Pith. sign in

REVIEW 3 major objections 5 minor 56 references

A Case Study of Cross-Lingual Zero-Shot Generalization for Classical Languages in LLMs

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

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

Pith's one-line read Large off-the-shelf LLMs generalize zero-shot to Sanskrit, Latin, and Ancient Greek across named-entity recognition, machine translation, and question answering, matching or beating fine-tuned baselines.

desk verdict Useful benchmark and evaluation for classical-language NLP, but the paper's strong zero-shot attribution claims are undercut by its own contamination caveat. read the letter →

arxiv 2505.13173 v2 pith:T3IAMDAO submitted 2025-05-19 cs.CL

V.S.D.S.Mahesh Akavarapu , Hrishikesh Terdalkar , Pramit Bhattacharyya , Shubhangi Agarwal , Vishakha Deulgaonkar , Pralay Manna , Chaitali Dangarikar , Arnab Bhattacharya This is my paper · ORCID
classification cs.CL
keywords cross-lingualzero-shotgeneralizationclassicallanguagesSanskritAncientGreekLatinnamedentityrecognitionmachinetranslationretrieval-augmentedgeneration
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 tests whether general-purpose large language models can handle three classical languages—Sanskrit, Ancient Greek, and Latin—without any fine-tuning on those languages. Across named-entity recognition, translation into English, and factoid question answering in Sanskrit, the largest models match or beat fine-tuned baselines on out-of-domain data, while smaller models fall behind on rare or abstract entity types. The paper argues that this performance is genuine cross-lingual generalization: the models were not instruction-tuned on these languages, and English prompts work better than native-language prompts, especially for Sanskrit. It also contributes a new 1,501-pair Sanskrit question-answering dataset and shows that retrieving relevant passages with BM25 plus a lemmatizer substantially improves exact-match answers. If the attribution holds, digital-humanities researchers can use existing LLMs directly for classical-language work instead of building specialized models.

What carries the argument

The operative mechanism is zero-shot prompting itself: no task-specific examples are supplied, so any success must come from the model's general linguistic competence. The paper manipulates four levers around that mechanism: model scale, prompt language, script, and retrieval context. The RAG pipeline is the empirical star: top-$k=4$ passages retrieved by BM25 from lemmatized source texts, using a Seq2Seq Sanskrit lemmatizer trained on the DCS corpus, raise exact-match QA scores sharply. The new 1,501-pair Sanskrit QA dataset, built from translated Hindi MCQ books on the Rāmāyaṇa and Bhāvaprakāśa nighaṇṭu, is the benchmark that makes the QA comparison possible. A knowledge-graph variant is explored but yields only modest gains, attributed to KG incompleteness.

What would settle it

Take the published Sanskrit QA pairs and the Ancient Greek MT test sentences, build matched items written only after the models' knowledge cut-off (same style, same sources, never posted online), and run the same zero-shot prompts on those fresh items; if accuracy on fresh items drops well below the published scores, the published results are contaminated and the cross-lingual claim collapses.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that high-capacity LLMs exhibit zero-shot cross-lingual generalization to three classical languages. GPT-4o and Llama-3.1-405B-instruct perform named entity recognition, translation into English, and factoid question answering in Sanskrit, Latin, and Ancient Greek at levels comparable to or above fine-tuned baselines on out-of-domain data, while their smaller counterparts degrade sharply on niche entity types and retrieval-based QA. Sanskrit QA scores improve markedly when BM25-retrieved context is provided, with a lemmatization step closing the gap caused by high inflection. English prompts outperform Sanskrit prompts for Sanskrit, which the paper reads as evidence that the models were not instruction-tuned on these languages; for Latin and Ancient Greek, larger models do equally well or better with native-language prompts. Devanagari input also beats IAST transliteration, consistent with transfer through shared script.

Load-bearing premise

The central claim collapses if models memorized any of the test items during pretraining or instruction tuning, and the paper's own Limitations section concedes that Ancient Greek MT shows anomalously high performance consistent with possible data contamination.

