A Split-then-Join Approach to Abstractive Summarization for Very Long Documents in a Low Resource Setting

Lhuqita Fazry

REVIEW 4 major objections 5 minor 27 references

A Split-then-Join Approach to Abstractive Summarization for Very Long Documents in a Low Resource Setting

T0 review · 4 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 The paper claims that splitting a very long document into 4,096-token chunks, summarizing each chunk, and selecting the chunk summary with the highest ROUGE-L score yields better whole-document summaries than truncating the document to…

desk verdict A sensible SPIN 3 selection rule is buried under an evaluation with no held-out split, so Table 3's gains are not yet evidence. read the letter →

arxiv 2505.06862 v1 pith:6VHIEC3H submitted 2025-05-11 cs.CL cs.IR

Lhuqita Fazry This is my paper · ORCID

classification cs.CLcs.IR

keywords abstractivesummarizationlongdocumentslow-resourceBIGBIRD-PEGASUSSPINROUGE-Ldocumentsplittingdataaugmentation

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 tries to establish that the usual fix for very long documents—cutting them off at the model's 4,096-token limit—throws away information needed for a good summary. It proposes a split-then-join procedure, called SPIN, that divides a document into 4,096-token chunks, summarizes each chunk, and then joins or selects those chunk summaries. On documents longer than 20,000 tokens from two large summarization corpora, the variant that keeps the single chunk summary with the highest ROUGE-L score, SPIN 3, outperforms both the truncated BIGBIRD-PEGASUS baseline and the other SPIN variants. A sympathetic reader would care because the result suggests that important summary content is often not in the document's opening section, and that chunk-level selection can recover it without a model that processes the whole document at once.

What carries the argument

The load-bearing mechanism is the split-then-join pipeline. During training, a document longer than 4,096 tokens is cut into chunks of at most 4,096 tokens, and the target summary is either split into the same number of parts or kept whole; each chunk is paired with the summary or summary part that gives the highest ROUGE-L recall, a longest-common-subsequence based overlap measure, producing many more training pairs from a small set. At inference, each chunk gets its own summary, and the variants differ only in how the chunk summaries are combined: SPIN 1 and SPIN 2 concatenate them, while SPIN 3 computes ROUGE-L between each chunk and its generated summary and keeps only the best one as the final summary. This selection rule is what carries the reported improvement.

What would settle it

Take a set of very long documents whose gold summaries require combining facts from widely separated chunks, apply SPIN 3, and compare with the truncated baseline; if ROUGE scores on that set fall below truncation, the independence assumption fails.

Watch

Extended reading notes

Core claim

The central claim, stated the way the author would state it, is that SPIN 3 is better than SPIN 1, SPIN 2, and BIGBIRD-PEGASUS for very long documents. In the paper's experiments on documents with more than 20,000 tokens, SPIN 3 reaches ROUGE-1 of 41.7 on scientific papers and 35.6 on patents, against 39.6 and 23.0 for the truncated BIGBIRD-PEGASUS baseline. The reason offered is that the most important information is not always at the head of a document; it can sit in the middle or near the end, so choosing the best chunk-level summary recovers content that truncation discards. The paper also reports that SPIN 1, which splits the target summary into parts matched to chunks, and SPIN 2, which pairs every chunk with the full summary but concatenates outputs, both perform worse than SPIN 3.

Load-bearing premise

The method assumes that a very long document can be cut into independent 4,096-token chunks, and that picking the single chunk summary with the highest ROUGE-L against its chunk also gives the best whole-document summary.

Editorial extensions

If this is right

If SPIN 3 is right, truncation is not a harmless preprocessing step for very long documents; discarding everything past 4,096 tokens can cost several ROUGE points.
The data-augmentation effect of splitting document-summary pairs means a low-resource long-document setting can be trained from fewer original examples.
Chunk summaries can be produced independently and in parallel, so inference cost scales with the number of chunks rather than with full-document attention.
The choice of selection metric matters: picking by ROUGE-L outperforms concatenating all chunk summaries, which is the paper's direct comparison between SPIN 3 and SPIN 2.

