Pith. sign in

REVIEW 4 major objections 6 minor 64 references

Emotion-aware Dual Cross-Attentive Neural Network with Label Fusion for Stance Detection in Misinformative Social Media Content

T0 review · 4 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

Pith's one-line read SPLAENet reports higher accuracy and F1 on three stance benchmarks by fusing emotion, attention, and label distance.

desk verdict A plausible new architecture whose headline 'SOTA' gains evaporate once you compare against the best baseline rather than the average of weak ones; worth a revise-and-resubmit, not a desk reject. read the letter →

arxiv 2505.23812 v1 pith:URAYKZX4 submitted 2025-05-27 cs.CL cs.AIcs.LG

classification cs.CLcs.AIcs.LG
keywords stancedetectiondualcross-attentionemotionsynthesislabelfusiondistancemetriclearningmisinformationsocialmediaRoBERTa
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

The paper aims to show that stance detection in misinformation-heavy social media improves when a model explicitly relates a source post to its reply in both directions, rather than encoding each text alone. To do this it introduces SPLAENet, which combines a two-stage cross-attention module with hierarchical attention, an emotion-alignment signal, and a label-fusion step that measures how close learned features sit to each stance label. On three public benchmarks with different class balance, it reports average accuracy gains between roughly 7 and 10 points and macro-F1 gains between about 11 and 17 points over the compared systems. The reason to care is that stance, whether a reply supports, denies, queries, or comments on a claim, is a practical signal for detecting how misinformation spreads and polarises opinion.

What carries the argument

The load-bearing mechanism is the dual cross-attention module: stage one uses each text's own queries and values but the other text's keys, and stage two uses each text's own queries and keys but the other text's values, with self-attention applied after each stage. This is followed by a hierarchical attention network that pools the attended token vectors into context vectors. Complementing it are three distance-based signals: the L2-normalized absolute difference of [CLS] representations, the absolute difference of top-K emotion embeddings produced by NRCLex and RoBERTa, and a label-fusion step that computes absolute differences between a projected feature vector and each stance-label embedding, concatenates them, and feeds the result into the classification head.

What would settle it

Run SPLAENet and the top officially reported systems for RumourEval, SemEval, and P-Stance under identical train, validation, and test splits with the same preprocessing and hyperparameter search; if SPLAENet's macro-F1 advantage over the strongest published system for each dataset is close to zero or negative, the paper's central significance claim is falsified.

Watch

Extended reading notes

Core claim

The central claim is that modelling the source-reply pair with a dual cross-attention pass, first letting each text attend to the other's keys and then to the other's values, with self-attention between the passes, extracts the inter-textual dependencies that single-encoder classifiers miss. On top of that, SPLAENet treats emotions as a relational signal: it embeds the top three NRCLex emotions of each text and takes their absolute difference as an emotion-divergence vector. It also computes an L2-normalized difference between the [CLS] representations of source and reply, and a label-fusion module that concatenates feature-to-label distances before classification. With these components the model reaches 86.50% accuracy and 51.52% macro-F1 on RumourEval, 75.26% and 72.50% on SemEval, and 85.67% and 85.58% on P-Stance, which the paper presents as the best overall performance among all systems it compares against on all three datasets.

Load-bearing premise

The claim that SPLAENet improves on existing methods rests on the systems in Table 8 being a fair and representative set, trained and evaluated under the same conditions; if stronger published systems were included or the baselines were run under different protocols, the reported gains could narrow substantially or vanish.

Editorial extensions

If this is right

  • On the reported numbers, one architecture covers imbalanced (RumourEval), moderately imbalanced (SemEval), and balanced (P-Stance) datasets without dataset-specific changes.
  • The ablations show every component helps: removing dual cross-attention, hierarchical attention, label fusion, emotion synthesis, or feature closeness lowers accuracy and macro-F1 on all three datasets.
  • The dual cross-attention and hierarchical attention pair is the biggest driver, since removing both costs about 17 F1 points on RumourEval and more than 5 F1 points on SemEval.
  • The label-fusion component makes its largest contribution on SemEval, while emotion synthesis is especially important on P-Stance, indicating the two mechanisms address different failure modes.
  • With a reported 383 million parameters and 0.046-second inference per sample, the method is positioned for deployment-scale content moderation, provided the gains reproduce against the strongest published systems.

