Pith. sign in

REVIEW 4 major objections 5 minor 40 references

ViFactCheck: A New Benchmark Dataset and Methods for Multi-domain News Fact-Checking in Vietnamese

T0 review · 4 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 ViFactCheck is the first publicly available benchmark for multi-domain Vietnamese news fact-checking, with 7,232 human-annotated claim-evidence pairs and a fine-tuned Gemma baseline reaching 89.90% macro F1 on gold evidence.

desk verdict ViFactCheck fills a real gap for Vietnamese news fact-checking, but the paper's own statistics are internally inconsistent and need reconciliation before the benchmark size and model rankings can be trusted. read the letter →

arxiv 2412.15308 v1 pith:CQAASVLE submitted 2024-12-19 cs.CL cs.IR

Tran Thai Hoa , Tran Quang Duy , Khanh Quoc Tran , Kiet Van Nguyen This is my paper · ORCID
classification cs.CLcs.IR
keywords fact-checkingVietnameseNLPbenchmarkdatasetclaimverificationevidence-basedmulti-domainnewslargelanguagemodelslow-resourcelanguages
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 introduces ViFactCheck, a new publicly available benchmark for fact-checking Vietnamese news claims across 12 topics and 9 online news outlets, built from 7,232 human-annotated claim-evidence pairs with high annotator agreement (Fleiss Kappa 0.83). It argues that this dataset fills a gap in low-resource fact-checking resources and demonstrates that fine-tuned language models, particularly Gemma, can reach strong performance, with a macro F1 of 89.90% when given gold evidence. The work matters because it provides a reusable testbed for developing and evaluating Vietnamese fact-checking systems, and it shows that the main difficulty lies in evidence retrieval rather than verification itself.

What carries the argument

The central artifact is ViFactCheck itself: a dataset whose construction pipeline starts from 1,000 news articles from nine licensed Vietnamese outlets, from which annotators generate six claims per article (two per label) and mark the supporting evidence spans. The task is three-way classification of a claim as Support, Refute, or Not Enough Information (NEI) against a given context or evidence. The evaluation machinery is the comparison of fine-tuned and zero-shot prompted models under two input settings: full article context versus gold evidence spans; the paper's headline numbers come from macro F1 on a held-out test set. A further mechanism is the word-overlap and semantic-similarity analysis (LCS, NWR, Jaccard, SBERT cosine) used to argue that claim-context pairs are hard (low overlap) while claim-evidence pairs are easy (high overlap, 63.52% Jaccard for Support), explaining why gold evidence boosts scores.

What would settle it

Replace each gold evidence paragraph in the test set with a similar-wording paragraph that does not support the claim, keeping the gold labels unchanged; if fine-tuned Gemma's macro F1 stays near 89.90%, the score reflects surface paraphrase overlap rather than genuine verification.

Watch

Extended reading notes

Core claim

ViFactCheck is presented as the first publicly available benchmark dataset built specifically for multi-domain Vietnamese news fact-checking. The dataset was created by having seven native Vietnamese-speaking annotators generate claims from news articles and label them Support, Refute, or Not Enough Information, with explicit evidence annotations and a requirement that some claims combine multiple pieces of evidence; agreement on a 10% relabeled subset reached a Fleiss Kappa of 0.83. Evaluating fine-tuned and prompted models on this resource, the paper reports that fine-tuned Gemma achieves the highest macro F1 of 89.90% in the gold-evidence setting and 85.94% when given full article context, outperforming all tested multilingual and Vietnamese-specific baselines. The authors conclude that the dataset is reliable, that fine-tuning markedly outperforms prompting, and that gold evidence consistently beats full context, locating the hardest part of fact-checking in evidence retrieval.

Load-bearing premise

The whole evaluation rests on the assumption that the 7,232 ground-truth labels are correct, but only 726 of them (10%) were ever checked by independent annotators, and the agreement statistic of 0.83 was measured on that small subset alone.

