Pith. sign in

REVIEW 4 major objections 5 minor 38 references

Human Genome Book: Words, Sentences and Paragraphs

T0 review · 4 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read An English-trained GPT-2 model is claimed to parse a human chromosome into a structured book of DNA words, sentences, paragraphs, chapters, and an English translation.

desk verdict A transparent but unvalidated construction of a 'genome book'—worth a conversation, not a citation. read the letter →

arxiv 2501.16982 v1 pith:ISJEB5TC submitted 2025-01-23 q-bio.OT

Wang Liang This is my paper · ORCID
classification q-bio.OT
keywords humangenomebookDNAlanguagemodelmultilingualtransferGPT-2bytepairencodingsegmentationDNA-to-EnglishtranslationGRCh38
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 claims that linguistic structure learned from English text can be carried over to DNA, so that a GPT-2 model fine-tuned on English sentence splitting, paragraphing, and summarization can parse a human chromosome into a hierarchical book. Processing chromosome 1 of GRCh38.p14 yields DNA words, sentences, paragraphs, sections, chapters, and a DNA-to-English translated version; the first 10-megabase volume alone is split into 15,238 paragraphs, 503 sections, and 38 chapters. If true, this offers a multi-level index for DNA search, compact genome summarization, and fault-tolerant identifiers. The authors are explicit that the book is illustrative, that the DNA-side segmentation relies on an untested transfer hypothesis, and that no biological interpretation is established.

What carries the argument

The central object is a single GPT-2-small causal language model with a 100,000-token byte-pair-encoding vocabulary trained jointly on DNA, protein, and English text, creating a shared embedding space in which DNA and English word vectors can be pulled together by fine-tuning. Task-specific versions reuse the same language-modeling head: paragraph boundaries are encoded as a learned end-of-paragraph token, sentence boundaries by predicting the period token, and summaries by completing TL;DR-style prompts. Paragraph embeddings are then dynamically clustered into sections and chapters, and DNA-to-English translation is a nearest-neighbor lookup in the shared embedding space.

What would settle it

Run the three fine-tuned models on a long chromosome region and on shuffled and reverse-complemented versions of the same region, and record where predicted sentence boundaries fall relative to annotated splice sites and transcription start sites. If the boundary overlap is indistinguishable from random placement and the book structure does not change under shuffling, the claimed English-to-DNA transfer is unsupported.

Watch

Extended reading notes

Core claim

The central claim is that multilingual transfer, shown on an English semantic-similarity task and checked on DNA similarity datasets (accuracy above 79 percent), extends to three higher-level linguistic tasks: paragraph boundary prediction, sentence boundary prediction, and summarization. A shared GPT-2 model pre-trained on DNA, protein, and English text, then fine-tuned on English examples, is applied directly to raw DNA to produce the book. The resulting structure is a transfer artifact: paragraph and sentence boundaries appear wherever the English-trained language model expects paragraph breaks and periods, paragraph embeddings are clustered into sections and chapters, and DNA words are mapped to about 600 English words by nearest-neighbor embedding similarity. The authors state that this translation is formal, not semantic, and that the segmentation-to-DNA transfer is a hypothesis rather than a directly verified ability.

Load-bearing premise

The book structure collapses if sentence splitting, paragraphing, and summarization abilities trained only on English do not actually transfer to DNA, a transfer the authors state cannot be directly verified on the DNA side.

Editorial extensions

If this is right

  • If the claimed segmentation transfer holds, the genome book gives a hierarchical search index: chapter-level search narrows to a region, paragraph-level search matches specific sequence blocks, and sentence-level search isolates smaller variants.
  • Summaries can act as compact, fault-tolerant genome identifiers that tolerate point mutations and support quick comparison between genomes.
  • Chapter and section summaries offer a compressed and more human-readable representation of raw genomic data, potentially reducing storage overhead.
  • Because the pipeline is not specific to chromosome 1, the same recipe could generate books for all human chromosomes and for other species.