Editorial extensions

If this is right

  • If the attribution holds, researchers studying classical texts can run NER, translation, and factoid QA with current large LLMs out of the box, without fine-tuning task-specific models.
  • Sanskrit QA, a nearly empty benchmark space, now has a 1,501-pair dataset that future classical-language evaluations can use as a common reference point.
  • Retrieval-augmented generation with lemmatized lexical retrieval is a practical recipe for improving LLM answers in highly inflected languages, so classical-language NLP systems should include a lemmatizer before retrieval.
  • Model scale is the main lever: large models separate confusable mythological and abstract entity classes and use context effectively, while small models do not.
  • The Devanagari-versus-IAST gap implies that transliteration choice changes measured performance, so classical-language benchmarks should report script explicitly.

Reading between the lines

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

  • A direct testable extension of the orthographic argument: if script sharing drives transfer, models pretrained on Devanagari-rich corpora should show a larger Devanagari-over-IAST gap for Sanskrit than models with little Devanagari exposure.
  • Because the QA pairs were translated from Hindi MCQ books, some measured 'comprehension' may be inherited from the translation process; generating questions natively in Sanskrit from primary texts would separate language competence from source-question artifacts.
  • The contamination caveat for Ancient Greek MT suggests a natural control: compare scores on the published test set against a fresh, post-cutoff set of aligned Ancient Greek sentences; a large drop would confirm memorization.
  • If cross-lingual transfer flows through lexical influence—Latin and Greek into English, Devanagari through Hindi and Marathi—then transfer strength should correlate with measurable vocabulary overlap, which could help predict which other under-resourced languages LLMs will handle zero-shot.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper studies zero-shot cross-lingual generalization of GPT-4o, GPT-4o-mini, Llama-3.1-405b-instruct, and Llama-3.1-8b-instruct on named entity recognition and machine translation into English for Sanskrit, Ancient Greek, and Latin, and on factoid question answering for Sanskrit using a newly introduced 1501-pair dataset. The authors report that larger models match or outperform fine-tuned baselines on the out-of-domain settings they test, that English prompts are usually better for smaller models, that RAG with BM25 and a custom lemmatizer improves Sanskrit QA, and that smaller models struggle on niche entity types and in using retrieved context. The paper attributes the observed behavior to cross-lingual generalization from pretraining, arguing that the models were not instruction tuned on these classical languages.

Significance. If the attribution to pure zero-shot cross-lingual generalization were established, this would be a useful contribution to multilingual LLM evaluation and digital humanities: it contributes a new Sanskrit QA dataset, evaluates several open and proprietary models under a uniform protocol, and reports informative context effects through RAG and KG-based prompting. The out-of-domain comparisons and confusion-matrix analyses are also useful empirical material. However, the central significance depends on the contamination assumption, which the paper itself concedes is questionable for Ancient Greek MT. Until that is addressed, the results are best read as a descriptive benchmark under a zero-shot prompting protocol rather than as proof that performance is solely due to cross-lingual generalization.

major comments (3)
  1. The central claim in Section 4 that zero-shot performance is 'attributable solely to cross-lingual generalization' is not supported by the evidence. The only support offered in Sections 1 and 2 is that datasets are recent relative to knowledge cut-offs and that documentation does not mention instruction tuning. The Limitations section then concedes that 'a few datasets we experimented were released within the models' knowledge cut-off dates raising the issue of data contamination,' and identifies Ancient Greek MT as 'anomalously high,' suggesting possible exposure. This is exactly the load-bearing condition for the paper's causal claim, and it is not tested. Release dates do not rule out contamination from source texts either, since Ramayana, Ayurvedic texts, Greek and Latin corpora are likely present in large-scale pretraining. The English-prompt advantage for Sanskrit only shows that instruction following is better in English; it does not show that Sanskrit task data or instruction examples were absent. The conclusion should either be reworded to report benchmark performance under a zero-shot prompting protocol, or the authors should add a direct contamination analysis (e.g., overlap checks, memorization probes, or held-out re-annotation) before claiming sole cross-lingual generalization.
  2. The conclusion that 'model scale is an important factor' (Abstract and Section 4) is weakened by the design: the large/small comparisons are between different model families and instruction-tuned variants (gpt-4o vs. gpt-4o-mini; llama-3.1-405b-instruct vs. llama-3.1-8b-instruct). Observed differences could reflect family-specific training data, alignment, decoding, or API behavior rather than parameter count alone. For example, in Table 7 the smaller models are absent from the LLM-KG comparison, and in Table 2 the smaller models show qualitatively different RAG gains from the larger models. The paper should either compare models matched on family and training procedure, or explicitly frame 'scale' as a confounded variable and support the conclusion with per-family analyses.
  3. The abstract states that LLMs perform 'equal to or better than fine-tuned baselines on out-of-domain data,' but the supporting evidence in Appendix E covers only two comparisons: one Sanskrit MT benchmark and one Latin NER benchmark. These are useful indicative results, but they do not support a general claim across all three languages and both tasks. The wording should be narrowed to the specific datasets and settings evaluated, or additional out-of-domain comparisons should be reported.
minor comments (5)
  1. There is a typo in Appendix F.1: 'Python packaage' should be 'Python package'.
  2. In Table 10, 'occurances' should be 'occurrences,' and in Table 11, 'Occuring' should be 'Occurring'.
  3. The phrase 'With regards to how-questions' should be 'With regard to how-questions' for grammatical correctness.
  4. The sentence 'This does not imply instruction tuning in these languages, since larger and smaller models likely saw comparable amounts of tuning data' is speculative; it should be presented as an untested assumption rather than an established fact.
  5. The QA evaluation relies on manually curated acceptable answers and manual evaluation for a subset of reasoning questions; the authors should clarify in the main text how these resources will be released so that the evaluation is fully reproducible.