Editorial extensions

If this is right

  • Vietnamese fact-checking now has a public, multi-domain testbed with 7,232 labeled claims, enabling consistent comparison of future systems.
  • Fine-tuned models with gold evidence far outperform both prompting and full-context inputs, meaning real gains in practice depend on building better evidence-retrieval pipelines.
  • Gemma-7B fine-tuned sets the current state of the art on this benchmark at 89.90% macro F1, a result the paper frames as establishing a new standard for the dataset.
  • Multi-evidence claims are measurably harder: all models drop in F1 when verification requires combining two or more evidence pieces.
  • Error analysis shows the residual failures concentrate in semantic ambiguity and complex inferential chains, giving a concrete target for future work.

Reading between the lines

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

  • The 63.52% Jaccard similarity between Support claims and their gold evidence suggests a large share of the 89.90% score may come from recognizing paraphrastic restatements rather than from multi-hop reasoning; spot-checking a random sample of correct predictions would reveal how much reasoning is actually happening.
  • The paper's reported train/dev/test split is inconsistent (6:2:2 in Section 4.2 versus 7:1:2 in Appendix G), and the table in Appendix G shows context counts that match neither ratio; before the dataset is used as a leaderboard, the canonical split needs to be fixed and documented.
  • The same annotation recipe—generate paraphrased claims, mark evidence spans, enforce multi-evidence examples—could be ported to other low-resource languages, giving a template for building fact-checking benchmarks where automatic translation of English datasets is not enough.
  • A natural extension would replace gold evidence with retrieved evidence at test time and measure end-to-end pipeline F1; based on the paper's own gap between full-context and gold-evidence scores, that end-to-end number would be considerably lower than 89.90%.
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

4 major / 5 minor

Summary. ViFactCheck is presented as the first publicly available multi-domain Vietnamese fact-checking benchmark. The authors describe collecting online news articles from nine Vietnamese outlets, generating claims labeled Support/Refute/NEI with annotated evidence, and validating labels via self-checking, cross-checking, and a Fleiss Kappa computation on 726 claims. They compare four pre-trained language models and several LLMs under fine-tuning and zero-shot prompting, using either full context or gold evidence, and report that fine-tuned Gemma achieves the best macro F1 of 89.90% with gold evidence. The paper also analyzes evidence retrieval, error types, topic and context-length effects, and training data size.

Significance. If the dataset statistics and split details are corrected and verified, ViFactCheck is a useful resource for Vietnamese NLP and low-resource fact-checking. It is one of the few Vietnamese datasets that covers multiple news domains and includes annotated evidence spans, and the public GitHub release of the dataset, checkpoints, and pipelines is a practical strength. The model comparison across PLMs, fine-tuned LLMs, and prompting methods is informative, and the evaluation is not circular: the reported F1 scores are measured on a held-out test set rather than quantities derived from the paper's own assumptions. However, the internal inconsistencies in the dataset size and partition ratios must be resolved before the benchmark's headline claims can be trusted.

