from the massive text embedding benchmark (MTEB) leaderboard and two baselines for ten diverse text classification tasks. Our findings reveal key insights: First, initializing the labeled pool using diversity-based sampling synergizes with high-quality embeddings, boosting performance in early AL iterations. Second, the choice of the optimal query strategy is sensitive to embedding quality. While the computationally inexpensive Margin sampling can achieve performance spikes on specific datasets, we find that strategies like Badge exhibit greater robustness across tasks. Importantly, their effectiveness is often enhanced when paired with higher-quality embeddings. Our results emphasize the need for context-specific evaluation of AL strategies, as performance heavily depends on embedding quality and the target task.

Figures

Figures reproduced from arXiv: 2506.01992 by the authors.

Figure 1
Figure 1. Winning IPS strategy frequency (counts), aggregated across 10 datasets and 7 embedding models for varying initial pool sizes (k0). Ties are excluded [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 3
Figure 3. Performance difference (TypiClust IPS vs. random [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Strategies’ pairwise win rates for random (left) and TypiClust (right) IPS. Values indicate the percentage of dataset-model-cycle combinations where the strategy outperforms its opponent. Takeaway: Impact of IPS on AL TypiClust IPS often increases early performance, particularly for complex tasks. This increase is larger for higher-quality embeddings (e.g., NV-EMBED-V2) but typically diminishes over 5-10 AL cycles. … view at source ↗
Figures from the paper (6 more)
Figure 5
Figure 5. Figure 5: Win rate of each strategy being the top performer per AL cycle, aggregated across all embeddings and datasets. Ties are excluded. General results. A primary observation is that higher￾quality embeddings from LLMs enhance the down￾stream classification performance in AL…
Figure 6
Figure 6. Figure 6: Performance comparison using NV-EMBED-V2, BGE, SFR, QWEN2.5, and BERT embeddings on Banks77 (top) and MNLI (bottom) datasets over AL cycles with random IPS. 8 [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: Win rates of being the best strategy per [PITH_FULL_IMAGE:figures/full_fig_p009_7.png]
Figure 8
Figure 8. Figure 8: Performance difference (TypiClust IPS vs. random IPS) over AL cycles (0-20). Rows [PITH_FULL_IMAGE:figures/full_fig_p014_8.png]
Figure 9
Figure 9. Figure 9: Performance comparison using of models and datasets over AL cycles with [PITH_FULL_IMAGE:figures/full_fig_p015_9.png]
Figure 10
Figure 10. Figure 10: Performance comparison using of models and datasets over AL cycles with [PITH_FULL_IMAGE:figures/full_fig_p016_10.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

46 extracted references · 35 canonical work pages

  1. [1]

    Ash, Chicheng Zhang, Akshay Krishnamurthy, John Langford, and Alekh Agarwal

    Jordan T. Ash, Chicheng Zhang, Akshay Krishnamurthy, John Langford, and Alekh Agarwal. Deep batch active learning by diverse, uncertain gradient lower bounds. In International Conference on Learning Representations (ICLR), 2020

    2020

  2. [2]

    Active Learning with LLMs for Partially Observed and Cost-Aware Scenarios

    Nicolás Astorga, Tennison Liu, Nabeel Seedat, and Mihaela van der Schaar. Active Learning with LLMs for Partially Observed and Cost-Aware Scenarios. InAdvances in Neural Information Processing Systems (NeurIPS), 2024

    work page 2024

  3. [3]

    Is margin all you need? An extensive empirical study of active learning on tabular data

    Dara Bahri, Heinrich Jiang, Tal Schuster, and Afshin Rostamizadeh. Is margin all you need? An extensive empirical study of active learning on tabular data. arXiv:2210.03822, 2022

    work page Pith review arXiv 2022

  4. [4]

    Language Models are Few-Shot Learners

    Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, et al. Language Models are Few-Shot Learners. In Advances in Neural Information Processing Systems (NeurIPS), 2020

    work page 2020

  5. [5]

    Emerging properties in self-supervised vision transformers

    Mathilde Caron, Hugo Touvron, Ishan Misra, Hervé Jégou, Julien Mairal, Piotr Bojanowski, and Armand Joulin. Emerging properties in self-supervised vision transformers. In International Conference on Computer Vision (ICCV), 2021

    work page 2021

  6. [6]