Circularity Check

0 steps flagged · score 1.0 of 10

No construction-level circularity: the paper reports measured benchmarks rather than deriving predictions from fitted inputs, so the central claim is not forced by construction; the unresolved contamination assumption is a validity caveat, not a circularity.

full rationale

This paper contains no mathematical derivation chain whose outputs reduce to its inputs. The central claim (Section 4) attributes zero-shot NER/MT/QA performance to cross-lingual generalization based on dataset release dates and an assumed absence of instruction tuning, and it is explicitly hedged in the Abstract: 'Assuming that models used such as GPT-4o and Llama-3.1 are not instruction fine-tuned on classical languages.' The Limitations section concedes that 'a few datasets we experimented were released within the models' knowledge cut-off dates raising the issue of data contamination' and that Ancient Greek MT may reflect exposure. That is an untested empirical assumption about data contamination, and the English-prompt superiority for Sanskrit is weak evidence for the attribution, but none of this is a case where a fitted parameter is renamed as a prediction or where a result is equivalent to its input by definition. The benchmarks are measured against external public datasets (Erdmann et al. 2019, Palladino et al. 2023, Myerston 2025, etc.) and against previously reported fine-tuned systems in Appendix E; the new Sanskrit QA dataset is a contributed resource, not a fitted target. The paper's self-citations (Terdalkar 2023 for the Sanskrit NER corpus; Terdalkar and Bhattacharya 2019/2021; Terdalkar et al. 2023 for knowledge graphs) supply data and tools, but they do not constitute a load-bearing uniqueness theorem or an ansatz smuggled in via citation. Accordingly, no specific circular step can be exhibited, and the honest finding is no significant circularity (score 1), with the contamination caveat recorded as a correctness risk rather than a circularity.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

No physical constants or new theoretical constructs are introduced. The ledger records validation-set hyperparameter tuning (RAG k, ToG limits) and the unverified contamination and instruction-tuning assumptions that carry the paper's central attribution to zero-shot cross-lingual generalization.

free parameters (2)
  • RAG context count k = 4
    Chosen by sweeping k in Figure 2 to maximize gpt-4o QA exact match on the test set; all reported RAG-BM25 numbers use this optimized value.
  • ToG depth and width limits = D=1, S=15, W=3
    Hand-selected for the LLM-KG experiments in Appendix F.1; the modest knowledge-graph QA results depend on these choices.
assumptions (3)
  • domain assumption The evaluated LLMs were not instruction-tuned on Sanskrit, Ancient Greek, or Latin.
    Section 1 and Section 4 state this as the basis for attributing results to cross-lingual generalization. The evidence is indirect: English prompts outperform Sanskrit prompts for Sanskrit, and documentation does not mention such tuning. This does not rule out exposure through pretraining or other tuning data.
  • domain assumption Datasets released after the models' knowledge cutoff were not seen during pretraining.
    Section 2 relies on cutoff dates to conclude the tasks are unseen. The Limitations section concedes that Ancient Greek MT may have been contaminated, so the assumption is not generally reliable.
  • domain assumption Observed large-versus-small model differences are due to model scale rather than model family, training data, or instruction-tuning differences.
    The paper compares gpt-4o with gpt-4o-mini and llama-3.1-405b with llama-3.1-8b. Only two model families are used, so scale is confounded with family and tuning recipe.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Case Study of Cross-Lingual Zero-Shot Generalization for Classical Languages in LLMs." pith.science (2026). https://pith.science/paper/T3IAMDAO