major comments (4)
  1. [§3.1, Appendix C, Table 9] The paper's own arithmetic on dataset size is internally inconsistent. Section 3.1 states that the initial corpus contained 1,000 articles, and Appendix C instructs annotators to generate six claims per article (two Support, two Refute, two NEI); that caps the dataset at 6,000 claims. Yet the abstract, Section 1, and Table 9 report 7,232 claims, and Table 9 lists 2,289 contexts (1,035+496+758). A 1,000-article corpus with six claims per article cannot produce 2,289 contexts or 7,232 claims. Please report the exact number of collected articles, the number of contexts after any filtering, the number of claims per article, and reconcile these totals.
  2. [§4.2 vs. Appendix G] The train/dev/test split is described inconsistently. Section 4.2 says the data were partitioned with a ratio of 6:2:2, while Appendix G states 7:1:2. The reported claim counts (5,062/723/1,447) match the 7:1:2 ratio (70/10/20%), but the reported context counts (1,035/496/758) match neither ratio. In addition, the paper never states whether the split is performed at the article/context level or at the claim level. This distinction matters for leakage: if claims derived from the same article appear in both training and test splits, the reported test performance may be optimistically biased. Please specify the split unit and give exact counts per subset.
  3. [§3.3] The Fleiss Kappa of 0.83 is computed on only 726 claims, which is 10% of the 7,232 total, and these claims were relabeled by three annotators who did not author them. No inter-annotator agreement or independent validation is reported for the remaining 90% of the dataset. The abstract's characterization of the whole dataset as 'rigorously vetted' with a 0.83 reliability score is therefore stronger than the reported evidence. Please either report IAA on a larger sample or explicitly frame the 0.83 as an estimate from a 10% subsample, and preferably include per-label agreement.
  4. [Table 2, Table 3, Section 4.3] The headline macro F1 of 89.90% is obtained in the gold-evidence setting. Table 2 shows that Support claims share 63.52% Jaccard similarity and 86.89% related-word similarity with their associated evidence, and the Refute and NEI rows also show high overlap. This raises the concern that a substantial share of the gold-evidence performance may reflect lexical or paraphrase matching rather than verification reasoning. To support the claim that Gemma is effective at fact-checking, please report performance broken down by low-overlap versus high-overlap claim-evidence pairs, and temper the 'new standard' wording, which is also problematic because no existing Vietnamese fact-checking benchmark is compared directly.
minor comments (5)
  1. [Figure 2 caption] The caption contains a typo: 'contruction process' should be 'construction process'.
  2. [Section 4.1] The text says the framework 'includes four SOTA open-source LLMs: Llama, Gemma, and Mistral', but only three names are listed; if the intended fourth is Gemini, the sentence should be corrected, and the label 'open-source' does not apply to Gemini.
  3. [Appendix H.2] The description of Gemma states it was trained on 'up to 6 billion tokens', which appears inconsistent with the publicly available Gemma technical report; please verify the number and correct it if needed.
  4. [Table 5] The column labeled 'F1 score' is ambiguous; please specify whether it is the macro F1 score and how it relates to the per-class Support/Refute/NEI columns.
  5. [Table 3 footnote] The statement that all performance differences are statistically significant at p < 0.01 is not accompanied by a description of the statistical test, the number of samples, or the multiple-comparison procedure; please provide these details.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the benchmark's performance numbers and annotation-quality estimates come from held-out or independently relabeled data, not from quantities fitted to make the conclusions true.

full rationale

The paper does not derive any target conclusion from its own assumptions. The central empirical claims—the 89.90% macro F1 for the fine-tuned Gemma model and the relative ranking of baselines—are measured on a test set (Section 4.3, Table 3), which is standard supervised evaluation and not a fitted input renamed as a prediction. The annotation-quality claim (Fleiss Kappa 0.83) is estimated from an independent relabeling exercise on 10% of the claims by annotators who did not author them (Section 3.3); this is an external-check procedure rather than a self-validating definition. Citations to prior Vietnamese fact-checking work (e.g., ViWikiFC, Duong et al.) are contextual and are not load-bearing for the paper's dataset or results. The paper does contain internal arithmetic inconsistencies—the 1,000-article corpus described in Section 3.1 with six claims per article in Appendix C implies no more than 6,000 claims, while the abstract and Table 9 report 7,232; Section 4.2 states a 6:2:2 split while Appendix G reports 7:1:2; and the context counts in Table 9 do not match either ratio. These are correctness and reproducibility concerns, not circular reasoning: no claim in the paper is true simply by virtue of being defined in terms of the evidence used to support it. Because no load-bearing step reduces to its own input by construction or by self-citation, the appropriate circularity score is 0.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central claim depends on the annotation pipeline (labels, evidence spans, gold standard), the split integrity, and the task's validity as a measure of fact-checking, each captured as an axiom above. The listed free parameters are disclosed training and retrieval settings that condition the reported F1 scores; they are choices made by the authors, not hidden degrees of freedom in a derivation. No invented entities are introduced. The deepest risk is not circularity but measurement validity: the benchmark is constructed and evaluated within a single team's pipeline, and the headline result is achieved in a high-lexical-overlap setting.

