Is External Information Useful for Stance Detection with LLMs?

Quang Minh Nguyen; Taegyoon Kim

REVIEW 3 major objections 6 minor 27 references

Is External Information Useful for Stance Detection with LLMs?

T0 review · 3 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read External information from Wikipedia or the web usually hurts LLM stance detection, cutting macro F1 by up to 27.9 percent.

desk verdict A wide, honest empirical study whose headline 'external info hurts' is entangled with a prompt-structure difference; fixable and worth sending to review. read the letter →

arxiv 2507.01543 v1 pith:V5DBESZD submitted 2025-07-02 cs.CL

Quang Minh Nguyen , Taegyoon Kim This is my paper · ORCID

classification cs.CL

keywords stancedetectionlargelanguagemodelsexternalinformationWikipediaexcerptswebsearchbiaschain-of-thoughtpromptingmacroF1

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper asks whether giving a large language model background information about a target—Wikipedia excerpts or web-search summaries—makes its stance detection better. Across eight LLMs and three tweet datasets, the answer is usually no: adding such information lowers macro F1 in most model–dataset combinations, with the largest drop at 27.9%. The paper traces the harm to a mechanism: LLMs align their prediction with the stance or sentiment they perceive in the supplied information rather than with the ground-truth stance of the text. Chain-of-thought prompting does not remove the effect, and fine-tuning reduces but does not fully eliminate it. A fine-tuned BERT baseline improves with the same information, so the finding is specific to how LLMs use context.

What carries the argument

The analytic instruments are the net adoption rate (NAR) and the harmful adoption rate (HAR), defined in Equation 1. NAR measures whether a model is more likely to produce a prediction matching the stance or sentiment it perceives in the external information when that information is present than when it is absent, and HAR measures whether such adoptions turn correct predictions into incorrect ones more often than the reverse. These rates convert the intuition that the model is swayed by the supplied text into a measurable quantity, giving a causal account of the performance drops observed in the main comparison.

What would settle it

A random-text control condition would settle the claim: present the same prompt with length-matched but stance-free filler text instead of Wikipedia or web-search excerpts, keeping the instruction sentence present in every condition. If macro F1 falls as much with filler as with real external information, then the degradation is a prompt-structure artifact, not evidence that external information biases LLM predictions.

Watch

Extended reading notes

Core claim

The central discovery is that external information from Wikipedia or web search degrades stance detection performance of LLMs in most cases rather than helping it. The paper shows this through a systematic comparison of eight LLMs on three datasets with 12 targets, where macro F1 drops by up to 27.9% relative to the no-information baseline. The degradation is explained by the models' tendency to adopt the stance and sentiment they perceive in the supplied information: net adoption rates are positive for nearly all model–dataset combinations, and the adoptions are more often harmful than helpful. The paper further shows that chain-of-thought prompting does not eliminate the degradation and that fine-tuning only narrows it, in contrast to a BERT-based baseline that benefits from the same external information.

Load-bearing premise

The load-bearing assumption is that the measured degradation comes from the content of the external information and not from the extra instruction sentence that is added only when external information is present; if the added sentence alone causes the drop, the central claim is not established.

Editorial extensions

If this is right

In zero-shot use, Wikipedia and web-search excerpts should not be assumed to help LLM stance classifiers; in most model–dataset pairs they lower macro F1, so systems should be evaluated with and without them.
The bias mechanism predicts that source reliability matters: a neutral or counter-stanced excerpt will push predictions in its direction even when the text itself says otherwise.
Chain-of-thought prompting does not cure the bias, so instruction-based warnings not to adopt the information's stance are not an adequate safeguard.
Fine-tuning with external information shrinks but does not eliminate the degradation, so mitigation requires more than continued training on the same prompt distribution.
The contrast with the BERT-based baseline indicates that the degradation is not a property of the task or datasets but of how LLMs attend to supplied context.

Reading between the lines

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

The same stance-adoption mechanism likely applies beyond three-way stance classification: any retrieved passage that carries a stance may shift an LLM's answer toward that stance.
A concrete extension would be to score each retrieved passage's stance and re-weight or filter passages before prompting; the paper does not test such a mitigation.
Taken further, a stance-tilted web source is not just a performance nuisance but a potential poisoning channel for deployed stance classifiers that consume external text.

