Pith. sign in

REVIEW 5 major objections 5 minor 50 references

BANER: Boundary-Aware LLMs for Few-Shot Named Entity Recognition

T0 review · 5 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read BANER claims that a LLaMA-2-7B fine-tuned with boundary-aware contrastive learning and LoRAHub composition outperforms prior methods on few-shot NER benchmarks.

desk verdict A plausible boundary-aware contrastive recipe for few-shot NER with LLMs, but the headline numbers are partly compromised by hidden-layer selection on the GUM test set. read the letter →

arxiv 2412.02228 v1 pith:KUHZWVBZ submitted 2024-12-03 cs.CL cs.AIcs.LG

Quanjiang Guo , Yihong Dong , Ling Tian , Zhao Kang , Yu Zhang , Sijie Wang This is my paper · ORCID
classification cs.CLcs.AIcs.LG
keywords few-shotnamedentityrecognitionboundary-awarecontrastivelearninglargelanguagemodelsLoRAdomainadaptationprototypicalnetworksspandetectioncross-domainNER
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

BANER is a two-stage approach to few-shot named entity recognition built on a large language model. The paper claims that adding a boundary-aware contrastive loss during the span-detection stage fixes the over- and under-detected spans that plague two-stage prototypical networks, while composing per-domain LoRA modules with LoRAHub aligns target-domain entity prototypes with source-domain knowledge during type classification. If correct, this means a 7-billion-parameter model can be adapted to new entity types with only a handful of labeled examples and outperform both one-stage and prior two-stage few-shot NER methods. The authors report gains over the strongest baselines on Few-NERD and on four cross-dataset benchmarks, and show the same recipe improves Mistral-7B and LLaMA-3-8B under the GUM 5-shot setting. That matters because general-purpose LLMs have not usually proven to be reliable few-shot information extractors, and few-shot NER is the setting where new entity vocabularies appear constantly.

What carries the argument

The load-bearing object is the boundary-aware contrastive loss $\mathcal{L}_{cl}$ computed from hidden states of the 25th layer of LLaMA-2. For each candidate span, the positive sample is the concatenation $[h_{b_i}, h_{e_i-1}]$ of the hidden states of the span's first and last token, the negative sample is the concatenation of the hidden states of the two tokens before and two tokens after the span, and the anchor is the embedding of the entity-type phrase in the prompt; the loss is $-\frac{1}{B}\sum_{i=1}^{B}\log\sigma(\text{sim}(o, pos_i) - \text{sim}(o, neg_i))$ with cosine similarities. This loss is added to the autoregressive span-generation loss with weight $\lambda = 0.001$. The second mechanism is LoRAHub composition $\hat{m} = (w_1 A_1 + \cdots + w_N A_N)(w_1 B_1 + \cdots + w_N B_N)$, learned by minimizing cross-entropy plus L1 regularization on the weights, which merges per-domain LoRA adapters so target-domain prototypes stay aligned with source-domain knowledge.

What would settle it

Take the span-detection stage with and without the boundary-aware contrastive loss under identical LoRA fine-tuning and evaluate on a held-out dataset such as WNUT 5-shot; if removing the loss does not reduce over- and under-detected spans or lower F1, the loss is not doing the claimed work. A second check is to repeat the hidden-layer comparison from Section 4.5 on a different dataset; if the 25th layer is not the best there, the layer choice was overfit to GUM.

Watch

Extended reading notes

Core claim

The paper's central claim is that few-shot NER improves when an LLM is taught entity boundaries directly and when type classification is adapted across domains in a parameter-efficient way. In the span-detection stage, BANER fine-tunes the LLM with the standard generation loss plus a boundary-aware contrastive loss: the embedding of the entity-type phrase is pulled toward the concatenated hidden states of the true span's first and last token and pushed away from the hidden states of tokens just outside the span. In the type-classification stage, a prototypical network is built on the LLM's span representations, separate LoRA modules are trained per target domain, and LoRAHub composes those modules to align target-domain prototypes with source-domain knowledge. Experiments on Few-NERD and on the OntoNotes-to-I2B2, CoNLL, WNUT, and GUM transfer settings report that BANER surpasses the compared one-stage and two-stage baselines, including average F1 gains of 5.2 points on Few-NERD intra and 2.3 and 5.1 points on the 1-shot and 5-shot cross-dataset settings.