Reading between the lines

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

  • The reported average gain is computed against the mean of the systems in Table 8, so the headline margins are a property of that comparison set; a comparison limited to the strongest published system for each dataset would almost certainly show smaller differences.
  • The label-fusion idea transfers naturally to other pairwise-text classification tasks, such as claim verification, argument mining, or reply-quality ranking, where the relationship between two texts rather than the content of either alone determines the label.
  • A cheap testable extension is to replace the one-word stance labels with paraphrased or descriptive label embeddings in the fusion module; the paper's own analysis says label proximity works best when labels are well separated, so richer label text could help the overlapping query and comment classes.
  • The emotion signal is tied to NRCLex's fixed lexicon, so a contextual emotion model trained on social media might strengthen the emotional-divergence vector on sarcastic or informal replies.
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

4 major / 6 minor

Summary. The paper proposes SPLAENet, a stance-detection architecture that combines RoBERTa textual features with a dual cross-attention mechanism, a hierarchical attention network, NRCLex-based emotion features, feature-closeness distance metrics, and a label-fusion module. The model is evaluated on RumourEval, SemEval-2016 Task 6A, and P-Stance, with reported average gains over baseline systems of up to 8.92% accuracy and 17.36% F1 on RumourEval. The main claimed contribution is that jointly modeling inter/intra-textual attention, emotional alignment, and label proximity yields consistent improvements across highly imbalanced, moderately imbalanced, and balanced benchmark datasets.

Significance. If the reported results are validated on the official benchmark splits and against a fair set of published state-of-the-art systems, SPLAENet would be a meaningful contribution: the dual cross-attention design is well motivated, the emotion-synthesis module is a plausible mechanism for stance discrimination, and the label-fusion idea is cleanly described. The paper also provides public code and a relatively extensive ablation study across three datasets, which are strengths. However, the central significance claim currently rests on a comparison protocol that inflates the reported gains, and on a modified RumourEval test split that prevents direct comparison with prior published results.

major comments (4)
  1. [§5.1.2, Table 5] The RumourEval evaluation is run on a modified test split (1,746 posts after preprocessing) rather than the official SemEval-2019 Task 7 test set (1,827 posts), with the total dataset reduced from 8,529 to 8,083 posts. Because the test set is different, the numbers previously published by state-of-the-art systems on the official split are not directly comparable, so the claimed superiority over those systems is not established. The authors should either rerun the model on the official test split or clearly frame the results as a new, differently preprocessed benchmark and avoid claiming state-of-the-art comparison with published RumourEval numbers.
  2. [Abstract and Table 8] The headline average gains (8.92% accuracy and 17.36% F1 on RumourEval, and similar figures on the other datasets) are computed against the arithmetic mean of all listed baselines, including zero-shot LLMs, StanceBERTa, and BERT variants with macro-F1 scores as low as 22.91. Against the strongest listed baseline the gains are much smaller: on RumourEval, +1.19 accuracy and +8.83 F1; on SemEval, +0.36 accuracy and +1.16 F1; on P-Stance, +1.06 accuracy and +1.19 F1. The "significant improvements over state-of-the-art" claim should be based on pairwise comparisons with the best available published systems, not on the mean of a heterogeneous baseline set.
  3. [§5.1.2, Table 4] The dataset counts in Table 4 are internally inconsistent: the RumourEval label counts sum to 6,165 + 819 + 567 + 553 = 8,104, not 8,083 as reported, and the percentages do not correspond to the listed counts. Table 5 also mixes label percentages with split percentages, showing train/val/test rows of 75.11/11.23/7.04 before preprocessing and 50.68/25.39/23.93 after, the latter matching the label distribution of Table 4 rather than the actual split proportions. These inconsistencies undermine the reproducibility of the dataset pipeline and should be corrected and reconciled with the official benchmark statistics.
  4. [§5.3.5, Statistical Analysis] The Friedman test results are reported as evidence that SPLAENet "significantly outperforms" competing methods, but the test only establishes that not all systems perform equally; it does not identify SPLAENet as significantly better than the best baseline. The p-value for P-Stance is 0.0477, which is borderline, and no post-hoc pairwise comparisons or corrections for multiple testing are provided. To support the central claim, the authors should report pairwise significance tests against the strongest baseline on each dataset.
minor comments (6)
  1. [§5.2.1] The sentence "SPLAENet also outperforms StanceBERTa, BERT, FLAN-T5, and RoBERTa [21]" appears to cite the wrong reference for RoBERTa; the correct citation is [20]. Additionally, DeBERTa is mentioned in the surrounding text but is not listed in this sentence.
  2. [§4.2.1, Eq. (3a)] Equation (3a) contains mismatched parentheses in the softmax argument: "softmax( Qs(Hs))Kr(Hr)T)" should be rewritten to make clear that the softmax is applied to Q_s K_r^T / sqrt(d_k). The same formatting issue affects later attention equations.
  3. [§5.2.3] The text says "As seen in the Table 11" when discussing P-Stance performance against existing state-of-the-art methods, but the comparative results are in Table 8; Table 11 reports ablations.
  4. [§5.3.1] In the ablation discussion, the sentence "integrating DCA mechanism leads to a 1.82% increase in accuracy compared to the proposed method" reverses the direction of the comparison; the increase should be relative to the variant without DCA, not relative to the full SPLAENet. The same direction reversal appears in the HAN ablation discussion.
  5. [§5.3.4, Table 12] Table 12 lists LKI-BART twice in the predictions block, and one row is labeled "ZSSD [61]" while the correct reference for ZSSD is [45]. The qualitative discussion also states that Post 3's reply "is categorized as a Query" while the ground truth is Deny, which contradicts the table and the surrounding text; this should be clarified.
  6. [§5.1.2] The P-Stance dataset description says it contains 30,000 annotated tweets, but Table 4 reports a total of 21,574. The authors should reconcile these numbers with the P-Stance paper.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: SPLAENet is a supervised classifier trained on external benchmarks; self-citations appear only as related work and as a label-encoding precedent, not as an unverified load-bearing premise.