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.

Desk Editor's Note

read the letter

Know this much: it's a genuinely useful empirical study, but the headline result is not as clean as it looks. The comparison that drives the paper—external info present vs. absent—also changes the prompt: the no-info condition omits the sentence 'The following information can be helpful: {wiki}' (or the 'do NOT automatically adopt' variant in the CoT prompts). So the observed F1 drops could be caused by the extra instruction, not by the semantic content of the Wikipedia or web text. That's a real confound, and the paper does not address it.

What is good: the scale. Eight LLMs, three datasets, twelve targets, four information types, plus CoT and LoRA experiments. The contrast with a fine-tuned BERT baseline is a nice control showing the effect is specific to these models. The NAR/HAR analysis is a sensible attempt to show that models move their predictions toward the stance/sentiment of the supplied text. The paper is honest about limitations and the code is released. The 'degrades in most cases' framing is accurate: the drops are consistent in direction if not always large.

The soft spots beyond the confound: single runs with no variance. Setting temperature to zero reduces stochasticity, but tokenizer/ordering effects and API nondeterminism can still produce spread; a few seeds would make the table sturdier. Also, the length-correlation analysis in Figure 2 doesn't rescue the main claim because length is not presence—every point in that figure already has the instruction sentence. A random-text control (e.g., an irrelevant Wikipedia passage or shuffled tokens) would isolate content from prompt form. Without it, the paper can still say 'external information as presented here' hurts, but it cannot say 'the information content' is the culprit.

I would not reject this; I would send it back for a reasonable revision. The empirical sweep is too wide to ignore, and the negative result, if it holds up under a proper control, matters for anyone building RAG-based classifiers. The fix is cheap. A serious referee should look at it, and the authors should be given the chance to lock the claim down.

Referee Report

3 major / 6 minor

Summary. The paper asks whether adding external information (Wikipedia excerpts or web search results) to LLM prompts improves stance detection. Across eight LLMs, three datasets, and twelve targets, the authors report that external information degrades performance in most configurations, with macro-F1 drops up to 27.9%, in contrast to a fine-tuned BERT baseline (WS-BERT) that generally improves. They propose net adoption rate (NAR) and harmful adoption rate (HAR) metrics to show that models align predictions with the stance or sentiment of the provided information, and they test chain-of-thought prompting and LoRA fine-tuning as mitigations. The paper concludes that external information often harms LLM stance detection and should be used with caution.

Significance. If the central claim is secure, this is a useful cautionary result for the growing practice of augmenting LLM prompts with retrieved text: it would show that knowledge injection, which helps fine-tuned BERT-style models, can backfire for prompted LLMs. The study's strengths are its breadth (eight models, three datasets, multiple information sources, CoT variations, and fine-tuning), the release of code, and the introduction of NAR/HAR as interpretable diagnostics. However, the headline conclusion is not yet fully supported because the no-information baseline differs from the information condition in prompt structure, and because single-run results lack uncertainty quantification. With additional control experiments and statistical rigor, this could be a strong contribution to LLM robustness and stance detection research.

major comments (3)