@misc{pith2026250513173,
  author       = {Pith},
  title        = {Pith review of: A Case Study of Cross-Lingual Zero-Shot Generalization for Classical Languages in LLMs},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/T3IAMDAO}},
  note         = {Machine review of arXiv:2505.13173}
}
read the original abstract

Large Language Models (LLMs) have demonstrated remarkable generalization capabilities across diverse tasks and languages. In this study, we focus on natural language understanding in three classical languages -- Sanskrit, Ancient Greek and Latin -- to investigate the factors affecting cross-lingual zero-shot generalization. First, we explore named entity recognition and machine translation into English. While LLMs perform equal to or better than fine-tuned baselines on out-of-domain data, smaller models often struggle, especially with niche or abstract entity types. In addition, we concentrate on Sanskrit by presenting a factoid question-answering (QA) dataset and show that incorporating context via retrieval-augmented generation approach significantly boosts performance. In contrast, we observe pronounced performance drops for smaller LLMs across these QA tasks. These results suggest model scale as an important factor influencing cross-lingual generalization. Assuming that models used such as GPT-4o and Llama-3.1 are not instruction fine-tuned on classical languages, our findings provide insights into how LLMs may generalize on these languages and their consequent utility in classical studies.

Figures

Figures reproduced from arXiv: 2505.13173 by the authors.

Figure 1
Figure 1. The three NLU tasks evaluated on the classical languages: Named-Entity Recognition (top-left), Machine [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Effect of k on RAG, denoting the number of top best matching text chunks retrieved, on the per￾formances of GPT-4o with retrievers based on BM25, averaged FastText (AvgFT) and GloVe (AvgGV) em￾beddings respectively of datasets Ram¯ aya ¯ n. a (left) and Bhavaprak ¯ a¯sanighan ´ . t .u (right). 4) from the original Sanskrit texts using BM25 (Sparck Jones, 1972; Robertson et al., 2009). We also compare BM25 with embed… view at source ↗
Figure 3
Figure 3. Zero-shot evaluation of LLMs on three NLU tasks for classical languages (language codes in ISO 639-2). [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Confusion matrices from the NER task in san (a-d), lat (e-h) and grc (i-l), all with <en> prompts, normalized across rows. (Muller et al., 2021; Fujinuma et al., 2022). To isolate this effect, we re-ran our Sanskrit NER and MT experiments (using English prompts) in Rom…
Figure 5
Figure 5. Figure 5: Overview of augmenting a LLM with a knowledge graph (KG) through Think-on-Graph (ToG) paradigm. [PITH_FULL_IMAGE:figures/full_fig_p014_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

56 extracted references · 30 canonical work pages

  1. [1]

    online" 'onlinestring :=

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

  2. [2]

    write newline

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

  3. [3]

    V.S.D.S.Mahesh Akavarapu and Arnab Bhattacharya. 2023. https://aclanthology.org/2023.wsc-csdh.6 Creation of a digital rig V edic index (anukramani) for computational linguistic tasks . In Proceedings of the Computational S anskrit & Digital Humanities: Selected papers presented at the 18th World S anskrit Conference , pages 89--96, Canberra, Australia (On...

    work page 2023

  4. [4]

    AnthropicAI. 2024. https://www.anthropic.com/ Claude-3.5-sonnet

    work page 2024

  5. [5]

    Marijke Beersmans, Evelien de Graaf, Tim Van de Cruys, and Margherita Fantoli. 2023. https://aclanthology.org/2023.alp-1.1/ Training and evaluation of named entity recognition models for classical L atin . In Proceedings of the Ancient Language Processing Workshop, pages 1--12, Varna, Bulgaria. INCOMA Ltd., Shoumen, Bulgaria

    work page 2023

  6. [6]

    Monica Berti. 2019. Digital classical philology: Ancient Greek and Latin in the digital revolution, volume 10. Walter de Gruyter GmbH & Co KG

    work page 2019

  7. [7]

    Piotr Bojanowski, Edouard Grave, Armand Joulin, and Tomas Mikolov. 2017. Enriching word vectors with subword information. Transactions of the association for computational linguistics, 5:135--146

    2017

  8. [8]

    Tom B Brown. 2020. Language models are few-shot learners. arXiv preprint arXiv:2005.14165

    arXiv 2020

Show all 56 references
  1. [9]

    Samuel Cahyawijaya, Holy Lovenia, and Pascale Fung. 2024. https://doi.org/10.18653/v1/2024.naacl-long.24 LLM s are few-shot in-context low-resource language learners . In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Ling...