Load-bearing premise

The method's gains rest on the assumption that pulling the entity-type phrase embedding toward the concatenated embeddings of a true span's first and last token, and away from tokens just outside the span, improves how the LLM perceives entity boundaries, and that the hidden layer chosen on the GUM 5-shot set transfers to other domains.

Editorial extensions

If this is right

  • A single 7B LLM can be turned into a competitive few-shot NER system by training only low-rank adapters, not full model weights.
  • Boundary-aware contrastive learning reduces false spans early, so the type-classification stage receives cleaner candidates and the two-stage pipeline errors do not compound.
  • LoRAHub composition means a model pretrained once on a source domain can be adapted to several target domains by training and then reweighting small domain-specific modules.
  • The same two enhancements transfer to other open LLMs, so the recipe is not tied to one backbone.
  • The method's largest margins are reported on dense-entity and intra-domain settings where boundary errors dominate, suggesting boundary perception is the bottleneck being addressed.

Reading between the lines

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

  • Because the hidden layer for the contrastive embeddings was selected on the GUM 5-shot set, the strongest test of the mechanism would hold out the layer choice on a second dataset; if layer 25 is not consistently best, part of the reported margin may come from fitting that evaluation set.
  • The contrastive loss uses only the span's first and last token hidden states; a natural variant would include interior span tokens or full-span pooling, which could improve recall on long entities or numeric spans.
  • LoRAHub's benefit likely grows with the number of target domains available at composition time; in a pure single-domain 5-shot transfer to one dataset its advantage over one fine-tuned LoRA module may be smaller than the ablation suggests.
  • A direct probe of the mechanism would inspect the hidden-state geometry before and after training: if the loss works as intended, the cosine similarity between entity-type phrase embeddings and true-span boundary embeddings should increase relative to outside-token embeddings.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

5 major / 5 minor

Summary. The paper proposes BANER, a two-stage few-shot NER framework built on LLaMA-2-7B with LoRA. The first stage performs span detection with an auto-regressive generation loss augmented by a boundary-aware contrastive loss that pulls entity-type phrase embeddings toward the embeddings of true span boundary tokens. The second stage performs type classification via prototypical networks, with per-domain LoRA modules composed by LoRAHub for cross-domain adaptation. The authors report state-of-the-art F1 scores on Few-NERD and four cross-dataset benchmarks, and provide ablations and experiments with Mistral-7B and LLaMA-3-8B.

Significance. If the reported results are unbiased, the paper demonstrates a practical recipe for making 7B-scale LLMs effective few-shot NER systems, with contributions in boundary-aware contrastive learning and LoRA-based domain adaptation. The release of code and data is a concrete strength, as are the ablations that isolate individual components and the evaluation across multiple LLM architectures. However, the significance is currently limited by evaluation-protocol issues: test-set-based selection of the contrastive layer, an underspecified LoRAHub weight-optimization protocol, and a main comparison that does not control for base-model scale. These concerns must be resolved before the state-of-the-art claim can be fully trusted.