[Appendix B / Section 4.1] The central comparison is confounded. In the non-CoT prompt, the information condition adds the sentence "The following information can be helpful: {wiki}." and the no-information condition omits this sentence entirely. In the zero-shot and few-shot CoT prompts, the information condition adds "Integrate the following external information and do NOT automatically adopt the stance of it: {wiki}.", and again this whole sentence is absent in the no-information baseline. Every performance difference in Table 1 therefore conflates the semantic content of the external information with a prompt-structure change: an extra instruction sentence, an explicit directive about stance adoption, and greater prompt length. The abstract and Section 4.1 attribute the degradation to "such information" / "external information," which is a causal claim about content. This cannot be sustained without a control condition that matches the prompt structure while varying only the content—for example, inserting a random or irrelevant text of similar length, or explicitly stating "No additional information is provided." The length-correlation analysis in Figure 2 does not resolve this, since all points in that figure already have the extra sentence present; length is not presence. This issue must be fixed or the claims rephrased as "adding an instruction sentence together with external information degrades performance."
[Section 3.1 / Table 1] The paper reports single runs with no variance estimates or significance tests. Section 3.1 justifies single runs by temperature zero producing "negligible variation in performance," but this does not eliminate sampling variability in the test set or sensitivity to prompt and decoding details, and API models can be nondeterministic. Many of the reported changes are small (e.g., -0.2% accuracy for Llama-3.1-8B on COVID-19-Stance with Wikipedia, Table 1), so without standard errors or bootstrap confidence intervals the sign of the change is not reliable. The aggregate claim "external information decreases performance in most cases" is literally a count of negative cells in Table 1, so the absence of uncertainty quantification is load-bearing. Please provide repeated runs or resampling-based confidence intervals, at least for the main no-info-vs-Wikipedia comparison, and report how many individual differences are statistically distinguishable from zero.
[Section 4.2, Eq. (1), and Figure 2] The NAR/HAR analysis and the length correlation inherit the confound described in the first major comment. The definition of NAR compares M(x|t,T) with M(x|∅,T), where the latter omits the entire instruction sentence; the observed positive net adoption could therefore reflect the model reacting to the added sentence rather than to the stance or sentiment of the information content. Figure 2 correlates length with prediction rates and relative performance only within information-present conditions, so it does not test whether the presence of the extra sentence is the cause. Additionally, s = M(t|∅,T) is the model's stance or sentiment of the information obtained through a separate, simpler prompt (Appendix B), which is a different task from stance detection of the tweet; the paper does not validate that this isolated perception matches how the model internally represents the information when it is embedded in the detection prompt. Please either add the missing control or explicitly limit the NAR/HAR claims to "under this prompt structure," and consider validating the perceived-stance prompt against a fixed external sentiment model.

minor comments (6)

[Appendix B] The omitted sentences are marked in red, which is not visible in a grayscale printout; please use a clearer notation such as brackets or a footnote indicator.
[Section 3.1] The claim of "negligible variation in performance" should be quantified; if possible, report the observed variance from a few pilot runs or a range rather than an informal assertion.
[Figure 2] The p-values are reported without multiple-comparison correction; with four correlations, the one significant result (p=0.004) should be interpreted cautiously, and the text should note this.
[Section 4.4 / Table A2] The fine-tuning analysis reports counts of instances with negative relative performance but does not distinguish large negative changes from small ones near zero; adding a magnitude summary (as in Table A3) in the main text would help the reader judge the practical relevance.
[Section 3.2] The external information for SemEval 2016 Task 6 was collected by the authors via the Wikipedia API; a brief description of the extraction process and any manual quality checks would improve reproducibility.
[Section 6] The Limitations section does not mention the prompt-structure difference between the no-info and info conditions; even if the control experiments are added, this limitation should be discussed explicitly.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central degradation finding is an empirical comparison against a no-information baseline, not a result forced by the paper's definitions or by self-citation.

full rationale

This paper is an empirical evaluation and contains no mathematical derivation chain whose conclusion is equivalent to its inputs by construction. The central claim that external information degrades LLM stance detection is supported by direct measurements of macro F1 and accuracy with and without external information (Table 1), where the no-information baseline is an independent condition rather than a fitted parameter or a model output reused as the prediction. The NAR/HAR metrics are self-referential in that s is defined as the model's own perceived stance or sentiment of the information (Equation 1: 's = M(t|∅,T) is the stance or sentiment ... M detects for t'), but this does not force the degradation conclusion; NAR merely quantifies how often predictions move toward the model's perception, and HAR separately measures whether such moves are harmful or helpful. The explanation that models adopt information stance is thus an interpretive statistic, not a tautology that determines the headline result. There are no load-bearing self-citations: the papers cited for external information and BERT baselines (He et al., 2022; Li et al., 2021; Mohammad et al., 2016) are prior work by other authors, and no uniqueness theorem or prior result by the present authors is invoked to rule out alternatives. The reviewer's concern that the no-information prompt omits the entire instruction sentence introducing the external information is a potential confound about causal attribution, but it is not circularity: the comparison is still an empirical measurement, and the confound does not make the reported degradation equivalent to the paper's own assumptions by definition. For these reasons, the appropriate finding is no significant circularity.