    2024 doi

  2. [10]

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

    2024 arXiv

  3. [11]

    Alexander Erdmann, David Joseph Wrisley, Benjamin Allen, Christopher Brown, Sophie Cohen-Bod \'e n \`e s, Micha Elsner, Yukun Feng, Brian Joseph, B \'e atrice Joyeux-Prunel, and Marie-Catherine de Marneffe. 2019. https://doi.org/10.18653/v1/N19-1231 Practical, efficient, and c...

    2019 doi

  4. [12]

    Thomas Finkenstaedt and Dieter Wolff. 1973. Ordered profusion: Studies in dictionaries and the English lexicon. C. Winter

    1973

  5. [13]

    Yoshinari Fujinuma, Jordan Boyd-Graber, and Katharina Kann. 2022. https://doi.org/10.18653/v1/2022.acl-long.106 Match the script, adapt if multilingual: Analyzing the effect of multilingual pretraining on cross-lingual transferability . In Proceedings of the 60th Annual Meetin...

    2022 doi

  6. [14]

    Google. 2024. https://deepmind.google/ Gemini-1.5-pro

    2024

  7. [15]

    Pawan Goyal, G \'e rard Huet, Amba Kulkarni, Peter Scharf, and Ralph Bunker. 2012. https://aclanthology.org/C12-1062/ A distributed platform for S anskrit processing . In Proceedings of COLING 2012 , pages 1011--1028, Mumbai, India. The COLING 2012 Organizing Committee

    2012

  8. [16]

    Janghoon Han, Changho Lee, Joongbo Shin, Stanley Jungkyu Choi, Honglak Lee, and Kyunghoon Bae. 2024. https://doi.org/10.18653/v1/2024.findings-acl.912 Deep exploration of cross-lingual zero-shot generalization in instruction tuning . In Findings of the Association for Computat...

    2024 doi

  9. [17]

    2010-2024

    Oliver Hellwig. 2010-2024. http://www.sanskrit-linguistics.org/dcs/index.php Dcs - the digitial corpus of sanskrit

    2010

  10. [18]

    Oliver Hellwig, Sebastian Nehrdich, and Sven Sellmer. 2023. Data-driven dependency parsing of vedic sanskrit. Language Resources and Evaluation, 57(3):1173--1206

    2023

  11. [19]

    Vladimir Karpukhin, Barlas Oguz, Sewon Min, Patrick Lewis, Ledell Wu, Sergey Edunov, Danqi Chen, and Wen-tau Yih. 2020. https://doi.org/10.18653/v1/2020.emnlp-main.550 Dense passage retrieval for open-domain question answering . In Proceedings of the 2020 Conference on Empiric...

    2020 doi

  12. [20]

    u ttler, Mike Lewis, Wen-tau Yih, Tim Rockt \

    Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich K \"u ttler, Mike Lewis, Wen-tau Yih, Tim Rockt \"a schel, et al. 2020. Retrieval-augmented generation for knowledge-intensive nlp tasks. Advances in Neural Information Proc...

    2020

  13. [21]

    Ayush Maheshwari, Ashim Gupta, Amrith Krishna, Atul Kumar Singh, Ganesh Ramakrishnan, Anil Kumar Gourishetty, and Jitin Singla. 2024. https://aclanthology.org/2024.lrec-main.1245/ Samayik: A benchmark and dataset for E nglish- S anskrit translation . In Proceedings of the 2024...

    2024

  14. [22]

    Christopher D Manning. 2008. Introduction to information retrieval

    2008

  15. [23]

    Dan Melamed, Ryan Green, and Joseph P

    I. Dan Melamed, Ryan Green, and Joseph P. Turian. 2003. https://aclanthology.org/N03-2021 Precision and recall of machine translation . In Companion Volume of the Proceedings of HLT - NAACL 2003 - Short Papers , pages 61--63

    2003

  16. [24]

    MistralAI. 2024. https://mistral.ai/ Mistra-large-2

    2024

  17. [25]

    Niklas Muennighoff, Thomas Wang, Lintang Sutawika, Adam Roberts, Stella Biderman, Teven Le Scao, M Saiful Bari, Sheng Shen, Zheng Xin Yong, Hailey Schoelkopf, Xiangru Tang, Dragomir Radev, Alham Fikri Aji, Khalid Almubarak, Samuel Albanie, Zaid Alyafeai, Albert Webson, Edward ...