major comments (5)
  1. [Section 4.5, Table 4, Figure 4] The hidden layer index (25th) for the boundary-aware contrastive loss is selected by comparing performance on the GUM 5-shot setting, and GUM 5-shot is itself one of the reported test sets in Table 4 and is included in the cross-dataset average. This is test-set-driven hyperparameter selection: the reported GUM 5-shot F1 and the cross-dataset averages are optimistically biased, and the same layer choice propagates into the ablation study in Table 5. Please re-select the layer on a validation split disjoint from all test sets (e.g., a held-out portion of the source-domain data or a GUM development split) and re-report the main tables and ablations under that protocol.
  2. [Section 3.2.3, Eq. (14), Section 4.1.4] The LoRAHub composition protocol is not fully specified. The text does not state whether the weights w are optimized on the support set, on a separate validation set, or on the query (test) set; the L1 coefficient alpha is never assigned a value in Section 4.1.4; and Eq. (14) contains a typographical error ('w1A1 ... + wN AN' should presumably be a sum). If w is fitted on the query set, the cross-dataset results are transductive and not comparable to the published baselines. Please specify the exact optimization procedure, the data split used to determine w, the value of alpha, and the initialization of w.
  3. [Section 3.1.2, Eqs. (6)-(8)] The boundary-aware contrastive loss is not written with consistent notation. In Eq. (6), L_cl is a sum over the batch index i, while Eqs. (7)-(8) define sim(o,pos_i) and sim(o,neg_i) as sums over m spans that return B-dimensional vectors. Moreover, e_o, e_posi, and e_negi have shapes B x 1 x D, B x 2 x D, and B x 4 x D, so the dot products in Eqs. (7)-(8) are not well-defined without a pooling or flattening operation. Please rewrite the loss with consistent indices and explicit tensor shapes (e.g., mean-pooling over the span dimensions) so that the method is reproducible.
  4. [Section 3.2.1, Eq. (10)] The definition of the span representation z_i used to compute prototypes is missing. Eq. (10) averages z_i, but the paper never states how z_i is extracted from the LLM hidden states (which layer, whether the span is mean-pooled, whether the entity type phrase embedding is used). Without this, the type classification stage cannot be reproduced. Please provide the exact construction of z_i.
  5. [Tables 3-4] The main comparison does not control for base-model scale: BANER is built on LLaMA-2-7B, while most baselines (ProtoBERT, NNShot, StructShot, ESD, DecomposedMetaNER, TadNER, TSFNER, BDCP) are based on substantially smaller pretrained encoders. The reported gains in Tables 3 and 4 may therefore reflect capacity rather than the proposed boundary-aware contrastive loss or LoRAHub composition. The authors should add same-backbone baselines (e.g., LLaMA-2-7B fine-tuned with LoRA in the same two-stage framework but without the proposed components) or restrict the claim to 'BANER outperforms prior methods' only under the caveat that the backbones differ.
minor comments (5)
  1. [Section 4.4, Figure 3] The text says the experiments are run 'under the GUM 5-shot setting' but then mentions both 1-shot and 5-shot scenarios; please clarify whether Figure 3 reports one or both settings.
  2. [Section 3.2.2, Eq. (12)] The quantity d(p,s) in Eq. (12) is a cosine similarity, not a distance; rename it to 'similarity' or use a true distance (e.g., 1 minus cosine similarity) to avoid ambiguity.
  3. [Contributions, Introduction] The claim that this is 'the first integration of LLM with contrastive learning for few-shot NER tasks' should be checked against a broader set of recent works; if uncertain, soften the claim to 'to our knowledge' with a more clearly scoped comparison.
  4. [Figure 2 caption] The caption contains an apparent typo: 'LoRAWiki DomainSocial Domain' appears to be a line-break error in the figure text; please fix the caption and the accompanying figure.
  5. [Section 4.1.4] The implementation details state that no validation set is used during source-domain training; given the test-set-based layer selection in Section 4.5, please add a validation split for hyperparameter selection as recommended in Major Comment 1 and describe how it is used.

Circularity Check

1 steps flagged · score 5.0 of 10

GUM 5-shot test set is used to select the contrastive-learning hidden layer and then reported as the GUM 5-shot result, biasing a key cross-dataset number.

  1. fitted input called prediction [Section 4.5 (Impact of Different Hidden Layers); Section 3.1.2, Eqs. 3-5; Table 4]
    "To determine which hidden layer's output in LlaMA-2 captures higher-level abstract information for constructing a better boundary-aware feature space, we compare overall performance by calculating the contrastive learning loss across different hidden layers under the GUM 5-shot setting. ... We observe that the highest F1 score is achieved when calculating the contrastive learning loss on the 25th layer. ... we extract entity type embedding eo, entity token embedding eposi and enegi, from outputs H in R^{B×L×D} of 25th hidden states layer in LlaMA-2."