full rationale

Walking the paper's derivation chain, the architecture is a standard empirical pipeline: RoBERTa encodes text and label names, dual cross-attention plus hierarchical attention produces context vectors, NRCLex supplies emotion features, an absolute-difference/L2-normalization module provides feature closeness, and a label-fusion module concatenates label-distance features before a dense softmax classifier. Each component is defined by the paper's own equations (e.g., Equations 3 through 17) and is optimized on external training data; the test predictions are not defined in terms of the fitted parameters in any self-referential way. The self-citations to the authors' prior work (Dar et al. [22], Rehman et al. [26], Anshul et al. [41], Chaudhari et al. [42], Raghaw et al. [25]) occur in the related-work review and as a contrast for the label-aware module; none of them is invoked as an unverified uniqueness theorem or as the reason the proposed architecture must work. No equation in the paper reduces the claimed result to an input by construction, and no fitted parameter is renamed as a prediction. The main caveats visible in the manuscript are benchmark-interpretation issues rather than circularity: Table 5 discloses that RumourEval was preprocessed to a different split (8,083 total, test 1,746 vs. the official 1,827), and the abstract's 'average gain' figures are arithmetic differences between SPLAENet and the mean of the 16 listed baselines in Table 8, several of which are zero-shot LLMs with low macro-F1. These choices affect whether the claim 'outperforms state of the art' is externally supported, but they do not make the derivation circular. The ablation study is empirical and could have gone the other way, which further confirms that the central claim is not forced by definition.

Assumptions & free parameters 6 free parameters · 6 assumptions · 0 invented entities

The central result depends on a set of tuned hyperparameters (learning rate, dropout, K, etc.) and on the validity of pretrained representations and emotion lexicons as inputs. No novel entities are postulated.

free parameters (6)
  • learning_rate = 2e-6
    Tuned over 2e-3 to 2e-7 (Section 5.1.1).
  • dropout_rate = 0.2
    Tuned over 0.1 to 0.5 (Section 5.1.1).
  • epochs = 10
    Tuned over 5 to 20 with early stopping (Section 5.1.1).
  • max_sequence_length_U = 50
    Set based on average text length (Section 4.1.1).
  • top_K_emotions = 3
    Set empirically, no details given (Section 4.4).
  • batch_size = 8
    Chosen hyperparameter (Section 5.1.1).
assumptions (6)
  • standard math Scaled dot-product attention is a valid mechanism for relating source and reply tokens.
    Used throughout the attention module, based on Vaswani et al. [18].
  • domain assumption RoBERTa-Large contextual embeddings capture semantics sufficient for stance detection.
    Relied on as the text encoder in Sections 4.1.1 and 4.4.
  • domain assumption NRCLex emotion scores, applied to short noisy tweets, reflect genuine emotional content.
    The emotion synthesis module depends on this (Section 4.4).
  • domain assumption The source-reply pair structure contains the information needed to infer stance.
    The entire problem formulation treats each pair as an independent classification unit (Section 3).
  • domain assumption RumourEval, SemEval, and P-Stance are representative benchmarks and their standard splits are used.
    All evaluation is on these three datasets (Section 5.1.2).
  • domain assumption Encoding top-K emotion words (e.g., 'joy', 'trust') with RoBERTa produces useful emotion representations.
    The emotion synthesis module feeds emotion words to RoBERTa and averages their CLS vectors (Equation 12).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Emotion-aware Dual Cross-Attentive Neural Network with Label Fusion for Stance Detection in Misinformative Social Media Content." pith.science (2026). https://pith.science/paper/URAYKZX4

@misc{pith2026250523812,
  author       = {Pith},
  title        = {Pith review of: Emotion-aware Dual Cross-Attentive Neural Network with Label Fusion for Stance Detection in Misinformative Social Media Content},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/URAYKZX4}},
  note         = {Machine review of arXiv:2505.23812}
}
read the original abstract