free parameters (3)
  • PLM fine-tuning hyperparameters = learning rate 5e-06, dropout 0.3, batch size 16, 10 epochs
    Section 4.2; disclosed training settings that condition the reported F1 scores. They are choices, not derived; a different grid would change the headline numbers.
  • LLM LoRA fine-tuning hyperparameters = LoRA rank 16, alpha 16, learning rate 2e-04, batch size 16, 5 epochs
    Section 4.2 and Appendix H.2; disclosed settings for the LLM fine-tuning that produced the 89.90% result.
  • Evidence retrieval top-K threshold = K = 5
    Section 4.4: 'an optimal K threshold of 5' is read off the plateau in Figure 4; a data-derived choice in the retrieval analysis, not a pre-registered constant.
assumptions (5)
  • domain assumption Fleiss Kappa on the relabeled 10% subset (n=726) is representative of label quality for the full 7,232-claim dataset
    Section 3.3: only 10% of claims received independent relabeling; the remaining 90% were self-checked and cross-checked without a second formal label, so the paper assumes the measured agreement generalizes.
  • domain assumption Train/dev/test partitions are leakage-free and the split unit is well-defined
    Section 4.2 states only a ratio (6:2:2) while Appendix G reports 7:1:2 with counts that match neither at context level; whether articles or contexts are held out as units is never stated. If claims from the same article appear in both train and test, scores would be inflated.
  • domain assumption Models cannot exploit annotation artifacts (for example, NEI claims being stylistically marked)
    Appendix F: 90.20% of NEI claims were generated by 'inferring sentences with unspecified information', a construction rule that could make the NEI class recognizable by style rather than by content; the benchmark's validity assumes this is not exploitable.
  • domain assumption The nine news outlets and the February-March 2023 window are representative of Vietnamese online news for fact-checking
    Section 3.1: the corpus is a two-month snapshot from licensed outlets; the dataset does not cover social media, user-generated content, or other periods, so benchmark conclusions are scoped to this distribution.
  • standard math Fleiss Kappa is a valid inter-annotator agreement measure for a three-label claim-verification task
    Section 3.3; standard practice in the fact-checking dataset literature (McHugh 2012; Thorne et al. 2018), not controversial.

how reviews work

0 comments
Cite this review

Pith. "Pith review of ViFactCheck: A New Benchmark Dataset and Methods for Multi-domain News Fact-Checking in Vietnamese." pith.science (2026). https://pith.science/paper/CQAASVLE

@misc{pith2026241215308,
  author       = {Pith},
  title        = {Pith review of: ViFactCheck: A New Benchmark Dataset and Methods for Multi-domain News Fact-Checking in Vietnamese},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CQAASVLE}},
  note         = {Machine review of arXiv:2412.15308}
}
read the original abstract

The rapid spread of information in the digital age highlights the critical need for effective fact-checking tools, particularly for languages with limited resources, such as Vietnamese. In response to this challenge, we introduce ViFactCheck, the first publicly available benchmark dataset designed specifically for Vietnamese fact-checking across multiple online news domains. This dataset contains 7,232 human-annotated pairs of claim-evidence combinations sourced from reputable Vietnamese online news, covering 12 diverse topics. It has been subjected to a meticulous annotation process to ensure high quality and reliability, achieving a Fleiss Kappa inter-annotator agreement score of 0.83. Our evaluation leverages state-of-the-art pre-trained and large language models, employing fine-tuning and prompting techniques to assess performance. Notably, the Gemma model demonstrated superior effectiveness, with an impressive macro F1 score of 89.90%, thereby establishing a new standard for fact-checking benchmarks. This result highlights the robust capabilities of Gemma in accurately identifying and verifying facts in Vietnamese. To further promote advances in fact-checking technology and improve the reliability of digital media, we have made the ViFactCheck dataset, model checkpoints, fact-checking pipelines, and source code freely available on GitHub. This initiative aims to inspire further research and enhance the accuracy of information in low-resource languages.