    The 25th hidden layer is selected by maximizing F1 on the GUM 5-shot test set (Section 4.5), and the same GUM 5-shot F1 (44.0) is then reported in Table 4 as an independent BANER result and included in the 5-shot cross-dataset average (53.9). The reported GUM number is therefore the objective used for model selection, not an unbiased estimate; it also drives the ablation gain for boundary-aware contrastive learning (Table 5: 44.0 vs 39.3 w/o BASD on GUM 5-shot). This is test-set-driven hyperparameter selection presented as a validated prediction.

full rationale

The paper is an empirical system built from standard components (two-stage span detection and classification, LoRA, prototypical networks, contrastive learning, LoRAHub composition); there is no derivational chain in which an output equation is identical to an input by construction, and no load-bearing self-citation. The one exhibited circularity is in the cross-dataset evaluation protocol: Section 4.5 chooses the hidden layer for the boundary-aware contrastive loss by comparing F1 on GUM 5-shot, and Table 4 reports that same GUM 5-shot F1 as evidence, while Section 4.1.4 states no validation set is used. This makes the GUM 5-shot entry and the 53.9 cross-dataset average optimistically biased and weakens the ablation claim for the boundary-aware loss. The Few-NERD results in Table 3 are not subject to this selection loop and provide independent support for the method's general effectiveness. The LoRAHub composition optimization is described as minimizing Lti on the target domain (Eqs. 13-14) without specifying a held-out split, which is a protocol ambiguity but is not separately counted as a circular step because the data used for the weight search is not explicitly identified. Overall, the central claim is partially, not wholly, circular: one headline number reduces to its own selection criterion, while the rest of the evaluation retains independent content.

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

The method introduces no new theoretical entities; all components come from prior work. The free parameters are mostly standard hyperparameters, except the hidden layer choice which is fit to an evaluation set.

free parameters (4)
  • lambda = 0.001
    Weight of the boundary-aware contrastive loss in Eq. 9. Set by hand, no sensitivity analysis.
  • alpha (LoRAHub L1 coefficient) = not specified
    Regularization coefficient in the LoRAHub objective (Section 3.2.3). The paper never gives its value.
  • Hidden layer index for contrastive embeddings = 25
    Chosen by comparing F1 across layers on GUM 5-shot (Section 4.5, Figure 4). This is a hyperparameter selected on an evaluation set.
  • LoRA rank r and scaling alpha = r=32, alpha=16
    Standard LoRA hyperparameters, chosen without tuning details.
assumptions (3)
  • ad hoc to paper The hidden state at the 25th layer of LLaMA-2-7B encodes boundary-relevant semantics for spans and entity type phrases.
    Assumed in Section 3.1.2 and selected empirically in Section 4.5.
  • domain assumption The contrastive loss between entity type embedding and span boundary embeddings will transfer to unseen entity types.
    Central to the span detection contribution; no theoretical or empirical proof beyond the reported F1.
  • ad hoc to paper LoRAHub composition of separately fine-tuned LoRA modules can align source and target domain prototype spaces.
    Borrowed from Huang et al. (2023), applied to few-shot NER without justification for why composition helps category alignment.

how reviews work

0 comments
Cite this review

Pith. "Pith review of BANER: Boundary-Aware LLMs for Few-Shot Named Entity Recognition." pith.science (2026). https://pith.science/paper/KUHZWVBZ

@misc{pith2026241202228,
  author       = {Pith},
  title        = {Pith review of: BANER: Boundary-Aware LLMs for Few-Shot Named Entity Recognition},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KUHZWVBZ}},
  note         = {Machine review of arXiv:2412.02228}
}
read the original abstract

Despite the recent success of two-stage prototypical networks in few-shot named entity recognition (NER), challenges such as over/under-detected false spans in the span detection stage and unaligned entity prototypes in the type classification stage persist. Additionally, LLMs have not proven to be effective few-shot information extractors in general. In this paper, we propose an approach called Boundary-Aware LLMs for Few-Shot Named Entity Recognition to address these issues. We introduce a boundary-aware contrastive learning strategy to enhance the LLM's ability to perceive entity boundaries for generalized entity spans. Additionally, we utilize LoRAHub to align information from the target domain to the source domain, thereby enhancing adaptive cross-domain classification capabilities. Extensive experiments across various benchmarks demonstrate that our framework outperforms prior methods, validating its effectiveness. In particular, the proposed strategies demonstrate effectiveness across a range of LLM architectures. The code and data are released on https://github.com/UESTC-GQJ/BANER.