The rapid evolution of social media has generated an overwhelming volume of user-generated content, conveying implicit opinions and contributing to the spread of misinformation. The method aims to enhance the detection of stance where misinformation can polarize user opinions. Stance detection has emerged as a crucial approach to effectively analyze underlying biases in shared information and combating misinformation. This paper proposes a novel method for \textbf{S}tance \textbf{P}rediction through a \textbf{L}abel-fused dual cross-\textbf{A}ttentive \textbf{E}motion-aware neural \textbf{Net}work (SPLAENet) in misinformative social media user-generated content. The proposed method employs a dual cross-attention mechanism and a hierarchical attention network to capture inter and intra-relationships by focusing on the relevant parts of source text in the context of reply text and vice versa. We incorporate emotions to effectively distinguish between different stance categories by leveraging the emotional alignment or divergence between the texts. We also employ label fusion that uses distance-metric learning to align extracted features with stance labels, improving the method's ability to accurately distinguish between stances. Extensive experiments demonstrate the significant improvements achieved by SPLAENet over existing state-of-the-art methods. SPLAENet demonstrates an average gain of 8.92\% in accuracy and 17.36\% in F1-score on the RumourEval dataset. On the SemEval dataset, it achieves average gains of 7.02\% in accuracy and 10.92\% in F1-score. On the P-stance dataset, it demonstrates average gains of 10.03\% in accuracy and 11.18\% in F1-score. These results validate the effectiveness of the proposed method for stance detection in the context of misinformative social media content.

Figures

Figures reproduced from arXiv: 2505.23812 by the authors.

Figure 1
Figure 1. The flow diagram of SPLAENet 4.1.1 Textual Feature Extraction Feature extraction entails converting raw text into numerical representations that can be effectively used for further analysis. To extract textual features from user-generated content, we employ a transformer-based deep learning model, RoBERTa [20]. It leverages several key strategies and techniques to enhance its performance and effectiveness in NLP tas… view at source ↗
Figure 2
Figure 2. The illustration of SPLAENet - Stance Prediction through a Label-fused dual cross-Attentive Emotion￾aware neural Network. The proposed methodology consists of six distinct stages: 1. Feature Extraction: In this stage, we extract textual features, denoted as Hs, Hr, from the respective input text Ts and Tr. 2. Feature Interaction: Textual features Hs and Hr are processed through our attention module to capture the in… view at source ↗
Figure 3
Figure 3. Emotion Synthesis Module to capture Emotion Alignment between Source and Reply Text [PITH_FULL_IMAGE:figures/full_fig_p013_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: The t-SNE Visualization of Intermediate Representations from SPLAENet of Datasets with and [PITH_FULL_IMAGE:figures/full_fig_p015_4.png]
Figure 5
Figure 5. Figure 5: Analysis of ROC curves on Datasets (a) RumorEval (b) SemEval (c) P-Stance [PITH_FULL_IMAGE:figures/full_fig_p028_5.png]
Figure 6
Figure 6. Figure 6: Ablation of (a) Attention and (b) Feature Importance on RumourEval Dataset [PITH_FULL_IMAGE:figures/full_fig_p030_6.png]
Figure 7
Figure 7. Figure 7: Ablation of (a) Attention and (b) Feature Importance on the SemEval Dataset [PITH_FULL_IMAGE:figures/full_fig_p031_7.png]
Figure 8
Figure 8. Figure 8: Ablation of (a) Attention and (b) Feature Importance on the P-Stance Dataset [PITH_FULL_IMAGE:figures/full_fig_p033_8.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

64 extracted references · 29 canonical work pages

  1. [1]

    Ruffo, A

    G. Ruffo, A. Semeraro, A. Giachanou, P. Rosso, Studying fake news spreading, polarisation dynamics, and manipulation by bots: A tale of networks and language, Computer Science Review 47 (2023) 100531.doi: https://doi.org/10.1016/j.cosrev.2022.100531

  2. [2]

    F. Olan, U. Jayawickrama, E. O. Arakpogun, J. Suklan, S. Liu, Fake news on social media: the impact on soci- ety, Information Systems Frontiers 26 (2) (2022) 443–458.doi:10.1007/s10796-022-10242-z

  3. [3]

    N. Luo, D. Xie, Y. Mo, F. Li, C. Teng, D. Ji, Joint rumour and stance identification based on se- mantic and structural information in social networks, Applied Intelligence 54.doi:10.1007/ s10489-023-05170-7

  4. [4]

    M. Umer, Z. Imtiaz, S. Ullah, A. Mehmood, G. S. Choi, B.-W. On, Fake news stance detection using deep learning architecture (cnn-lstm), IEEE Access 8 (2020) 156695–156706.doi:10.1109/ACCESS. 2020.3019735. 37