Figures

Figures reproduced from arXiv: 2412.15308 by the authors.

Figure 1
Figure 1. An example of the Vietnamese fact-checking task. [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. The ViFactCheck dataset contruction process. [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Comparison of the labeling pipelines in the FEVER [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (11 more)
Figure 4
Figure 4. Figure 4: Comparative performance of various text retrieval models across different Top-K settings. [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Distributions of errors. The analysis revealed significant challenges in handling Semantic Ambiguity and Complex Inferential Chains, both of which are pivotal for refining NLP technologies. Seman￾tic Ambiguity issues particularly highlight the necessity for context-awa…
Figure 6
Figure 6. Figure 6: Label Studio UI for our annotation task. [PITH_FULL_IMAGE:figures/full_fig_p010_6.png]
Figure 7
Figure 7. Figure 7: The ratio of combining different rules to create claims in ViFactCheck. [PITH_FULL_IMAGE:figures/full_fig_p013_7.png]
Figure 8
Figure 8. Figure 8: Topic distribution on ViFactCheck dataset. [PITH_FULL_IMAGE:figures/full_fig_p013_8.png]
Figure 10
Figure 10. Figure 10: Examples of Evidence Retrieval Failure [PITH_FULL_IMAGE:figures/full_fig_p015_10.png]
Figure 11
Figure 11. Figure 11: Examples of Semantic Ambiguity. • Inference Hallucination: Incorrect classifications pro￾duced by the model, despite the correct extraction and availability of relevant evidence (as shown in [PITH_FULL_IMAGE:figures/full_fig_p016_11.png]
Figure 12
Figure 12. Figure 12: Examples of Inference Hallucination. • Complex Inferential Chain: Errors resulting from the necessity to synthesize insights across multiple sources or evidences through sequential reasoning (as shown in [PITH_FULL_IMAGE:figures/full_fig_p016_12.png]
Figure 15
Figure 15. Figure 15: The effect of the topic on the test set. [PITH_FULL_IMAGE:figures/full_fig_p017_15.png]
Figure 16
Figure 16. Figure 16: The impact of training data size on test set. [PITH_FULL_IMAGE:figures/full_fig_p017_16.png]
Figure 17
Figure 17. Figure 17: The comparison of training times per epoch for various baseline models. [PITH_FULL_IMAGE:figures/full_fig_p018_17.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

40 extracted references · 26 canonical work pages

  1. [1]

    Aly, R.; Guo, Z.; Schlichtkrull, M.; Thorne, J.; Vlachos, A.; Christodoulopoulos, C.; Cocarascu, O.; and Mittal, A. 2021. FEVEROUS: Fact Extraction and VERification Over Unstructured and Structured information. In Vanschoren, J.; and Yeung, S., eds., Proceedings of the Neural Information Processing Systems Track on Datasets and Benchmarks, volume 1. Curran

    work page 2021

  2. [2]

    Augenstein, I.; Lioma, C.; Wang, D.; Chaves Lima, L.; Hansen, C.; Hansen, C.; and Simonsen, J. G. 2019. M ulti FC : A Real-World Multi-Domain Dataset for Evidence-Based Fact Checking of Claims. In Inui, K.; Jiang, J.; Ng, V.; and Wan, X., eds., Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International...

    work page 2019

  3. [3]