Reading between the lines

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

  • A test the paper does not run: compare predicted DNA sentence and paragraph boundaries against annotated functional landmarks such as splice sites, transcription start sites, and conserved elements; chance-level overlap would mean the book structure reflects token statistics rather than biological grammar.
  • An extension of the translation step: because 19,000 DNA tokens collapse onto roughly 600 English words via embedding proximity, the English book is a formal projection; any biological reading of the translated text would need independent validation.
  • A second stress test: shuffle or reverse-complement a chromosome region and rerun the pipeline; if the paragraph and chapter structure barely changes, the models are keying on local sequence statistics rather than long-range genomic context.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The paper proposes a method to treat the human genome as a book with DNA 'words,' 'sentences,' and 'paragraphs.' The authors pre-train a GPT-2 model on DNA, protein, and English data with a unified BPE tokenizer; fine-tune it on an English semantic-similarity task to enable 'language transfer' to DNA; then fine-tune it further on English paragraph segmentation, sentence splitting, and summarization tasks. The resulting models are applied to human chromosome 1 (described in places as the whole genome GRCh38.p14) to produce a hierarchical structure of volumes, paragraphs, sections, and chapters, with an 'English version' obtained by mapping DNA tokens to English words via embedding similarity. The central claim is that the resulting genome book is a linguistically structured, searchable representation of the genome.

Significance. If the claimed transfer of English paragraph, sentence, and summarization abilities to DNA were validated, the genome-book representation could offer a novel indexing and search scheme for genomic data. The paper is unusually transparent about its assumptions: Fig. 1 explicitly states that the segmentation and summarization abilities 'cannot be directly verified' and are hypothesized. This transparency is a strength, as is the availability of a pipeline and GitHub link. However, the current evidence does not establish the central claim: the only quantitative transfer validation (Table 2) concerns a similarity-judgment task, not segmentation, sentence splitting, or summarization; the hierarchical structure is never compared with known genomic annotations; and the DNA-to-English mapping is explicitly disclaimed as semantically meaningful. The potential significance is therefore contingent on validation that the paper does not provide.