Figures

Figures reproduced from arXiv: 2412.02228 by the authors.

Figure 1
Figure 1. (a) shows under/over-detected false spans, (b) [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Overall structure of the proposed BANER. (a) Entity span detection with pre-training in the source domain. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. F1 Score for different LLMs under the GUM [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: F1 Score, Recall, and Precision for different [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 5
Figure 5. Figure 5: Example of the prompt in entity span detection. [PITH_FULL_IMAGE:figures/full_fig_p014_5.png]
Figure 6
Figure 6. Figure 6: Example of the prompt in entity entity classification. [PITH_FULL_IMAGE:figures/full_fig_p014_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

50 extracted references · 22 canonical work pages

  1. [1]

    URL: " 'urlintro :=

    ENTRY address author booktitle chapter edition editor howpublished institution journal key month note number organization pages publisher school series title type volume year eprint doi pubmed url lastchecked label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block STRINGS urlintro eprinturl eprintpr...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    Tommaso Mario Buonocore, Claudio Crema, Alberto Redolfi, Riccardo Bellazzi, and Enea Parimbelli. 2023. Localizing in-domain adaptation of transformer-based biomedical language models. Journal of Biomedical Informatics, 144:104431

    work page 2023

  4. [4]

    Chiu and Eric Nichols

    Jason P.C. Chiu and Eric Nichols. 2016. https://doi.org/10.1162/tacl_a_00104 Named entity recognition with bidirectional LSTM - CNN s . Transactions of the Association for Computational Linguistics, 4:357--370

    doi:10.1162/tacl_a_00104 2016

  5. [5]

    Leyang Cui, Yu Wu, Jian Liu, Sen Yang, and Yue Zhang. 2021. https://doi.org/10.18653/v1/2021.findings-acl.161 Template-based named entity recognition using BART . In Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, pages 1835--1845, Online. Association for Computational Linguistics

    doi:10.18653/v1/2021.findings-acl.161 2021

  6. [6]

    Sarkar Snigdha Sarathi Das, Arzoo Katiyar, Rebecca Passonneau, and Rui Zhang. 2022. https://doi.org/10.18653/v1/2022.acl-long.439 CONT ai NER : Few-shot named entity recognition via contrastive learning . In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 6338--6353, Dublin, Ireland. A...

    doi:10.18653/v1/2022.acl-long.439 2022

  7. [7]

    Leon Derczynski, Eric Nichols, Marieke van Erp, and Nut Limsopatham. 2017. https://doi.org/10.18653/v1/W17-4418 Results of the WNUT 2017 shared task on novel and emerging entity recognition . In Proceedings of the 3rd Workshop on Noisy User-generated Text, pages 140--147, Copenhagen, Denmark. Association for Computational Linguistics

    doi:10.18653/v1/w17-4418 2017

  8. [8]

    Ning Ding, Guangwei Xu, Yulin Chen, Xiaobin Wang, Xu Han, Pengjun Xie, Haitao Zheng, and Zhiyuan Liu. 2021. https://doi.org/10.18653/v1/2021.acl-long.248 Few- NERD : A few-shot named entity recognition dataset . In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural...

    doi:10.18653/v1/2021.acl-long.248 2021

Show all 50 references
  1. [9]

    Jinyuan Fang, Xiaobin Wang, Zaiqiao Meng, Pengjun Xie, Fei Huang, and Yong Jiang. 2023. Manner: A variational memory-augmented model for cross domain few-shot named entity recognition. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (...