    Making your first choice: To address cold start problem in medical active learning

    Liangyu Chen, Yutong Bai, Siyu Huang, Yongyi Lu, Bihan Wen, Alan Yuille, and Zongwei Zhou. Making your first choice: To address cold start problem in medical active learning. In Medical Imaging with Deep Learning (MIDL), 2024

    work page 2024

  7. [7]

    BERT: Pre-training of deep bidirectional transformers for language understanding

    Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of NAACL-HLT, 2019

    work page 2019

  8. [8]

    Revisiting active learning in the era of vision foundation models

    Sanket Rajan Gupte, Josiah Aklilu, Jeffrey J Nirschl, and Serena Yeung-Levy. Revisiting active learning in the era of vision foundation models. Transactions on Machine Learning Research (TMLR), 2024

    work page 2024

Show all 46 references
  1. [9]

    Active learning on a budget: Opposite strategies suit high and low budgets

    Guy Hacohen, Avihu Dekel, and Daphna Weinshall. Active learning on a budget: Opposite strategies suit high and low budgets. In International Conference on Machine Learning (ICML), volume 162, 2022

    2022

  2. [10]

    A Framework and Benchmark for Deep Batch Active Learning for Regression

    David Holzmüller, Viktor Zaverkin, Johannes Kästner, and Ingo Steinwart. A Framework and Benchmark for Deep Batch Active Learning for Regression. Journal of Machine Learning Research (JMLR), 2023

    2023

  3. [11]

    Fast fishing: Approximating bait for efficient and scalable deep active image classification

    Denis Huseljic, Paul Hahn, Marek Herde, Lukas Rauch, and Bernhard Sick. Fast fishing: Approximating bait for efficient and scalable deep active image classification. InJoint European Conference on Machine Learning and Knowledge Discovery in Databases (ECML PKDD) . Springer, 2024

    2024

  4. [12]

    The interplay of uncertainty modeling and deep active learning: An empirical analysis in image classification

    Denis Huseljic, Marek Herde, Yannick Nagel, Lukas Rauch, Paulius Strimaitis, and Bernhard Sick. The interplay of uncertainty modeling and deep active learning: An empirical analysis in image classification. Transactions on Machine Learning Research (TMLR), 2024

    2024

  5. [13]

    Randomness is the root of all evil: more reliable evaluation of deep active learning

    Yilin Ji, Daniel Kaestner, Oliver Wirth, and Christian Wressnegger. Randomness is the root of all evil: more reliable evaluation of deep active learning. In IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2023

    2023

  6. [14]

    LLMs in the Loop: Leveraging Large Language Model Annotations for Active Learning in Low-Resource Languages

    Nataliia Kholodna, Sahib Julka, Mohammad Khodadadi, Muhammed Nurullah Gumus, and Michael Granitzer. LLMs in the Loop: Leveraging Large Language Model Annotations for Active Learning in Low-Resource Languages. In Joint European Conference on Machine Learning and Knowledge Disco...

    2024

  7. [15]

    scikit-activeml: A library and toolbox for active learning algorithms

    Daniel Kottke, Marek Herde, Tuan Pham Minh, Alexander Benz, Pascal Mergard, Atal Rogh- man, Christoph Sandrock, and Bernhard Sick. scikit-activeml: A library and toolbox for active learning algorithms. Preprints, 2021. 10

    2021

  8. [16]

    NV-Embed: Improved Techniques for Training LLMs as Generalist Embedding Models

    Chankyu Lee, Rajarshi Roy, Mengyao Xu, Jonathan Raiman, Mohammad Shoeybi, Bryan Catanzaro, and Wei Ping. NV-Embed: Improved Techniques for Training LLMs as Generalist Embedding Models. In International Conference on Learning Representations (ICLR), 2025

    2025

  9. [17]

    Navigating the Pitfalls of Active Learning Evaluation: A Systematic Framework for Meaningful Performance Assessment

    Carsten Lüth, Till Bungert, Lukas Klein, and Paul Jaeger. Navigating the Pitfalls of Active Learning Evaluation: A Systematic Framework for Meaningful Performance Assessment. In Advances in Neural Information Processing Systems (NeurIPS), 2024

    2024

  10. [18]

    On the importance of effectively adapting pretrained language models for active learning