    V.; Tran, T

    Bui, T. V.; Tran, T. O.; and Le-Hong, P. 2020. Improving Sequence Tagging for V ietnamese Text using Transformer-based Neural Models. In Nguyen, M. L.; Luong, M. C.; and Song, S., eds., Proceedings of the 34th Pacific Asia Conference on Language, Information and Computation, 13--20. Hanoi, Vietnam: Association for Computational Linguistics

    work page 2020

  4. [4]

    Conneau, A.; Khandelwal, K.; Goyal, N.; Chaudhary, V.; Wenzek, G.; Guzm \'a n, F.; Grave, E.; Ott, M.; Zettlemoyer, L.; and Stoyanov, V. 2020. Unsupervised Cross-lingual Representation Learning at Scale. In Jurafsky, D.; Chai, J.; Schluter, N.; and Tetreault, J., eds., Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics...

    2020

  5. [5]

    Devlin, J.; Chang, M.-W.; Lee, K.; and Toutanova, K. 2019. BERT : Pre-training of Deep Bidirectional Transformers for Language Understanding. In Proceedings of the 2019 Conference of the North A merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) , 4171--4186. Minneapolis, Minneso...

    2019

  6. [6]

    T.; Ho, V

    Duong, H. T.; Ho, V. H.; and Do, P. 2023. Fact-checking Vietnamese Information Using Knowledge Graph, Datalog, and KG-BERT. ACM Transactions on Asian and Low-Resource Language Information Processing, 22(10): 1--23

    work page 2023

  7. [7]

    Fleiss, J. L. 1971. Measuring nominal scale agreement among many raters. Psychological bulletin, 76(5): 378

    1971

  8. [8]

    Gupta, A.; and Srikumar, V. 2021. X -Fact: A New Benchmark Dataset for Multilingual Fact Checking. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 2: Short Papers), 675--682. Online: Association for Computational Linguistics

    work page 2021

Show all 40 references
  1. [9]

    Hu, X.; Guo, Z.; Wu, G.; Liu, A.; Wen, L.; and Yu, P. 2022. CHEF : A Pilot C hinese Dataset for Evidence-Based Fact-Checking. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, 3362...

    2022

  2. [10]

    P.; and Ji, H

    Huang, K.-H.; Chan, H. P.; and Ji, H. 2023. Zero-shot Faithful Factual Error Correction. In Rogers, A.; Boyd-Graber, J.; and Okazaki, N., eds., Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 5660--5676. Toronto,...

    2023

  3. [11]

    Q.; Sablayrolles, A.; Mensch, A.; Bamford, C.; Chaplot, D

    Jiang, A. Q.; Sablayrolles, A.; Mensch, A.; Bamford, C.; Chaplot, D. S.; Casas, D. d. l.; Bressand, F.; Lengyel, G.; Lample, G.; Saulnier, L.; et al. 2023. Mistral 7B. arXiv preprint arXiv:2310.06825

    2023 arXiv

  4. [12]

    Khouja, J. 2020. Stance Prediction and Claim Verification: An A rabic Perspective. In Christodoulopoulos, C.; Thorne, J.; Vlachos, A.; Cocarascu, O.; and Mittal, A., eds., Proceedings of the Third Workshop on Fact Extraction and VERification (FEVER), 8--17. Online: Association...

    2020

  5. [13]

    Kim, J.; Park, S.; Kwon, Y.; Jo, Y.; Thorne, J.; and Choi, E. 2023. F act KG : Fact Verification via Reasoning on Knowledge Graphs. In Rogers, A.; Boyd-Graber, J.; and Okazaki, N., eds., Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (V...