    2023

  2. [10]

    Alexander Fritzler, Varvara Logacheva, and Maksim Kretov. 2019. https://doi.org/10.1145/3297280.3297378 Few-shot classification in named entity recognition task . In Proceedings of the 34th ACM/SIGAPP Symposium on Applied Computing, SAC '19, page 993–1000, New York, NY, USA. A...

    2019

  3. [11]

    David Grangier and Dan Iter. 2022. The trade-offs of domain adaptation for neural language models. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 3802--3813

    2022

  4. [12]

    Yanzhu Guo, Virgile Rennard, Christos Xypolopoulos, and Michalis Vazirgiannis. 2021. Bertweetfr: Domain adaptation of pre-trained language models for french tweets. In Proceedings of the Seventh Workshop on Noisy User-generated Text (W-NUT 2021), pages 445--450. Association fo...

    2021

  5. [13]

    Chengsong Huang, Qian Liu, Bill Yuchen Lin, Tianyu Pang, Chao Du, and Min Lin. 2023. Lorahub: Efficient cross-task generalization via dynamic lora composition. arXiv preprint arXiv:2307.13269

    2023 arXiv

  6. [14]

    Yucheng Huang, Kai He, Yige Wang, Xianli Zhang, Tieliang Gong, Rui Mao, and Chen Li. 2022. https://aclanthology.org/2022.coling-1.222 COPNER : Contrastive learning with prompt guiding for few-shot named entity recognition . In Proceedings of the 29th International Conference o...

    2022

  7. [15]

    Zhiheng Huang, Wei Xu, and Kai Yu. 2015. https://arxiv.org/abs/1508.01991 Bidirectional lstm-crf models for sequence tagging . arXiv preprint arXiv:1508.01991

    2015 arXiv

  8. [16]

    Bin Ji, Shasha Li, Shaoduo Gan, Jie Yu, Jun Ma, Huijun Liu, and Jing Yang. 2022. https://aclanthology.org/2022.coling-1.159 Few-shot named entity recognition with entity-level prototypical network enhanced by dispersedly distributed prototypes . In Proceedings of the 29th Inte...

    2022

  9. [17]

    Shengjie Ji and Fang Kong. 2024. https://aclanthology.org/2024.lrec-main.116 A novel three-stage framework for few-shot named entity recognition . In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-CO...

    2024

  10. [18]

    Albert Q Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lucile Saulnier, et al. 2023. Mistral 7b. arXiv preprint arXiv:2310.06825

    2023 arXiv

  11. [19]

    Prannay Khosla, Piotr Teterwak, Chen Wang, Aaron Sarna, Yonglong Tian, Phillip Isola, Aaron Maschinot, Ce Liu, and Dilip Krishnan. 2020. https://proceedings.neurips.cc/paper/2020/file/d89a66c7c80a29b1bdbab0f2a1a94af8-Paper.pdf Supervised contrastive learning . In Advances in N...

    2020

  12. [20]

    Guillaume Lample, Miguel Ballesteros, Sandeep Subramanian, Kazuya Kawakami, and Chris Dyer. 2016. https://doi.org/10.18653/v1/N16-1030 Neural architectures for named entity recognition . In Proceedings of the 2016 Conference of the North A merican Chapter of the Association fo...

    2016 doi

  13. [21]

    Dong-Ho Lee, Akshen Kadakia, Kangmin Tan, Mahak Agarwal, Xinyu Feng, Takashi Shibuya, Ryosuke Mitani, Toshiyuki Sekiya, Jay Pujara, and Xiang Ren. 2022. https://doi.org/10.18653/v1/2022.acl-long.192 Good examples make a faster learner: Simple demonstration-based learning for l...

    2022 doi

  14. [22]

    Yongqi Li, Yu Yu, and Tieyun Qian. 2023. https://doi.org/10.18653/v1/2023.findings-emnlp.598 Type-aware decomposed framework for few-shot named entity recognition . In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 8911--8927, Singapore. Associati...