    Katerina Margatina, Loic Barrault, and Nikolaos Aletras. On the importance of effectively adapting pretrained language models for active learning. In Annual Meeting of the Association for Computational Linguistics (ACL), 2022

    2022

  11. [19]

    ALPBench: A Benchmark for Active Learning Pipelines on Tabular Data

    Valentin Margraf, Marcel Wever, Sandra Gilhuber, Gabriel Marques Tavares, Thomas Seidl, and Eyke Hüllermeier. ALPBench: A Benchmark for Active Learning Pipelines on Tabular Data. arXiv preprint arXiv:2406.17322, 2024

    2024 arXiv

  12. [20]

    Sfr-embedding-2: Advanced text embedding with multi-stage training, 2024

    Rui Meng, Ye Liu, Shafiq Rayhan Joty, Caiming Xiong, Yingbo Zhou, and Semih Yavuz. Sfr-embedding-2: Advanced text embedding with multi-stage training, 2024

    2024

  13. [21]

    MTEB: Massive text embedding benchmark

    Niklas Muennighoff, Nouamane Tazi, Loïc Magne, and Nils Reimers. MTEB: Massive text embedding benchmark. In Conference of the European Chapter of the Association for Compu- tational Linguistics (EACL), 2023

    2023

  14. [22]

    Towards Robust and Reproducible Active Learning Using Neural Networks

    Prateek Munjal, Nasir Hayat, Munawar Hayat, Jamshid Sourati, and Shadab Khan. Towards Robust and Reproducible Active Learning Using Neural Networks. In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022

    2022

  15. [23]

    Gpt-4 technical report

    OpenAI. Gpt-4 technical report. Technical Report arXiv:2303.08774, OpenAI, 2023

    2023 arXiv

  16. [24]

    Maxime Oquab, Timothée Darcet, Théo Moutakanni, Huy V . V o, Marc Szafraniec, Vasil Khalidov, Pierre Fernandez, Daniel Haziza, Francisco Massa, Alaaeldin El-Nouby, Mido Assran, Nicolas Ballas, Wojciech Galuba, Russell Howes, Po-Yao Huang, Shang-Wen Li, Ishan Misra, Michael Rab...

    2024

  17. [25]

    Pedregosa, G

    F. Pedregosa, G. Varoquaux, A. Gramfort, V . Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V . Dubourg, J. Vanderplas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, and E. Duchesnay. Scikit-learn: Machine learning in Python. Journal of Machine Learni...

    2011

  18. [26]

    Improving language understanding by generative pre-training, 2018

    Alec Radford, Karthik Narasimhan, Tim Salimans, and Ilya Sutskever. Improving language understanding by generative pre-training, 2018. OpenAI Technical Report

    2018

  19. [27]

    Language models are unsupervised multitask learners, 2019

    Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever. Language models are unsupervised multitask learners, 2019. OpenAI Blog

    2019

  20. [28]

    Activeglae: A benchmark for deep active learning with transformers

    Lukas Rauch, Matthias Aßenmacher, Denis Huseljic, Moritz Wirth, Bernd Bischl, and Bernhard Sick. Activeglae: A benchmark for deep active learning with transformers. In Joint European Conference on Machine Learning and Knowledge Discovery in Databases (ECML PKDD) , 2023

    2023

  21. [29]

    Towards deep active learning in avian bioacoustics

    Lukas Rauch, Denis Huseljic, Moritz Wirth, Jens Decke, Bernhard Sick, and Christoph Scholz. Towards deep active learning in avian bioacoustics. arXiv:2406.18621, 2024

    2024 arXiv

  22. [30]

    Sentence-bert: Sentence embeddings using siamese bert- networks

    Nils Reimers and Iryna Gurevych. Sentence-bert: Sentence embeddings using siamese bert- networks. In Conference on Empirical Methods in Natural Language Processing and Interna- tional Joint Conference on Natural Language Processing (EMNLP-IJCNLP), 2019

    2019

  23. [31]

    Revisiting uncertainty-based query strategies for active learning with transformers

    Christopher Schröder, Andreas Niekler, and Martin Potthast. Revisiting uncertainty-based query strategies for active learning with transformers. InFindings of the Association for Computational Linguistics (ACL), 2022. 11

    2022

  24. [32]

    Active learning for convolutional neural networks: A core-set approach

    Ozan Sener and Silvio Savarese. Active learning for convolutional neural networks: A core-set approach. In International Conference on Learning Representations (ICLR), 2018