  5. [5]

    W. Li, Y. Xu, G. Wang, Stance detection of microblog text based on two-channel cnn-gru fusion network, IEEE Access 7 (2019) 145944–145952.doi:10.1109/ACCESS.2019.2944136

  6. [6]

    Q. Sun, Z. Wang, S. Li, Q. Zhu, G. Zhou, Stance detection via sentiment information and neural network model, Frontiers of Computer Science 13.doi:10.1007/s11704-018-7150-9

  7. [7]

    H. W. Hanley, Z. Durumeric, Tata: Stance detection via topic-agnostic and topic-aware embeddings, in: EMNLP 2023 - 2023 Conference on Empirical Methods in Natural Language Processing, Proceedings, 2023. doi:10.18653/v1/2023.emnlp-main.694

  8. [8]

    J. Zou, X. Zhao, F. Xie, B. Zhou, Z. Zhang, L. Tian, Zero-shot stance detection via sentiment-stance con- trastive learning, in: 2022 IEEE 34th International Conference on Tools with Artificial Intelligence (ICTAI), 2022, pp. 251–258.doi:10.1109/ICTAI56018.2022.00044

Show all 64 references
  1. [9]

    Bhutani, N

    B. Bhutani, N. Rastogi, P. Sehgal, A. Purwar, Fake news detection using sentiment analysis, in: 2019 Twelfth International Conference on Contemporary Computing (IC3), 2019, pp. 1–5.doi:10.1109/IC3. 2019.8844880

  2. [10]

    Q. Sun, Z. Wang, Q. Zhu, G. Zhou, Stance detection with hierarchical attention network, in: COLING 2018 - 27th International Conference on Computational Linguistics, Proceedings, 2018, (accessed 2025-04-04). URLhttps://aclanthology.org/C18-1203/

  3. [11]

    Huang, J

    W. Huang, J. Yang, A multi-stance detection method by fusing sentiment features, Applied Sciences 14 (9). doi:10.3390/app14093916

  4. [12]

    Karande, R

    H. Karande, R. Walambe, V. Benjamin, K. Kotecha, T. Raghu, Stance detection with bert embeddings for credibility analysis of information on social media, PeerJ Computer Science 7 (2021) e467.doi:10. 7717/peerj-cs.467

  5. [13]

    Rashed, M

    A. Rashed, M. Kutlu, K. Darwish, T. Elsayed, C. Bayrak, Embeddings-based clustering for target specific stances: The case of a polarized turkey, Proceedings of the International AAAI Conference on Web and Social Media 15 (1) (2021) 537–548.doi:10.1609/icwsm.v15i1.18082

  6. [14]

    Y. Fu, X. Li, Y. Li, S. Wang, D. Li, J. Liao, J. Zheng, Incorporate opinion-towards for stance detection, Knowledge-Based Systems 246 (2022) 108657.doi:https://doi.org/10.1016/j.knosys. 2022.108657

  7. [15]

    T. Y. Santosh, S. Bansal, A. Saha, Can siamese networks help in stance detection?, in: Proceedings of the ACM India Joint International Conference on Data Science and Management of Data, CODS-COMAD ’19, Association for Computing Machinery, New York, NY, USA, 2019, p. 306–309.d...

  8. [16]

    Y. Yang, B. Wu, K. Zhao, W. Guo, Tweet stance detection: A two-stage dc-bilstm model based on semantic attention, in: 2020 IEEE Fifth International Conference on Data Science in Cyberspace (DSC), 2020, pp. 22–29.doi:10.1109/DSC50466.2020.00012

  9. [17]

    Q. Pu, F. Huang, F. Li, J. Wei, S. Jiang, Integrating emotional features for stance detection aimed at social network security: A multi-task learning approach, Electronics 14 (1).doi:10.3390/ electronics14010186

  10. [18]

    Vaswani, N

    A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. Kaiser, I. Polosukhin, Attention is all you need, in: Advances in Neural Information Processing Systems, Vol. 2017-December, 2017, pp. 6000 – 6010, (accessed 2025-04-04).doi:10.48550/arXiv.1706.03762

  11. [19]

    Prakash, H

    A. Prakash, H. Tayyar Madabushi, Incorporating count-based features into pre-trained models for improved stance detection, in: G. Da San Martino, C. Brew, G. L. Ciampaglia, A. Feldman, C. Leberknight, P. Nakov (Eds.), Proceedings of the 3rd NLP4IF Workshop on NLP for Internet ...

  12. [20]

    Z. Liu, W. Lin, Y. Shi, J. Zhao, A robustly optimized bert pre-training approach with post-training, in: Chinese Computational Linguistics: 20th China National Conference, CCL 2021, Hohhot, China, Au- gust 13–15, 2021, Proceedings, Springer-Verlag, Berlin, Heidelberg, 2021, p....