    2023 doi

  15. [23]

    Ilya Loshchilov and Frank Hutter. 2019. https://openreview.net/forum?id=Bkg6RiCqY7 Decoupled weight decay regularization . In 7th International Conference on Learning Representations, ICLR 2019, New Orleans, LA, USA, May 6-9, 2019 . OpenReview.net

    2019

  16. [24]

    Ruotian Ma, Xin Zhou, Tao Gui, Yiding Tan, Linyang Li, Qi Zhang, and Xuanjing Huang. 2022 a . https://doi.org/10.18653/v1/2022.naacl-main.420 Template-free prompt tuning for few-shot NER . In Proceedings of the 2022 Conference of the North American Chapter of the Association f...

    2022 doi

  17. [25]

    Tingting Ma, Huiqiang Jiang, Qianhui Wu, Tiejun Zhao, and Chin-Yew Lin. 2022 b . https://doi.org/10.18653/v1/2022.findings-acl.124 Decomposed meta-learning for few-shot named entity recognition . In Findings of the Association for Computational Linguistics: ACL 2022, pages 158...

    2022 doi

  18. [26]

    Xuezhe Ma and Eduard Hovy. 2016. https://doi.org/10.18653/v1/P16-1101 End-to-end sequence labeling via bi-directional LSTM - CNN s- CRF . In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1064--1074, Berli...

    2016 doi

  19. [27]

    Yubo Ma, Yixin Cao, Yong Hong, and Aixin Sun. 2023. Large language model is not a good few-shot information extractor, but a good reranker for hard samples! In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 10572--10601

    2023

  20. [28]

    Alexander Rietzler, Sebastian Stabinger, Paul Opitz, and Stefan Engl. 2020. Adapt or get left behind: Domain adaptation through bert language model finetuning for aspect-target sentiment classification. In Proceedings of the Twelfth Language Resources and Evaluation Conference...

    2020

  21. [29]

    Minho Ryu, Geonseok Lee, and Kichun Lee. 2022. Knowledge distillation for bert unsupervised domain adaptation. Knowledge and Information Systems, 64(11):3113--3128

    2022

  22. [30]

    Vin Sachidananda, Jason Kessler, and Yi-An Lai. 2021. Efficient domain adaptation of language models via adaptive tokenization. In Proceedings of the Second Workshop on Simple and Efficient Natural Language Processing, pages 155--165

    2021

  23. [31]

    Yongliang Shen, Xinyin Ma, Zeqi Tan, Shuai Zhang, Wen Wang, and Weiming Lu. 2021. https://doi.org/10.18653/v1/2021.acl-long.216 Locate and label: A two-stage identifier for nested named entity recognition . In Proceedings of the 59th Annual Meeting of the Association for Compu...

    2021 doi

  24. [32]

    Jake Snell, Kevin Swersky, and Richard Zemel. 2017. https://proceedings.neurips.cc/paper/2017/file/cb8da6767461f2812ae4290eac7cbc42-Paper.pdf Prototypical Networks for Few -shot Learning . In Advances in Neural Information Processing Systems , volume 30. Curran Associates, Inc

    2017

  25. [33]

    Amber Stubbs and \"O zlem Uzuner. 2015. https://www.sciencedirect.com/science/article/pii/S1532046415001823 Annotating longitudinal clinical narratives for de-identification: The 2014 i2b2/uthealth corpus . Journal of biomedical informatics, 58:S20--S29

    2015

  26. [34]

    Tjong Kim Sang and Fien De Meulder

    Erik F. Tjong Kim Sang and Fien De Meulder. 2003. https://aclanthology.org/W03-0419 Introduction to the C o NLL -2003 shared task: Language-independent named entity recognition . In Proceedings of the Seventh Conference on Natural Language Learning at HLT - NAACL 2003 , pages 142--147

    2003

  27. [35]

    Peiyi Wang, Runxin Xu, Tianyu Liu, Qingyu Zhou, Yunbo Cao, Baobao Chang, and Zhifang Sui. 2022. https://doi.org/10.18653/v1/2022.naacl-main.369 An enhanced span-based decomposition method for few-shot sequence labeling . In Proceedings of the 2022 Conference of the North Ameri...