    2018

  25. [33]

    Active learning literature survey

    Burr Settles. Active learning literature survey. Computer Sciences Technical Report 1648, University of Wisconsin–Madison, 2009

    2009

  26. [34]

    Active learning helps pretrained models learn the intended task

    Alex Tamkin, Dat Pham Nguyen, Salil Deshpande, Jesse Mu, and Noah Goodman. Active learning helps pretrained models learn the intended task. In Advances in Neural Information Processing Systems (NeurIPS), 2022

    2022

  27. [35]

    Llms are also effective embedding models: An in-depth overview

    Chongyang Tao, Tao Shen, Shen Gao, Junshuo Zhang, Zhen Li, Zhengwei Tao, and Shuai Ma. Llms are also effective embedding models: An in-depth overview. arXiv:2412.12591, 2024

    2024 arXiv

  28. [36]

    Qwen2.5: A party of foundation models

    Qwen Team. Qwen2.5: A party of foundation models. arXiv:2412.15115, 2024

    2024 arXiv

  29. [37]

    Attention is All you Need

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is All you Need. In Advances in Neural Information Processing Systems (NeurIPS), 2017

    2017

  30. [38]

    Smarter, better, faster, longer: A modern bidirectional encoder for fast, memory efficient, and long context finetuning and inference

    Benjamin Warner, Antoine Chaffin, Benjamin Clavié, Orion Weller, Oskar Hallström, Said Taghadouini, Alexis Gallagher, Raja Biswas, Faisal Ladhak, Tom Aarsen, Nathan Cooper, Griffin Adams, Jeremy Howard, and Iacopo Poli. Smarter, better, faster, longer: A modern bidirectional e...

    2024 arXiv

  31. [39]

    A Cross-Domain Benchmark for Active Learning

    Thorben Werner, Johannes Burchert, Maximilian Stubbemann, and Lars Schmidt-Thieme. A Cross-Domain Benchmark for Active Learning. In Advances in Neural Information Processing Systems (NeurIPS), 2024

    2024

  32. [40]

    C-pack: Packed resources for general chinese embeddings

    Shitao Xiao, Zheng Liu, Peitian Zhang, Niklas Muennighoff, Defu Lian, and Jian-Yun Nie. C-pack: Packed resources for general chinese embeddings. In International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR), 2024

    2024

  33. [41]

    Active learning through a covering lens

    Ofer Yehuda, Avihu Dekel, Guy Hacohen, and Daphna Weinshall. Active learning through a covering lens. In Advances in Neural Information Processing Systems (NeurIPS), 2022

    2022

  34. [42]

    Cold-start active learning through self-supervised language modeling

    Michelle Yuan, Hsuan-Tien Lin, and Jordan Boyd-Graber. Cold-start active learning through self-supervised language modeling. In Conference on Empirical Methods in Natural Language Processing (EMNLP), 2020

    2020

  35. [43]

    A Comparative Survey: Benchmarking for Pool-based Active Learning

    Xueying Zhan, Huan Liu, Qing Li, and Antoni B Chan. A Comparative Survey: Benchmarking for Pool-based Active Learning. In International Joint Conferences on Artificial Intelligence (IJCAI), 2021

    2021

  36. [44]

    A comparative survey of deep active learning

    Xueying Zhan, Qingzhong Wang, Kuan-hao Huang, Haoyi Xiong, Dejing Dou, and Antoni B Chan. A comparative survey of deep active learning. arXiv:2203.13450, 2022

    2022 arXiv

  37. [45]

    Jasper and stella: distillation of sota embedding models

    Dun Zhang, Jiacheng Li, Ziyang Zeng, and Fulong Wang. Jasper and stella: distillation of sota embedding models. arXiv:2412.19048, 2025

    2025 arXiv

  38. [46]

    LabelBench: A Comprehensive Framework for Benchmarking Adaptive Label-Efficient Learning

    Jifan Zhang, Yifang Chen, Gregory Canal, Arnav Mohanty Das, Gantavya Bhatt, Stephen Mussmann, Yinglun Zhu, Jeff Bilmes, Simon Shaolei Du, Kevin Jamieson, et al. LabelBench: A Comprehensive Framework for Benchmarking Adaptive Label-Efficient Learning. Journal of Data-centric Ma...

    2024

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.