    2023 doi

  18. [26]

    Benjamin Muller, Antonios Anastasopoulos, Beno \^i t Sagot, and Djam \'e Seddah. 2021. https://doi.org/10.18653/v1/2021.naacl-main.38 When being unseen from m BERT is just the beginning: Handling new languages with multilingual language models . In Proceedings of the 2021 Conf...

    2021 doi

  19. [27]

    Jacobo Myerston. 2025. NEReus : A named entity corpus of ancient greek. https://github.com/jmyerston/NEReus. [Online; accessed 01-Feb-2025]

    2025

  20. [28]

    Sebastian Nehrdich and Oliver Hellwig. 2022. https://aclanthology.org/2022.lt4hala-1.3/ Accurate dependency parsing and tagging of L atin . In Proceedings of the Second Workshop on Language Technologies for Historical and Ancient Languages, pages 20--25, Marseille, France. Eur...

    2022

  21. [29]

    Sebastian Nehrdich, Oliver Hellwig, and Kurt Keutzer. 2024. https://doi.org/10.18653/v1/2024.findings-emnlp.805 One model is all you need: B y T 5- S anskrit, a unified model for S anskrit NLP tasks . In Findings of the Association for Computational Linguistics: EMNLP 2024, pa...

    2024 doi

  22. [30]

    OpenAI. 2024. https://openai.com/ Gpt-4o

    2024

  23. [31]

    OpenAI, Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al. 2023. https://arxiv.org/abs/2303.08774 Gpt-4 technical report . ArXiv, abs/2303.08774

    2023 arXiv

  24. [32]

    Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, et al. 2022. Training language models to follow instructions with human feedback. Advances in neural information processing systems, 3...

    2022

  25. [33]

    Wright, Anise d’Orange Ferreira, and Michel Ferreira dos Reis

    Chiara Palladino, Farnoosh Shamsian, Tariq Yousef, David J. Wright, Anise d’Orange Ferreira, and Michel Ferreira dos Reis. 2023. https://doi.org/10.5334/johd.131 Translation alignment for ancient greek: Annotation guidelines and gold standards . Journal of Open Humanities Data

    2023 doi

  26. [34]

    Jeffrey Pennington, Richard Socher, and Christopher Manning. 2014. https://doi.org/10.3115/v1/D14-1162 G lo V e: Global vectors for word representation . In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing ( EMNLP ) , pages 1532--1543, Doh...

    2014 doi

  27. [35]

    Rekha Phull and Gaurav Phull. 2017. Ayurveda Amrtam: MCQs on Laghutrayi & Medical Research in Ayurveda . Chaukhamba Surabharati Prakashana

    2017

  28. [36]

    Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, and Peter J Liu. 2020. Exploring the limits of transfer learning with a unified text-to-text transformer. Journal of machine learning research, 21(140):1--67

    2020

  29. [37]

    Ramkumar Rai. 1965. Valmiki-Ramayana Kosha: Descriptive Index to the Names and Subjects of Ramayana. Chowkhamba Sanskrit Series Office

    1965

  30. [38]

    Manmatha Natha Ray. 1984. An Index to the Proper Names Occuring in Valmiki's Ramayana . The Princess of Wales Sarasvati Bhavana studies: Reprint series. Sampurnanand Sanskrit University

    1984

  31. [39]

    Nils Reimers and Iryna Gurevych. 2019. https://doi.org/10.18653/v1/D19-1410 Sentence- BERT : Sentence embeddings using S iamese BERT -networks . In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference...

    2019 doi

  32. [40]

    Frederick Riemenschneider and Anette Frank. 2023. https://doi.org/10.18653/v1/2023.acl-long.846 Exploring large language models for classical philology . In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1...