Reading between the lines

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

The paper does not test whether chunks are truly independent; a natural extension would be to build summaries that require evidence from multiple chunks and measure whether SPIN 3 still beats truncation.
Selecting by ROUGE-L against the chunk may implicitly favor summaries that stay lexically close to the chunk, which could matter for abstractive quality even if ROUGE improves.
The same chunk-and-select recipe could be applied to other long-input encoders, not just the BIGBIRD-PEGASUS architecture, to see whether the gain transfers.

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.

Desk Editor's Note

read the letter

Colleague,

The paper has a real idea buried under an unsupported evaluation. SPIN 3, selecting the chunk summary with highest ROUGE-L against its source chunk, is a simple but sensible extension of DANCER; the paper cites DANCER and frames itself as an adaptation. The empirical comparison of three splitting strategies on very long (>20k token) documents is new as far as I know. The motivation is practical: BIGBIRD-PEGASUS caps at 4,096 tokens and truncation is a poor default.

What the paper does well: the split-then-augment idea is a reasonable low-resource answer to domain shift, and the author is honest about building on DANCER. The code link is a plus.

The soft spots are load-bearing. The manuscript never says there is a held-out test set. Section 4.2 says the model was fine-tuned on the filtered arXiv and BigPatent documents; Table 3 then reports scores. With only a few thousand documents and a 576M-parameter model, those numbers could be memorization. No seeds, no error bars, no significance tests. The baseline (truncated BIGBIRD-PEGASUS) is not described in enough detail to know whether it was fine-tuned on the same data, with what hyperparameters, or on the same documents. That makes the central 'SPIN 3 is better' claim not independently verifiable.

There is also a circularity concern: ROUGE-L is used to pair summary parts to document parts (Algorithm 1), to select the final summary in SPIN 3, and ROUGE is the evaluation metric. That does not force the result, but it can inflate scores in ways the paper doesn't discuss. SPIN 1's ROUGE-1 drops to 11.7 from the 39.6 baseline, which suggests the part-pairing or training objective may be broken; the paper doesn't explain this. The pseudo-code has undefined variables (K, nparts), and the prose is full of typos.

As it stands, the numbers in Table 3 cannot be trusted. The paper is a draft, not a paper. For whom is it useful? Someone working on long-document summarization with BIGBIRD-PEGASUS might take the SPIN 3 selection rule as a trick to test. But the lack of a train/test split is a foundational problem, not a revision-level concern. I would not send this to peer review until the author reruns with held-out documents, multiple seeds, and ablations that separate augmentation from chunk selection. I'd tell the author to resubmit when that exists.

Referee Report

4 major / 5 minor

Summary. The manuscript proposes SPIN (Split-then-Join), an approach for abstractive summarization of very long documents whose length exceeds the 4,096-token input limit of BIGBIRD-PEGASUS. Documents are split into 4,096-token parts in training; three variants pair document parts with either split summary parts selected by ROUGE-L (SPIN 1) or with the full target summary (SPIN 2 and SPIN 3). At inference, SPIN 1 and SPIN 2 concatenate per-part summaries, while SPIN 3 selects the per-part summary with the highest ROUGE-L against its source chunk. The authors filter arXiv and BigPatent to documents over 20,000 tokens, fine-tune BIGBIRD-PEGASUS on the filtered sets, and report ROUGE-1/2/L scores in Table 3, claiming that SPIN 3 outperforms SPIN 1, SPIN 2, and the BIGBIRD-PEGASUS truncation baseline.