    2022 doi

  28. [36]

    Shuhe Wang, Xiaofei Sun, Xiaoya Li, Rongbin Ouyang, Fei Wu, Tianwei Zhang, Jiwei Li, and Guoyin Wang. 2023. Gpt-ner: Named entity recognition via large language models. arXiv preprint arXiv:2304.10428

    2023 arXiv

  29. [37]

    Yaqing Wang, Haoda Chu, Chao Zhang, and Jing Gao. 2021. https://doi.org/10.18653/v1/2021.findings-emnlp.139 Learning from language description: Low-shot named entity recognition via decomposed framework . In Findings of the Association for Computational Linguistics: EMNLP 2021...

    2021 doi

  30. [38]

    Ralph Weischedel, Martha Palmer, Mitchell Marcus, Eduard Hovy, Sameer Pradhan, Lance Ramshaw, Nianwen Xue, Ann Taylor, Jeff Kaufman, Michelle Franchini, et al. 2013. https://catalog.ldc.upenn.edu/LDC2013T19 Ontonotes release 5.0 ldc2013t19 . Linguistic Data Consortium, Philade...

    2013

  31. [39]

    Sam Wiseman and Karl Stratos. 2019. Label-agnostic sequence labeling by copying nearest neighbors. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 5363--5369

    2019

  32. [40]

    Tingyu Xie, Qi Li, Jian Zhang, Yan Zhang, Zuozhu Liu, and Hongwei Wang. 2023. Empirical study of zero-shot ner with chatgpt. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 7935--7956

    2023

  33. [41]

    Xiaojun Xue, Chunxia Zhang, Tianxiang Xu, and Zhendong Niu. 2024. Robust few-shot named entity recognition with boundary discrimination and correlation purification. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38, pages 19341--19349

    2024

  34. [42]

    Yi Yang and Arzoo Katiyar. 2020. https://doi.org/10.18653/v1/2020.emnlp-main.516 Simple and effective few-shot named entity recognition with structured nearest neighbor learning . In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)...

    2020 doi

  35. [43]

    Amir Zeldes. 2017. https://doi.org/10.1007/s10579-016-9343-x The gum corpus: Creating multilayer resources in the classroom . Lang. Resour. Eval., 51(3):581–612

    2017 doi

  36. [44]

    Xinghua Zhang, Bowen Yu, Yubin Wang, Tingwen Liu, Taoyu Su, and Hongbo Xu. 2022. https://dl.acm.org/doi/abs/10.1145/3477495.3531976 Exploring modular task decomposition in cross-domain named entity recognition . In Proceedings of the 45th International ACM SIGIR Conference on ...

    2022

  37. [45]

    Yabin Zhang, Haojian Zhang, Bin Deng, Shuai Li, Kui Jia, and Lei Zhang. 2021. Semi-supervised models are strong unsupervised domain adaptation learners. arXiv preprint arXiv:2106.00417

    2021 arXiv

  38. [46]

    Yu Zhang, Kehai Chen, Xuefeng Bai, Zhao Kang, Quanjiang Guo, and Min Zhang. 2024 a . Question-guided knowledge graph re-scoring and injection for knowledge graph question answering. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 8972--8985

    2024

  39. [47]

    Yu Zhang and Zhao Kang. 2024. Tpn: Transferable proto-learning network towards few-shot document-level relation extraction. In 2024 International Joint Conference on Neural Networks (IJCNN), pages 1--9. IEEE

    2024

  40. [48]

    Zhen Zhang, Yuhua Zhao, Hang Gao, and Mengting Hu. 2024 b . Linkner: Linking local named entity recognition models to large language models using uncertainty. In Proceedings of the ACM on Web Conference 2024, pages 4047--4058

    2024

  41. [49]

    Wenxuan Zhou, Sheng Zhang, Yu Gu, Muhao Chen, and Hoifung Poon. 2024. https://openreview.net/forum?id=r65xfUb76p Universal NER : Targeted distillation from large language models for open named entity recognition . In The Twelfth International Conference on Learning Representations

    2024

  42. [50]

    Xudong Zhu, Zhao Kang, and Bei Hui. 2024. Fcds: Fusing constituency and dependency syntax into document-level relation extraction. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), page...

    2024

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.