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

The study is empirical; the central claim depends on experimental assumptions rather than fitted parameters or new entities. The main assumptions are that temperature-zero outputs are deterministic and single-run results are representative, that the no-information condition is a valid control, and that the chosen Wikipedia/web excerpts represent the external information used in practice.

assumptions (3)

domain assumption LLM outputs are deterministic at temperature 0, so single runs are representative.
Section 3.1 states temperature 0 is used and results are based on single runs; no variance or multiple seeds are reported.
domain assumption The no-information baseline is a valid control, despite the prompt sentence being omitted.
Appendix B shows the no-info prompt lacks the sentence 'The following information can be helpful:', so the comparison may be confounded by prompt structure rather than content alone.
domain assumption Wikipedia excerpts (from He et al., 2022) and web-search summaries are typical external information for stance detection.
Section 3.2 describes the external information sources; the relevance of these particular excerpts to real-world deployment is assumed.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Is External Information Useful for Stance Detection with LLMs?." pith.science (2026). https://pith.science/paper/V5DBESZD

@misc{pith2026250701543,
  author       = {Pith},
  title        = {Pith review of: Is External Information Useful for Stance Detection with LLMs?},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/V5DBESZD}},
  note         = {Machine review of arXiv:2507.01543}
}

read the original abstract

In the stance detection task, a text is classified as either favorable, opposing, or neutral towards a target. Prior work suggests that the use of external information, e.g., excerpts from Wikipedia, improves stance detection performance. However, whether or not such information can benefit large language models (LLMs) remains an unanswered question, despite their wide adoption in many reasoning tasks. In this study, we conduct a systematic evaluation on how Wikipedia and web search external information can affect stance detection across eight LLMs and in three datasets with 12 targets. Surprisingly, we find that such information degrades performance in most cases, with macro F1 scores dropping by up to 27.9\%. We explain this through experiments showing LLMs' tendency to align their predictions with the stance and sentiment of the provided information rather than the ground truth stance of the given text. We also find that performance degradation persists with chain-of-thought prompting, while fine-tuning mitigates but does not fully eliminate it. Our findings, in contrast to previous literature on BERT-based systems which suggests that external information enhances performance, highlight the risks of information biases in LLM-based stance classifiers. Code is available at https://github.com/ngqm/acl2025-stance-detection.

Figures

Figures reproduced from arXiv: 2507.01543 by the authors.

**Figure 1.** Left: Predictions are more likely to move towards information stance and sentiment than they are to move away. Right: When a model adopts information stance or sentiment, predictions are more likely to be wrong (harmful) than to be correct (helpful). mance changes with zero-shot (Kojima et al., 2022) and few-shot (Wei et al., 2022b) chain-of-thought (CoT) prompting. Our prompts explicitly instruct the model to incor… view at source ↗

**Figure 2.** Top: There is some correlation between external information length and FAVOR predictions but not NONE predictions. Bottom: There is no correlation between external information length and performance change due to such information. Details on correlation calculations are in Appendix D. 4.3 Variations in Prompting Strategies COVID-19 -Stance P-Stance SemEval 2016 Task 6 -15% 0 (No Info.) +15% Relative Accuracy -15% 0… view at source ↗

**Figure 4.** Models become more robust but do not com [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗

Figures from the paper (1 more)

**Figure 3.** Figure 3: Use of external information degrades stance [PITH_FULL_IMAGE:figures/full_fig_p005_3.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

27 extracted references · 15 canonical work pages

[1]

Anthropic. 2024. https://www-cdn.anthropic.com/de8ba9b01c9ab7cbabf5c33b80b7bbc618857627/Model_Card_Claude_3.pdf The Claude 3 Model Family : Opus , Sonnet , Haiku . Technical report, Anthropic

work page 2024
[2]

Michael Han Daniel Han and Unsloth team. 2023. http://github.com/unslothai/unsloth Unsloth

2023
[3]

Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Amy Yang, Angela Fan, and 1 others. 2024. The llama 3 herd of models. arXiv preprint arXiv:2407.21783

arXiv 2024
[4]

Kyle Glandt, Sarthak Khanal, Yingjie Li, Doina Caragea, and Cornelia Caragea. 2021. Stance detection in covid-19 tweets. In Proceedings of the 59th annual meeting of the association for computational linguistics and the 11th international joint conference on natural language processing (long papers), volume 1

work page 2021
[5]