major comments (4)
  1. [Fig. 1; Sections 2.4–2.6] The central construction rests on the hypothesis that paragraph segmentation, sentence splitting, and summarization abilities transfer from English to DNA. Fig. 1's caption concedes that these abilities 'cannot be directly verified' and are hypothesized, and Sections 2.4–2.6 provide no DNA-side evaluation for any of the three tasks. In particular, Section 2.5 detects sentence boundaries by predicting the '.' token, but DNA sequences contain no periods; without a control showing that these predictions align with known functional boundaries, the resulting 'sentences' are arbitrary artifacts of the model's output distribution. If the transfer hypothesis fails, the genome book collapses to tokenized DNA processed by an English-tuned language model, so this missing validation is load-bearing for the paper's main claim.
  2. [Table 2; Section 2.3] The only quantitative evidence for language transfer is the accuracy reported on DNA150s, DNA150, and DNA50 after fine-tuning on the English PAWSX similarity task. These datasets are introduced in the author's own previous work (ref. 30), and the table reports no baselines, error bars, or statistical significance. Since the underlying model was pre-trained on DNA, protein, and English data, the high accuracies may reflect DNA-specific pretraining rather than transfer from English. This table does not establish that English semantic-similarity abilities transfer to DNA, and it certainly does not validate transfer of segmentation, sentence splitting, or summarization, which are the abilities used in the main construction.
  3. [Section 2.7] The hierarchical structure of the book—paragraphs, sections, chapters—is produced by applying the fine-tuned models and by 'dynamic clustering' of paragraph vectors, but the clustering algorithm, distance metric, and cluster-number selection are not specified. More importantly, the resulting structure is never compared with known genomic annotations such as genes, exons, regulatory elements, or chromatin domains. Without any external ground truth, the claimed 'paragraphs,' 'sections,' and 'chapters' are not demonstrated to correspond to meaningful genomic units; the hierarchy could be an arbitrary consequence of the tokenizer and the clustering parameters.
  4. [Section 2.8] The DNA-to-English mapping is based on cosine similarity of embedding vectors, and the section explicitly states that the relationship 'does not imply semantic similarity.' The mapping reduces approximately 19,000 DNA-related English words to about 600 unique English vocabulary items. Given this explicit disclaimer and the massive vocabulary collapse, the 'English version' of the genome book cannot be described as a translation in any meaningful sense; it is a projection onto a small set of English tokens. This undermines the abstract's statement that an 'English-translated version' of the book was created.
minor comments (5)
  1. [Section 2.7 vs. Abstract/Introduction] The abstract and introduction state that the GRCh38.p14 human genome was processed into a book, but Section 2.7 says 'we focused primarily on processing Chromosome 1.' Please clarify whether the book covers the entire genome or only one chromosome, and if only chromosome 1, adjust the claims accordingly.
  2. [Section 2.7, last paragraph] The text says 'our pre-trained model has a maximum input size of 256 tokens,' but Section 2.2 states that GPT-2 accepts a maximum length of 1024 tokens. This inconsistency should be resolved.
  3. [Section 2.4, learning rate] The learning rate appears as '5×10−55×10−5' in the text; this is likely a typo for 5e-5. Please correct it.
  4. [References] References 24 and 27 both cite the same arXiv identifier 2407.11435; one of these citations is likely incorrect. Please verify the bibliography.
  5. [Section 2.5] The statement 'Experimental results demonstrated that this method can efficiently and accurately identify sentence boundaries' is not supported by any reported experiments or quantitative results in the paper; please either provide the evidence or soften the claim.

Circularity Check

1 steps flagged · score 4.0 of 10

The genome-book construction is transparently hypothetical, but the only quantitative evidence for the load-bearing language-transfer premise comes from the author's own earlier DNA similarity datasets, making the central support self-referential rather than externally validated.

  1. self citation load bearing [Section 2.3 (Table 2) and reference [30]]
    "Additionally, we employed the DNA similarity sequence judgment dataset from the same study to validate the language transfer capability. The results of testing the transfer of English language capabilities to DNA in the gpt2-gene-eng-ft model are as follows: Table.2 ... DNA150s, DNA150, and DNA50 are DNA sequence similarity judgment datasets constructed with varying lengths and strategies."

    The paper's central premise is that English-language capabilities transfer to DNA. The only quantitative demonstration is Table 2, whose test sets are taken from reference [30], the same author's prior work ('the same study'). No external, independently constructed benchmark is provided, and the construction of DNA150s/DNA150/DNA50 is not described beyond 'varying lengths and strategies.' Because the downstream genome-book segmentation, sentence splitting, and summarization models are all justified by this transfer premise, the load-bearing evidence reduces to a self-citation chain: the author's own datasets validate the author's own model, and the unvalidated transfer is then used to generate the book. The paper's Fig.

full rationale

The paper is unusually transparent: Fig. 1 states that the transfer of segmentation, sentence-splitting, and summarization abilities to DNA 'cannot be directly verified' and is hypothesized, and the conclusion describes the study as 'primarily illustrative.' Those admissions reduce (but do not eliminate) circularity because the authors do not hide the assumption. The genuinely load-bearing step is the claim that English-to-DNA transfer exists; this is supported only by Table 2, whose DNA similarity datasets come from the author's own earlier work (reference [30]). No external benchmark is cited, and the dataset construction is not specified, so the central premise rests on a self-citation rather than on independent falsifiable evidence. The construction of the genome book itself is a deterministic application of the fine-tuned models to GRCh38.p14; it is not a hidden re-use of a fitted parameter, and the DNA-to-English translation is explicitly disclaimed as non-semantic ('does not imply semantic similarity'). Thus there is no equation-level circularity, but the evidential support for the key transfer premise is self-referential. Score 4 reflects a load-bearing self-citation with a central artifact that still has independent content as a concrete construction.