  13. [21]

    P. He, X. Liu, J. Gao, W. Chen, Deberta: Decoding-enhanced bert with disentangled attention, in: ICLR 2021 - 9th International Conference on Learning Representations, 2021, (accessed 2025-04-04). URLhttps://openreview.net/pdf?id=sE7-XhLxHA

  14. [22]

    S. S. Dar, M. K. Karandikar, M. Z. U. Rehman, S. Bansal, N. Kumar, A contrastive topic-aware attentive framework with label encodings for post-disaster resource classification, Knowledge-Based Systems 304 (2024) 112526.doi:10.1016/j.knosys.2024.112526

  15. [23]

    Kawintiranon, L

    K. Kawintiranon, L. Singh, Knowledge enhanced masked language model for stance detection, in: K. Toutanova, A. Rumshisky, L. Zettlemoyer, D. Hakkani-Tur, I. Beltagy, S. Bethard, R. Cotterell, 38 T. Chakraborty, Y. Zhou (Eds.), Proceedings of the 2021 Conference of the North Am...

  16. [24]

    Singh, S

    A. Singh, S. S. Dar, R. Singh, N. Kumar, A hybrid similarity-aware graph neural network with transformer for node classification, Expert Systems with Applications 279 (2025) 127292.doi:10.1016/j.eswa. 2025.127292

  17. [25]

    C. S. Raghaw, A. Sharma, S. Bansal, M. Z. U. Rehman, N. Kumar, Cotconet: An optimized coupled transformer-convolutional network with an adaptive graph reconstruction for leukemia detection, Computers in Biology and Medicine 179 (2024) 108821.doi:https://doi.org/10.1016/j. comp...

  18. [26]

    M. Z. U. Rehman, S. Zahoor, A. Manzoor, M. Maqbool, N. Kumar, A context-aware attention and graph neu- ral network-based multimodal framework for misogyny detection, Information Processing & Management 62 (1) (2025) 103895.doi:https://doi.org/10.1016/j.ipm.2024.103895

  19. [27]

    A. Liu, Ca-moeit: Generalizable face anti-spoofing via dual cross-attention and semi-fixed mixture- of-expert, International Journal of Computer Vision 132 (11) (2024) 5439–5452.doi:10.1007/ s11263-024-02135-2

  20. [28]

    Y. Li, Y. Li, S. Zhang, G. Liu, Y. Chen, R. Shang, L. Jiao, An attention-based, context-aware multimodal fusion method for sarcasm detection using inter-modality inconsistency, Knowledge-Based Systems 287 (2024) 111457.doi:10.1016/j.knosys.2024.111457

  21. [29]

    Z. He, N. Mokhberian, K. Lerman, Infusing knowledge from Wikipedia to enhance stance detection, in: J. Barnes, O. De Clercq, V. Barriere, S. Tafreshi, S. Alqahtani, J. Sedoc, R. Klinger, A. Balahur (Eds.), Proceedings of the 12th Workshop on Computational Approaches to Subject...

  22. [30]

    Hardalov, A

    M. Hardalov, A. Arora, P. Nakov, I. Augenstein, Few-shot cross-lingual stance detection with sentiment- based pre-training, in: Proceedings of the 36th AAAI Conference on Artificial Intelligence, AAAI 2022, Vol. 36, 2022.doi:10.1609/aaai.v36i10.21318

  23. [31]

    Zhang, D

    B. Zhang, D. Ding, L. Jing, How would stance detection techniques evolve after the launch of chatgpt?, arXiv preprint arXiv:2212.14548(accessed 2025-04-04).doi:10.48550/arXiv.2212.14548

  24. [32]

    X. Lan, C. Gao, D. Jin, Y. Li, Stance detection with collaborative role-infused llm-based agents, Proceedings of the International AAAI Conference on Web and Social Media 18 (1) (2024) 891–903.doi:10.1609/ icwsm.v18i1.31360

  25. [33]

    Zhang, Y

    Z. Zhang, Y. Li, J. Zhang, H. Xu, LLM-driven knowledge injection advances zero-shot and cross-target stance detection, in: K. Duh, H. Gomez, S. Bethard (Eds.), Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Hu...

  26. [34]

    Mohammad, P

    S. Mohammad, P. Turney, Emotions evoked by common words and phrases: Using Mechanical Turk to create an emotion lexicon, in: D. Inkpen, C. Strapparava (Eds.), Proceedings of the NAACL HLT 2010 Workshop on Computational Approaches to Analysis and Generation of Emotion in Text, ...