    2023

  6. [14]

    M.; Baum, M

    Lazer, D. M.; Baum, M. A.; Benkler, Y.; Berinsky, A. J.; Greenhill, K. M.; Menczer, F.; Metzger, M. J.; Nyhan, B.; Pennycook, G.; Rothschild, D.; et al. 2018. The science of fake news. Science, 359(6380): 1094--1096

    2018

  7. [15]

    T.; To, L

    Le, H. T.; To, L. T.; Nguyen, M. T.; and Van Nguyen, K. 2024. ViWikiFC: Fact-Checking for Vietnamese Wikipedia-Based Textual Knowledge Source. arXiv preprint arXiv:2405.07615

    2024

  8. [16]

    Liu, Y.; Ott, M.; Goyal, N.; Du, J.; Joshi, M.; Chen, D.; Levy, O.; Lewis, M.; Zettlemoyer, L.; and Stoyanov, V. 2019. RoBERTa: A Robustly Optimized BERT Pretraining Approach. arXiv:1907.11692

    2019 arXiv

  9. [17]

    Liu, Z.; Xiong, C.; Sun, M.; and Liu, Z. 2020. Fine-grained Fact Verification with Kernel Graph Attention Network. In Jurafsky, D.; Chai, J.; Schluter, N.; and Tetreault, J., eds., Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 7342--7...

    2020

  10. [18]

    Loshchilov, I.; and Hutter, F. 2019. Decoupled Weight Decay Regularization. In 7th International Conference on Learning Representations, ICLR 2019, New Orleans, LA, USA, May 6-9, 2019 . OpenReview.net

    2019

  11. [19]

    McCoy, T.; Pavlick, E.; and Linzen, T. 2019. Right for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inference. In Korhonen, A.; Traum, D.; and M \`a rquez, L., eds., Proceedings of the 57th Annual Meeting of the Association for Computational Linguisti...

    2019

  12. [20]

    McHugh, M. 2012. Interrater reliability: the kappa statistic. Biochem Med (Zagreb), 22(3): 276--282

    2012

  13. [21]

    Q.; and Tuan Nguyen, A

    Nguyen, D. Q.; and Tuan Nguyen, A. 2020. P ho BERT : Pre-trained language models for V ietnamese. In Findings of the Association for Computational Linguistics: EMNLP 2020, 1037--1042. Association for Computational Linguistics

    2020

  14. [22]

    Nie, Y.; Chen, H.; and Bansal, M. 2019. Combining fact extraction and verification with neural semantic matching networks. In Proceedings of the AAAI conference on artificial intelligence, volume 33, 6859--6866

    2019

  15. [23]

    N rregaard, J.; and Derczynski, L. 2021. D an FEVER : claim verification dataset for D anish. In Dobnik, S.; and vrelid, L., eds., Proceedings of the 23rd Nordic Conference on Computational Linguistics (NoDaLiDa), 422--428. Reykjavik, Iceland (Online): Link \"o ping University...

    2021

  16. [24]

    T.; Wang, W

    Pan, L.; Wu, X.; Lu, X.; Luu, A. T.; Wang, W. Y.; Kan, M.-Y.; and Nakov, P. 2023. Fact-Checking Complex Claims with Program-Guided Reasoning. In Rogers, A.; Boyd-Graber, J.; and Okazaki, N., eds., Proceedings of the 61st Annual Meeting of the Association for Computational Ling...

    2023

  17. [25]

    Reid, M.; Savinov, N.; Teplyashin, D.; Lepikhin, D.; Lillicrap, T.; Alayrac, J.-b.; Soricut, R.; Lazaridou, A.; Firat, O.; Schrittwieser, J.; et al. 2024. Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context. arXiv preprint arXiv:2403.05530

    2024 arXiv

  18. [26]

    Reimers, N.; and Gurevych, I. 2019. Sentence- BERT : Sentence Embeddings using S iamese BERT -Networks. In Inui, K.; Jiang, J.; Ng, V.; and Wan, X., eds., Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Con...

    2019

  19. [27]

    Robertson, S.; Zaragoza, H.; et al. 2009. The probabilistic relevance framework: BM25 and beyond. Foundations and Trends in Information Retrieval , 3(4): 333--389

    2009

  20. [28]

    Schuster, T.; Fisch, A.; and Barzilay, R. 2021. Get Your Vitamin C ! Robust Fact Verification with Contrastive Evidence. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, 624--643....