Daya Guo, Qihao Zhu, Dejian Yang, Zhenda Xie, Kai Dong, Wentao Zhang, Guanting Chen, Xiao Bi, Yu Wu, YK Li, and 1 others. 2024. Deepseek-coder: When the large language model meets programming--the rise of code intelligence. arXiv preprint arXiv:2401.14196

arXiv 2024
[6]

Zihao He, Negar Mokhberian, and Kristina Lerman. 2022. https://doi.org/10.18653/v1/2022.wassa-1.7 Infusing knowledge from W ikipedia to enhance stance detection . In Proceedings of the 12th Workshop on Computational Approaches to Subjectivity, Sentiment & Social Media Analysis , pages 71--77, Dublin, Ireland. Association for Computational Linguistics

doi:10.18653/v1/2022.wassa-1.7 2022
[7]

Edward J Hu, yelong shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. 2022. https://openreview.net/forum?id=nZeVKeeFYf9 Lo RA : Low-rank adaptation of large language models . In International Conference on Learning Representations

2022
[8]

Aaron Hurst, Adam Lerer, Adam P Goucher, Adam Perelman, Aditya Ramesh, Aidan Clark, AJ Ostrow, Akila Welihinda, Alan Hayes, Alec Radford, and 1 others. 2024. Gpt-4o system card. arXiv preprint arXiv:2410.21276

arXiv 2024

Show all 27 references

[9]

Shima Imani, Liang Du, and Harsh Shrivastava. 2023. https://doi.org/10.18653/v1/2023.acl-industry.4 M ath P rompter: Mathematical reasoning using large language models . In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 5: Indus...

2023 doi
[10]

Junsol Kim, James Evans, and Aaron Schein. 2025. https://openreview.net/forum?id=rwqShzb9li Linear representations of political perspective emerge in large language models . In The Thirteenth International Conference on Learning Representations

2025
[11]

Takeshi Kojima, Shixiang Shane Gu, Machel Reid, Yutaka Matsuo, and Yusuke Iwasawa. 2022. Large language models are zero-shot reasoners. Advances in neural information processing systems, 35:22199--22213

2022
[12]

Gonzalez, Hao Zhang, and Ion Stoica

Woosuk Kwon, Zhuohan Li, Siyuan Zhuang, Ying Sheng, Lianmin Zheng, Cody Hao Yu, Joseph E. Gonzalez, Hao Zhang, and Ion Stoica. 2023. Efficient memory management for large language model serving with pagedattention. In Proceedings of the ACM SIGOPS 29th Symposium on Operating S...

2023
[13]

Xiaochong Lan, Chen Gao, Depeng Jin, and Yong Li. 2024. Stance detection with collaborative role-infused llm-based agents. In Proceedings of the International AAAI Conference on Web and Social Media, volume 18, pages 891--903

2024
[14]

Ang Li, Bin Liang, Jingqian Zhao, Bowen Zhang, Min Yang, and Ruifeng Xu. 2023. Stance detection on social media with background knowledge. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 15703--15717

2023
[15]

Ang Li, Jingqian Zhao, Bin Liang, Lin Gui, Hui Wang, Xi Zeng, Kam-Fai Wong, and Ruifeng Xu. 2024. Mitigating biases of large language models in stance detection with calibration. arXiv preprint arXiv:2402.14296

2024 arXiv
[16]

Yingjie Li, Tiberiu Sosea, Aditya Sawant, Ajith Jayaraman Nair, Diana Inkpen, and Cornelia Caragea. 2021. P-stance: A large dataset for stance detection in political domain. In Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, pages 2355--2365

2021
[17]

The geometry of truth: Emergent linear structure in large language model representations of true/false datasets

Samuel Marks and Max Tegmark. The geometry of truth: Emergent linear structure in large language model representations of true/false datasets. In First Conference on Language Modeling
[18]

Saif Mohammad, Svetlana Kiritchenko, Parinaz Sobhani, Xiaodan Zhu, and Colin Cherry. 2016. Semeval-2016 task 6: Detecting stance in tweets. In Proceedings of the 10th international workshop on semantic evaluation (SemEval-2016), pages 31--41

2016
[19]

OpenAI . 2025 a . https://platform.openai.com/docs/guides/tools-web-search?api-mode=chat Openai api web search

2025
[20]

OpenAI . 2025 b . https://github.com/openai/tiktoken tiktoken

2025
[21]

Melanie Sclar, Yejin Choi, Yulia Tsvetkov, and Alane Suhr. 2024. https://openreview.net/forum?id=RIu5lyNXjT Quantifying language models' sensitivity to spurious features in prompt design or: How i learned to start worrying about prompt formatting . In The Twelfth International...