Assumptions & free parameters 7 free parameters · 5 assumptions · 2 invented entities

The pipeline rests on several assumptions from the same author's prior work, on manually chosen structural parameters (25 segments, 3 levels), and on an untested transfer hypothesis. No new physical entities are postulated; the book units and translation dictionary are model-derived constructs without independent biological evidence.

free parameters (7)
  • 25 chromosome segments ('volumes') = 25
    Chromosome 1 is split into 25 segments of about 10 Mb each; the number is chosen for book-like structure, not from data (Section 2.7).
  • number of hierarchy levels = 3
    The hierarchy is capped at paragraphs, sections, and chapters because 38 chapters 'align well with the typical structure of a book' (Section 2.7).
  • BPE vocabulary size = 100,000
    Vocabulary size for the unified DNA/protein/English tokenizer; no ablation is provided (Section 2.2).
  • pretraining epochs = 3 to 5
    The model was trained for a stated range of 3 to 5 epochs with a dynamic learning rate, so the exact schedule is not fixed (Section 2.2).
  • paragraph segmentation learning rate = 5e-5
    Initial learning rate for the paragraph-end model; other fine-tuning rates are not reported (Section 2.4).
  • summary input length = 256 tokens
    Summarization uses only the first 256 tokens of each sequence, a cutoff chosen without experimentation (Section 2.7).
  • clustering parameters for sections/chapters = unspecified
    Dynamic clustering is mentioned but no algorithm, distance metric, or number-of-clusters criterion is given (Section 2.7).
assumptions (5)
  • domain assumption A unified BPE tokenizer over DNA, protein, and English enables cross-domain transfer.
    Used throughout Section 2.2; no experiment isolates whether the unified tokenizer is necessary for the reported similarity transfer.
  • domain assumption Fine-tuning on English semantic similarity transfers to DNA sequence similarity.
    Reported in Table 2, but only on DNA datasets from the author's own ref 30, with no independent benchmark (Section 2.3).
  • ad hoc to paper English paragraph, sentence, and summarization capabilities transfer to DNA.
    Explicitly hypothesized in Fig. 1 caption and applied in Sections 2.4-2.6 with no DNA-side validation.
  • domain assumption Nearest-neighbor embedding similarity provides a useful DNA-to-English translation.
    Used in Section 2.8; the authors themselves state it does not imply semantic similarity.
  • domain assumption Clustering paragraph vectors yields meaningful sections and chapters.
    Applied in Section 2.7; no comparison with known genome features such as genes or regulatory domains.
invented entities (2)
  • The human genome book with DNA words, sentences, paragraphs, sections, and chapters
    purpose: Provides a hierarchical, searchable representation of genomic sequence
    The units are generated under an untested transfer hypothesis and are not tied to any independent biological annotation; the English translation is explicitly non-semantic (Section 2.8, Fig. 1 caption).
  • DNA-to-English translation dictionary
    purpose: Creates an 'English version' of the genome book
    Derived solely from embedding similarity in the author's model; collapses about 19,000 DNA words to about 600 English words, and is stated not to imply semantic equivalence (Section 2.8).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Human Genome Book: Words, Sentences and Paragraphs." pith.science (2026). https://pith.science/paper/ISJEB5TC

@misc{pith2026250116982,
  author       = {Pith},
  title        = {Pith review of: Human Genome Book: Words, Sentences and Paragraphs},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ISJEB5TC}},
  note         = {Machine review of arXiv:2501.16982}
}
read the original abstract