  27. [35]

    Kumar, V

    A. Kumar, V. T. Narapareddy, V. Aditya Srikanth, A. Malapati, L. B. M. Neti, Sarcasm detection using multi- head attention based bidirectional lstm, IEEE Access 8 (2020) 6388–6397.doi:10.1109/ACCESS. 2019.2963630

  28. [36]

    P. Li, J. Gu, J. Kuen, V. I. Morariu, H. Zhao, R. Jain, V. Manjunatha, H. Liu, Selfdoc: Self-supervised document representation learning, in: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2021.doi:10.1109/CVPR46437.2021.00560

  29. [37]

    Q. Sun, Z. Wang, Q. Zhu, G. Zhou, Stance detection with hierarchical attention network, in: E. M. Ben- der, L. Derczynski, P. Isabelle (Eds.), Proceedings of the 27th International Conference on Computational Linguistics, Association for Computational Linguistics, Santa Fe, Ne...

  30. [38]

    Zhang, M

    B. Zhang, M. Yang, X. Li, Y. Ye, X. Xu, K. Dai, Enhancing cross-target stance detection with transferable semantic-emotion knowledge, in: D. Jurafsky, J. Chai, N. Schluter, J. Tetreault (Eds.), Proceedings of the 58th Annual Meeting of the Association for Computational Linguis...

  31. [39]

    Conforti, J

    C. Conforti, J. Berndt, M. T. Pilehvar, C. Giannitsarou, F. Toxvaerd, N. Collier, Incorporating stock market signals for Twitter stance detection, in: S. Muresan, P. Nakov, A. Villavicencio (Eds.), Proceedings of the 60th Annual Meeting of the Association for Computational Lin...

  32. [40]

    Ramesh, S

    D. Ramesh, S. K. Sanampudi, An automated essay scoring systems: a systematic literature review, Artificial Intelligence Review 55 (3) (2021) 2495–2527.doi:10.1007/s10462-021-10068-2

  33. [41]

    Anshul, G

    A. Anshul, G. S. Pranav, M. Z. U. Rehman, N. Kumar, A multimodal framework for depression detection during covid-19 via harvesting social media, IEEE Transactions on Computational Social Systems 11 (2) (2024) 2872–2888.doi:10.1109/TCSS.2023.3309229

  34. [42]

    Chaudhari, P

    P. Chaudhari, P. Nandeshwar, S. Bansal, N. Kumar, Mahaemosen: Towards emotion-aware multimodal marathi sentiment analysis, ACM Transactions on Asian and Low-Resource Language Information Pro- cessing 22 (9) (2023) 1–24.doi:10.1145/3618057

  35. [43]

    Babanejad, H

    N. Babanejad, H. Davoudi, A. An, M. Papagelis, Affective and contextual embedding for sarcasm detection, in: D. Scott, N. Bel, C. Zong (Eds.), Proceedings of the 28th International Conference on Computational Linguistics, International Committee on Computational Linguistics, B...

  36. [44]

    Tromp, M

    E. Tromp, M. Pechenizkiy, Rule-based emotion detection on social media: Putting tweets on plutchik’s wheel, arXiv preprint arXiv:1412.4682(accessed 2025-04-04).doi:10.48550/arXiv.1412.4682

  37. [45]

    Liang, Z

    B. Liang, Z. Chen, L. Gui, Y. He, M. Yang, R. Xu, Zero-shot stance detection via contrastive learning, in: WWW 2022 - Proceedings of the ACM Web Conference 2022, 2022.doi:10.1145/3485447. 3511994

  38. [46]

    C. Zhao, Y. Li, C. Caragea, Y. Zhang, ZeroStance: Leveraging ChatGPT for open-domain stance detection via dataset generation, in: L.-W. Ku, A. Martins, V. Srikumar (Eds.), Findings of the Association for Com- putational Linguistics: ACL 2024, Association for Computational Ling...

  39. [47]

    G. Wang, C. Li, W. Wang, Y. Zhang, D. Shen, X. Zhang, R. Henao, L. Carin, Joint embedding of words and labels for text classification, in: I. Gurevych, Y. Miyao (Eds.), Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Pape...

  40. [48]

    P. Yang, F. Luo, S. Ma, J. Lin, X. Sun, A deep reinforced sequence-to-set model for multi-label classification, in: A. Korhonen, D. Traum, L. Màrquez (Eds.), Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, Association for Computational ...

  41. [49]

    Gorrell, E

    G. Gorrell, E. Kochkina, M. Liakata, A. Aker, A. Zubiaga, K. Bontcheva, L. Derczynski, SemEval-2019 task 7: RumourEval, determining rumour veracity and support for rumours, in: J. May, E. Shutova, A. Herbe- lot, X. Zhu, M. Apidianaki, S. M. Mohammad (Eds.), Proceedings of the ...