2024
[22]

Jason Wei, Yi Tay, Rishi Bommasani, Colin Raffel, Barret Zoph, Sebastian Borgeaud, Dani Yogatama, Maarten Bosma, Denny Zhou, Donald Metzler, and 1 others. 2022 a . Emergent abilities of large language models. Transactions on Machine Learning Research

2022
[23]

Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, and 1 others. 2022 b . Chain-of-thought prompting elicits reasoning in large language models. Advances in neural information processing systems, 35:24824--24837

2022
[24]

Maxwell Weinzierl and Sanda Harabagiu. 2024. Tree-of-counterfactual prompting for zero-shot stance detection. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 861--880

2024
[25]

Haoyang Wen and Alexander G Hauptmann. 2023. Zero-shot and few-shot stance detection on varied topics via conditional generation. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 1491--1499

2023
[26]

An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, and 1 others. 2024. Qwen2. 5 technical report. arXiv preprint arXiv:2412.15115

2024 arXiv
[27]

Qinglin Zhu, Bin Liang, Jingyi Sun, Jiachen Du, Lanjun Zhou, and Ruifeng Xu. 2022. Enhancing zero-shot stance detection via targeted background knowledge. In Proceedings of the 45th international ACM SIGIR conference on research and development in information retrieval, pages ...

2022

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [1]

Anthropic. 2024. https://www-cdn.anthropic.com/de8ba9b01c9ab7cbabf5c33b80b7bbc618857627/Model_Card_Claude_3.pdf The Claude 3 Model Family : Opus , Sonnet , Haiku . Technical report, Anthropic

work page 2024

[2] [2]

Michael Han Daniel Han and Unsloth team. 2023. http://github.com/unslothai/unsloth Unsloth

2023

[3] [3]

Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Amy Yang, Angela Fan, and 1 others. 2024. The llama 3 herd of models. arXiv preprint arXiv:2407.21783

arXiv 2024

[4] [4]

Kyle Glandt, Sarthak Khanal, Yingjie Li, Doina Caragea, and Cornelia Caragea. 2021. Stance detection in covid-19 tweets. In Proceedings of the 59th annual meeting of the association for computational linguistics and the 11th international joint conference on natural language processing (long papers), volume 1

work page 2021

[5] [5]

Daya Guo, Qihao Zhu, Dejian Yang, Zhenda Xie, Kai Dong, Wentao Zhang, Guanting Chen, Xiao Bi, Yu Wu, YK Li, and 1 others. 2024. Deepseek-coder: When the large language model meets programming--the rise of code intelligence. arXiv preprint arXiv:2401.14196

arXiv 2024

[6] [6]

Zihao He, Negar Mokhberian, and Kristina Lerman. 2022. https://doi.org/10.18653/v1/2022.wassa-1.7 Infusing knowledge from W ikipedia to enhance stance detection . In Proceedings of the 12th Workshop on Computational Approaches to Subjectivity, Sentiment & Social Media Analysis , pages 71--77, Dublin, Ireland. Association for Computational Linguistics

doi:10.18653/v1/2022.wassa-1.7 2022

[7] [7]

Edward J Hu, yelong shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. 2022. https://openreview.net/forum?id=nZeVKeeFYf9 Lo RA : Low-rank adaptation of large language models . In International Conference on Learning Representations

2022

[8] [8]

Aaron Hurst, Adam Lerer, Adam P Goucher, Adam Perelman, Aditya Ramesh, Aidan Clark, AJ Ostrow, Akila Welihinda, Alan Hayes, Alec Radford, and 1 others. 2024. Gpt-4o system card. arXiv preprint arXiv:2410.21276

arXiv 2024