Since the completion of the human genome sequencing project in 2001, significant progress has been made in areas such as gene regulation editing and protein structure prediction. However, given the vast amount of genomic data, the segments that can be fully annotated and understood remain relatively limited. If we consider the genome as a book, constructing its equivalents of words, sentences, and paragraphs has been a long-standing and popular research direction. Recently, studies on transfer learning in large language models have provided a novel approach to this challenge.Multilingual transfer ability, which assesses how well models fine-tuned on a source language can be applied to other languages, has been extensively studied in multilingual pre-trained models. Similarly, the transfer of natural language capabilities to "DNA language" has also been validated. Building upon these findings, we first trained a foundational model capable of transferring linguistic capabilities from English to DNA sequences. Using this model, we constructed a vocabulary of DNA words and mapped DNA words to their English equivalents.Subsequently, we fine-tuned this model using English datasets for paragraphing and sentence segmentation to develop models capable of segmenting DNA sequences into sentences and paragraphs. Leveraging these models, we processed the GRCh38.p14 human genome by segmenting, tokenizing, and organizing it into a "book" comprised of genomic "words," "sentences," and "paragraphs." Additionally, based on the DNA-to-English vocabulary mapping, we created an "English version" of the genomic book. This study offers a novel perspective for understanding the genome and provides exciting possibilities for developing innovative tools for DNA search, generation, and analysis.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

38 extracted references · 24 canonical work pages

  1. [1]

    Direct Prediction of Paragraph Markers: Train the model to predict whether the next token is a paragraph marker after each token

    work page

  2. [2]

    [Original Text] TL;DR: [Summary]

    Binary Classification Problem: Transform the task into a binary classification problem, where the model determines if a given position marks a paragraph boundary. This involves adding an extra classification layer after each token, outputting 0 or 1 to indicate the absence or presence of a paragraph boundary. We opted for the relatively simpler Strategy 1...

    work page

  3. [3]

    Chromosome Volume

    Our approach involved dividing and structuring the chromosome into multiple hierarchical levels, as detailed below: Chromosome Part: We first divided Chromosome 1 into 25 segments, each containing approximately 10 MB of DNA sequence. This initial division is referred to as the "Chromosome Volume" level. DNA Paragraphs: For each segment, we applied a segme...

    work page

  4. [4]

    Melissa Sanabria ， etc.The human genome’s vocabulary as proposed by the DNA language model GROVER.bioRxiv 2023.07.19.549677; doi: https://doi.org/10.1101/2023.07.19.549677

    work page doi:10.1101/2023.07.19.549677 2023

  5. [5]

    The linguistics of DNA: words, sentences, grammar, phonetics, and semantics

    Ji S. The linguistics of DNA: words, sentences, grammar, phonetics, and semantics. Linguistics. 1999 May 18

    work page 1999

  6. [6]

    Statistical and linguistic features of DNA sequences

    Havlin S, Buldyrev SV, Goldberger AL, Mantegna RN, Peng CK, Simons M, Stanley HE. Statistical and linguistic features of DNA sequences. Fractals. 1995 Jun;3(02):269-84

    work page 1995

  7. [7]

    A book of life?: How the genome became an information system and DNA a language

    Kay LE. A book of life?: How the genome became an information system and DNA a language. Perspectives in biology and medicine. 1998;41(4):504-28

    work page 1998

  8. [8]

    Segmenting DNA sequence into words based on statistical language model

    Liang W. Segmenting DNA sequence into words based on statistical language model. Nature Precedings. 2012 Feb 27:1-

    work page 2012

Show all 38 references
  1. [9]

    Is DNA a language?