  42. [50]

    S. M. Mohammad, S. Kiritchenko, P. Sobhani, X. Zhu, C. Cherry, Semeval-2016 task 6: Detecting stance in tweets, in: SemEval 2016 - 10th International Workshop on Semantic Evaluation, Proceedings, 2016. doi:10.18653/v1/s16-1003

  43. [51]

    Y. Li, T. Sosea, A. Sawant, A. J. Nair, D. Inkpen, C. Caragea, P-stance: A large dataset for stance detection in political domain, in: C. Zong, F. Xia, W. Li, R. Navigli (Eds.), Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, Association for Computa...

  44. [52]

    Fajcik, L

    M. Fajcik, L. Burget, P. Smrz, But-fit at semeval-2019 task 7: Determining the rumour stance with pre- trained deep bidirectional transformers, in: NAACL HLT 2019 - International Workshop on Semantic Eval- uation, SemEval 2019, Proceedings of the 13th Workshop, 2019.doi:10.186...

  45. [53]

    A. Q. Jiang, A. Sablayrolles, A. Mensch, C. Bamford, D. S. Chaplot, D. d. l. Casas, F. Bressand, G. Lengyel, G. Lample, L. Saulnier, et al., Mistral 7b, arXiv preprint arXiv:2310.06825(accessed 2025-04-04). URLhttps://arxiv.org/abs/2310.06825

  46. [54]

    URLhttps://openai.com/blog/introducing-gpt-4-turbo

    OpenAI, Gpt-3.5 turbo release, (accessed 2025-04-04) (2023). URLhttps://openai.com/blog/introducing-gpt-4-turbo

  47. [55]

    Dubey, A

    A. Dubey, A. Jauhri, A. Pandey, A. Kadian, A. Al-Dahle, A. Letman, A. Mathur, A. Schelten, A. Yang, A. Fan, et al., The llama 3 herd of models, arXiv preprint arXiv:2407.21783(accessed 2025-04-04). URLhttps://arxiv.org/abs/2407.21783 40

  48. [56]

    Devlin, M.-W

    J. Devlin, M.-W. Chang, K. Lee, K. Toutanova, BERT: Pre-training of deep bidirectional transformers for language understanding, in: J. Burstein, C. Doran, T. Solorio (Eds.), Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational L...

  49. [57]

    H. W. Chung, L. Hou, S. Longpre, B. Zoph, Y. Tay, W. Fedus, Y. Li, X. Wang, M. Dehghani, S. Brahma, et al., Scaling instruction-finetuned language models, Journal of Machine Learning Research 25 (70) (2024) 1–53, (accessed 2025-04-04). URLhttps://www.jmlr.org/papers/volume25/2...

  50. [58]

    Raffel, N

    C. Raffel, N. Shazeer, A. Roberts, K. Lee, S. Narang, M. Matena, Y. Zhou, W. Li, P. J. Liu, Exploring the limits of transfer learning with a unified text-to-text transformer, Journal of Machine Learning Research 21 (2020) 1–67, (accessed 2025-04-04). URLhttps://jmlr.org/papers...

  51. [59]

    Z. Yao, W. Yang, F. Wei, Enhancing zero-shot stance detection with contrastive and prompt learning, En- tropy 26 (4).doi:10.3390/e26040325

  52. [60]

    Q. Chen, R. Zhang, Y. Zheng, Y. Mao, Dual contrastive learning: Text classification via label-aware data augmentation, arXiv preprint arXiv:2201.08702(accessed 2025-04-04). URLhttps://arxiv.org/abs/2201.08702

  53. [61]

    Liang, Q

    B. Liang, Q. Zhu, X. Li, M. Yang, L. Gui, Y. He, R. Xu, Jointcl: A joint contrastive learning framework for zero-shot stance detection, in: Proceedings of the Annual Meeting of the Association for Computational Linguistics, Vol. 1, 2022.doi:10.18653/v1/2022.acl-long.7

  54. [62]

    X. Wang, Y. Wang, S. Cheng, P. Li, Y. Liu, DEEM: Dynamic experienced expert modeling for stance de- tection, in: N. Calzolari, M.-Y. Kan, V. Hoste, A. Lenci, S. Sakti, N. Xue (Eds.), Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language ...

  55. [63]

    Burnham, Stance detection: a practical guide to classifying political beliefs in text, Political Science Research and Methods (2024) 1–18doi:10.1017/psrm.2024.35

    M. Burnham, Stance detection: a practical guide to classifying political beliefs in text, Political Science Research and Methods (2024) 1–18doi:10.1017/psrm.2024.35

  56. [64]

    M. Mets, A. Karjus, I. Ibrus, M. Schich, Automated stance detection in complex topics and small languages: The challenging case of immigration in polarizing news media, PLOS ONE 19 (4) (2024) e0302380.doi: 10.1371/journal.pone.0302380. 41

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.