    2021

  21. [29]

    Soleimani, A.; Monz, C.; and Worring, M. 2020. BERT for Evidence Retrieval and Claim Verification. In Jose, J. M.; Yilmaz, E.; Magalh \ a es, J.; Castells, P.; Ferro, N.; Silva, M. J.; and Martins, F., eds., Advances in Information Retrieval, 359--366. Cham: Springer Internati...

    2020

  22. [30]

    S.; Love, J.; et al

    Team, G.; Mesnard, T.; Hardin, C.; Dadashi, R.; Bhupatiraju, S.; Pathak, S.; Sifre, L.; Rivi \`e re, M.; Kale, M. S.; Love, J.; et al. 2024. Gemma: Open models based on gemini research and technology. arXiv preprint arXiv:2403.08295

    2024 arXiv

  23. [31]

    Thorne, J.; Vlachos, A.; Christodoulopoulos, C.; and Mittal, A. 2018. FEVER : a Large-scale Dataset for Fact Extraction and VER ification. In Walker, M.; Ji, H.; and Stent, A., eds., Proceedings of the 2018 Conference of the North A merican Chapter of the Association for Compu...

    2018

  24. [32]

    Touvron, H.; Lavril, T.; Izacard, G.; Martinet, X.; Lachaux, M.-A.; Lacroix, T.; Rozi \`e re, B.; Goyal, N.; Hambro, E.; Azhar, F.; et al. 2023. Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971

    2023 arXiv

  25. [33]

    Vlachos, A.; and Riedel, S. 2014. Fact Checking: Task definition and dataset construction. In Danescu-Niculescu-Mizil, C.; Eisenstein, J.; McKeown, K.; and Smith, N. A., eds., Proceedings of the ACL 2014 Workshop on Language Technologies and Computational Social Science , 18--...

    2014

  26. [34]

    Q.; Nguyen, D

    Vu, T.; Nguyen, D. Q.; Nguyen, D. Q.; Dras, M.; and Johnson, M. 2018. V n C ore NLP : A V ietnamese Natural Language Processing Toolkit. In Liu, Y.; Paek, T.; and Patwardhan, M., eds., Proceedings of the 2018 Conference of the North A merican Chapter of the Association for Com...

    2018

  27. [35]

    Wang, W. Y. 2017. `` Liar, Liar Pants on Fire '' : A New Benchmark Dataset for Fake News Detection. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), 422--426. Vancouver, Canada: Association for Computational L...

    2017

  28. [36]

    Wang, Y.; Li, Y.; Wang, Y.; Mi, F.; Zhou, P.; Wang, X.; Liu, J.; Jiang, X.; and Liu, Q. 2022. Pan More Gold from the Sand: Refining Open-domain Dialogue Training with Noisy Self-Retrieval Generation. In Calzolari, N.; Huang, C.-R.; Kim, H.; Pustejovsky, J.; Wanner, L.; Choi, K...

    2022

  29. [37]

    Wu, S.; Xin, C.; Lin, H.; Han, X.; Liu, C.; Chen, J.; Yang, F.; Wan, G.; and Sun, L. 2023. Ambiguous Learning from Retrieval: Towards Zero-shot Semantic Parsing. In Rogers, A.; Boyd-Graber, J.; and Okazaki, N., eds., Proceedings of the 61st Annual Meeting of the Association fo...

    2023

  30. [38]

    Zhong, W.; Xu, J.; Tang, D.; Xu, Z.; Duan, N.; Zhou, M.; Wang, J.; and Yin, J. 2020. Reasoning Over Semantic-Level Graph for Fact Checking. In Jurafsky, D.; Chai, J.; Schluter, N.; and Tetreault, J., eds., Proceedings of the 58th Annual Meeting of the Association for Computati...

    2020

  31. [39]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all...

  32. [40]

    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 gl...

Pith tools

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