    Tsonis AA, Elsner JB, Tsonis PA. Is DNA a language?. Journal of theoretical Biology. 1997 Jan 7;184(1):25-9

    1997

  2. [10]

    Word organization in coding DNA: a mathematical model

    Mukhopadhyay I, Som A, Sahoo S. Word organization in coding DNA: a mathematical model. Theory in Biosciences. 2006 Jun 16;125(1):1-7

    2006

  3. [11]

    Testing DNA code words properties of regular languages

    Zaccagnino R, Zizza R, Zottoli C. Testing DNA code words properties of regular languages. Theoretical Computer Science. 2015 Dec 10;608:84-97

    2015

  4. [12]

    Learning the language of DNA

    Theodoris CV. Learning the language of DNA. Science. 2024 Nov 15;386(6723):729-30

    2024

  5. [13]

    Distinguishing word identity and sequence context in DNA language models

    Sanabria M, Hirsch J, Poetsch AR. Distinguishing word identity and sequence context in DNA language models. BMC bioinformatics. 2024 Sep 13;25(1):301

    2024

  6. [14]

    Generative language models on nucleotide sequences of human genes

    İhtiyar, M.N., Özgür, A. Generative language models on nucleotide sequences of human genes. Sci Rep 14, 22204 (2024). https://doi.org/10.1038/s41598-024-72512-x

    2024 doi

  7. [15]

    Durrant, Brian Kang, Dhruva Katrekar, David B

    Nguyen, Eric, Michael Poli, Matthew G. Durrant, Brian Kang, Dhruva Katrekar, David B. Li, Liam J. Bartie et al. Sequence modeling and design from molecular to genome scale with Evo. Science 386, no. 6723 (2024): eado9336

    2024

  8. [16]

    Zhou, et al., Dnabert-2: Efficient foundation model and benchmark for multi-species genome, arXiv preprint arXiv:2306.15006 (2023)

    Z. Zhou, et al., Dnabert-2: Efficient foundation model and benchmark for multi-species genome, arXiv preprint arXiv:2306.15006 (2023)

    2023 arXiv

  9. [17]

    Nguyen, et al., Hyenadna: Long-range genomic sequence modeling at single nucleotide resolution, Advances in neural information processing systems 36 (2024)

    E. Nguyen, et al., Hyenadna: Long-range genomic sequence modeling at single nucleotide resolution, Advances in neural information processing systems 36 (2024)

    2024

  10. [18]

    Yang, et al., scbert as a large-scale pretrained deep language model for cell type annotation of single-cell rna-seq data, Nature Machine Intelligence 4, 852–866 (2022)

    F. Yang, et al., scbert as a large-scale pretrained deep language model for cell type annotation of single-cell rna-seq data, Nature Machine Intelligence 4, 852–866 (2022)

    2022

  11. [19]

    A. Elnaggar, et al., Prottrans: Toward understanding the language of life through selfsupervised learning, IEEE transactions on pattern analysis and machine intelligence 44, 7112–7127 (2021)

    2021

  12. [20]

    Brandes, D

    N. Brandes, D. Ofer, Y. Peleg, N. Rappoport, M. Linial, Proteinbert: a universal deeplearning model of protein sequence and function, Bioinformatics 38, 2102–2110 (2022)

    2022

  13. [21]

    Lin, et al., Evolutionary-scale prediction of atomic-level protein structure with a language model, Science 379, 1123–1130 (2023)

    Z. Lin, et al., Evolutionary-scale prediction of atomic-level protein structure with a language model, Science 379, 1123–1130 (2023)

    2023

  14. [22]

    & Davuluri, R

    Ji, Y., Zhou, Z., Liu, H. & Davuluri, R. V. DNABERT: Pre-trained bidirectional encoder representations from transformers model for DNA-language in genome. Bioinformatics 37, 2112–2120 (2021). doi:doi.org/10.1093/bioinformatics/btab083

    2021 doi

  15. [23]

    doi: https://doi.org/10.1101/2023.06.12.544594

    Veniamin Fishman, et al.GENA-LM: A Family of Open-Source Foundational DNA Language Models for Long Sequences.BioRxiv 2023.06.12.544594. doi: https://doi.org/10.1101/2023.06.12.544594

    2023 doi

  16. [24]