    2023 doi

  33. [41]

    Stephen Robertson, Hugo Zaragoza, et al. 2009. The probabilistic relevance framework: Bm25 and beyond. Foundations and Trends in Information Retrieval , 3(4):333--389

    2009

  34. [42]

    Gill Rosenthal. 2023. Machina cognoscens: Neural machine translation for latin, a case-marked free-order language. Master's thesis, University of Chicago

    2023

  35. [43]

    Siba Sankar Sahu and Sukomal Pal. 2023. Building a text retrieval system for the sanskrit language: Exploring indexing, stemming, and searching issues. Computer Speech & Language, 81:101518

    2023

  36. [44]

    Jivnesh Sandhan, Laxmidhar Behera, and Pawan Goyal. 2023. https://doi.org/10.18653/v1/2023.eacl-main.158 Systematic investigation of strategies tailored for low-resource settings for low-resource dependency parsing . In Proceedings of the 17th Conference of the European Chapte...

    2023 doi

  37. [45]

    Rajendra Pratap Singh. 2009. 1000 Ramayana Prashnottari . Prabhat Prakashan

    2009

  38. [46]

    Karen Sparck Jones. 1972. A statistical interpretation of term specificity and its application in retrieval. Journal of documentation, 28(1):11--21

    1972

  39. [47]

    Jiashuo Sun, Chengjin Xu, Lumingyuan Tang, Saizhuo Wang, Chen Lin, Yeyun Gong, Lionel Ni, Heung-Yeung Shum, and Jian Guo. 2024. https://openreview.net/forum?id=nnVO1PvbTv Think-on-graph: Deep and responsible reasoning of large language model on knowledge graph . In The Twelfth...

    2024

  40. [48]

    Hrishikesh Terdalkar. 2023. Sanskrit Knowledge-based Systems: Annotation and Computational Tools. Ph.D. thesis, Indian Institute of Technology Kanpur

    2023

  41. [49]

    Hrishikesh Terdalkar and Arnab Bhattacharya. 2019. https://aclanthology.org/W19-7508 Framework for question-answering in S anskrit through automated construction of knowledge graphs . In Proceedings of the 6th International Sanskrit Computational Linguistics Symposium, pages 9...

    2019

  42. [50]

    Hrishikesh Terdalkar and Arnab Bhattacharya. 2021. https://doi.org/10.1145/3468264.3473113 Sangrahaka: A tool for annotating and querying knowledge graphs . In Proceedings of the 29th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundation...

    2021

  43. [51]

    Hrishikesh Terdalkar, Arnab Bhattacharya, Madhulika Dubey, S Ramamurthy, and Bhavna Naneria Singh. 2023. https://aclanthology.org/2023.wsc-csdh.11 Semantic annotation and querying framework based on semi-structured ayurvedic text . In Proceedings of the Computational S anskrit...

    2023

  44. [52]

    Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yasmine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava, Shruti Bhosale, et al. 2023. Llama 2: Open foundation and fine-tuned chat models. arXiv preprint arXiv:2307.09288

    2023 arXiv

  45. [53]

    Martin Volk, Dominic Philipp Fischer, Lukas Fischer, Patricia Scheurer, and Phillip Benjamin Str \"o bel. 2024. https://aclanthology.org/2024.lt4hala-1.15/ LLM -based machine translation and summarization for L atin . In Proceedings of the Third Workshop on Language Technologi...

    2024

  46. [54]

    Yizhong Wang, Swaroop Mishra, Pegah Alipoormolabashi, Yeganeh Kordi, Amirreza Mirzaei, Atharva Naik, Arjun Ashok, Arut Selvan Dhanasekaran, Anjana Arunkumar, David Stap, Eshaan Pathak, Giannis Karamanolakis, Haizhi Lai, Ishan Purohit, Ishani Mondal, Jacob Anderson, Kirby Kuzni...

    2022

  47. [55]

    Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Clement Delangue, Anthony Moi, Pierric Cistac, Tim Rault, Remi Louf, Morgan Funtowicz, Joe Davison, Sam Shleifer, Patrick von Platen, Clara Ma, Yacine Jernite, Julien Plu, Canwen Xu, Teven Le Scao, Sylvain Gugger, Mari...

    2020 doi

  48. [56]

    Yao Xu, Shizhu He, Jiabei Chen, Zihao Wang, Yangqiu Song, Hanghang Tong, Guang Liu, Jun Zhao, and Kang Liu. 2024. https://doi.org/10.18653/v1/2024.emnlp-main.1023 Generate-on-graph: Treat LLM as both agent and KG for incomplete knowledge graph question answering . In Proceedin...

    2024 doi

Pith tools

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