    BioRxiv 2023.07.11.548628; doi: https://doi.org/10.1101/2023.07.11.548628

    Daoan Zhang, et al.DNAGPT: A Generalized Pretrained Tool for Multiple DNA Sequence Analysis Tasks. BioRxiv 2023.07.11.548628; doi: https://doi.org/10.1101/2023.07.11.548628

    2023 doi

  17. [25]

    Wang, B. et al. Pre-trained Language Models in Biomedical Domain: A Systematic Survey. ACM Comput. Surv.56, 1–52 (2023).doi:doi.org/10.1145/3611651

    2023 doi

  18. [26]

    & Gao, G

    Chen, Z., Wei, L. & Gao, G. Foundation models for bioinformatics. Quant. Biol. (2024) doi:10.1002/qub2.69

    2024 doi

  19. [28]

    Fei Yuan,etc.How Vocabulary Sharing Facilitates Multilingualism in LLaMA?arXiv (2024) doi:https://doi.org/10.48550/arXiv.2306.06688

    2024 doi arXiv

  20. [29]

    Alexis Conneau, Shijie Wu, Haoran Li, Luke Zettlemoyer, and Veselin Stoyanov. 2020. Emerging Cross-lingual Structure in Pretrained Language Models. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 6022–6034, Online. Association ...

    2020

  21. [30]

    Specialists: An Empirical Revisiting on Multilingual Transfer Ability

    Jiacheng Ye, Xijia Tao, Lingpeng Kong.Language Versatilists vs. Specialists: An Empirical Revisiting on Multilingual Transfer Ability. arXiv (2024) doi:10.48550/arxiv.2407.11435

    2024 doi arXiv

  22. [31]

    Hetong Wang, Pasquale Minervini, and Edoardo Ponti. 2024. Probing the Emergence of Cross-lingual Alignment during LLM Training. In Findings of the Association for Computational Linguistics: ACL 2024, pages 12159–12173, Bangkok, Thailand. Association for Computational Linguistics

    2024

  23. [32]

    arXiv(2020) https://arxiv.org/abs/2003.11080

    Junjie Hu, etc.XTREME: A Massively Multilingual Multi-task Benchmark for Evaluating Cross-lingual Generalization. arXiv(2020) https://arxiv.org/abs/2003.11080

    2020 arXiv

  24. [33]

    Can linguists better understand DNA?

    Liang W. Can linguists better understand DNA?. arXiv preprint arXiv:2412.07678. 2024 Dec 10

    2024 arXiv

  25. [34]

    BioRxiv 2024.07.01.600583; doi: https://doi.org/10.1101/2024.07.01.600583

    Thomas Hayes, et al.Simulating 500 million years of evolution with a language model. BioRxiv 2024.07.01.600583; doi: https://doi.org/10.1101/2024.07.01.600583

    2024 doi

  26. [35]

    LucaOne: Generalized Biological Foundation Model with Unified Nucleic Acid and Protein Language

    Yong He, et al. LucaOne: Generalized Biological Foundation Model with Unified Nucleic Acid and Protein Language. BioRxiv 2024.05.10.592927; doi: https://doi.org/10.1101/2024.05.10.592927

    2024 doi

  27. [36]

    Zhang, K., Zhou, R., Adhikarla, E. et al. A generalist vision–language foundation model for diverse biomedical tasks. Nat Med (2024). https://doi.org/10.1038/s41591-024-03185-2

    2024 doi

  28. [37]

    Minghao Xu, Xinyu Yuan, Santiago Miret, and Jian Tang. 2023. ProtST: multi-modality learning of protein sequences and biomedical texts. In Proceedings of the 40th International Conference on Machine Learning (ICML'23), Vol. 202. JMLR.org, Article 1615, 38749–38767

    2023

  29. [38]

    Zeyuan Wang, etc. 2024. InstructProtein: Aligning Human and Protein Language via Knowledge Instruction. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1114–1136, Bangkok, Thailand. Association for Compu...

    2024

  30. [39]

    paper github, https://github.com/maris205/genome_book

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.