Significance. If the reported gains were obtained on held-out documents, the paper would offer a remarkably simple recipe for low-resource very-long-document summarization: fine-tune on chunked data and select the best chunk summary by ROUGE-L. The main strengths are the clearly stated splitting scheme, the public source-code link, and the use of two standard long-document datasets. However, the significance is conditional because the experimental protocol does not establish generalization: the manuscript never describes a train/test split, and the ROUGE-L selection rule is part of both the training-pair construction and the inference-time choice, while ROUGE is also the evaluation metric. As submitted, the central empirical claim is not supported.

major comments (4)

[Section 4, Table 3] No train/test split is described anywhere in the manuscript. Section 4.1 filters arXiv and BigPatent to documents over 20,000 tokens, and Section 4.2 states that the model was fine-tuned on those datasets, but there is no statement of a held-out evaluation set, seed count, or error bars. With only 3,533 arXiv and 6,589 BigPatent documents and a 576M-parameter model, scores computed on the training documents could reflect memorization rather than summarization ability. The central claim in Section 5 that SPIN 3 is better than the baselines therefore lacks the required evidence. The authors must specify the exact split, report results on the held-out portion, and provide variance estimates.
[Section 3.2, Algorithm 1, Eq. (1), Table 3] ROUGE-L is embedded twice in the pipeline: Algorithm 1 uses ROUGE-L recall to pair document parts with summary parts in training, and SPIN 3 uses ROUGE-L at inference to select the final summary, while the evaluation in Table 3 is also based on ROUGE. This does not force the reported numbers by construction, but it means the inference-time selection criterion is the same metric family used for evaluation, which can inflate scores by selecting the most extractive or most memorized chunk summary rather than the most informative one. The authors should evaluate with a metric or protocol that is not used in selection, and should report abstractiveness or overlap statistics to show the selected summaries are not simply extractive fragments.
[Section 3.1.2, Section 3.2, Algorithm 1] The description of SPIN 2 and SPIN 3 is inconsistent with Algorithm 1. Section 3.1.2 says that in SPIN 2 and SPIN 3 the target summary is not split and each document part is paired with the full target summary, but Algorithm 1 always splits the target summary into `n_parts` pieces and pairs each document part with a split summary part using ROUGE-L. As written, the algorithm cannot reproduce SPIN 2 or SPIN 3. Furthermore, Algorithm 1 uses undefined variables: `K` in line 5, `nparts` in line 9, and `lk` is computed but not used. The manuscript must provide a precise, self-contained description of each variant so the experiments are reproducible.
[Section 5, Table 3] The explanation for SPIN 2's poor performance contradicts the method definition. Section 5 says 'by splitting summaries with a fixed length equal to the number of document parts make the document-summary pair become unrelevant,' but Section 3.1.2 states that SPIN 2 does not split the target summary. This contradiction must be resolved, and the actual training procedure for the baseline ('standard training') is also under-specified: the manuscript does not state how truncation was applied, how many epochs were run, what learning rate and optimizer were used, or whether the baseline received the same augmented data. Without this information, the comparison in Table 3 is not independently reproducible or verifiable.

minor comments (5)

[Title and Abstract] There are several typos and formatting errors, including 'V ery' in the title, 'reasearch', 'it's', and 'respectivelly'. These should be corrected.
[Table 3 caption] The caption reads 'Result comparison on dataset BigPatent' but the table reports results for both arXiv and BigPatent. The caption should name both datasets.
[Section 4.1] The statement that 'only arXiv and BigPatent contain enough samples' among summarization datasets is unsupported; a citation or a brief survey of dataset statistics is needed.
[Section 3.2] The inference procedure is described only in prose; a pseudocode block for the generation and selection steps of SPIN 1, SPIN 2, and SPIN 3 would improve clarity and reproducibility.
[References] Some references lack page ranges or are formatted inconsistently (for example, the entry for Nenkova 2011 and the Radford et al. entry). The reference list should be checked against the venue's style.

Circularity Check

0 steps flagged · score 0.0 of 10

No constructional circularity: ROUGE-L is used as a selection and pairing heuristic, not as a definitional substitute for the reference-based evaluation.

full rationale

The paper's derivation chain is: (i) training pairs are built by splitting long documents and summaries and matching each document part to the summary part with the highest ROUGE-L (Eq. 1, Algorithm 1); (ii) a fine-tuned BIGBIRD-PEGASUS generates per-chunk summaries; (iii) SPIN 3 selects the chunk summary with the highest ROUGE-L against its source chunk; (iv) Table 3 reports ROUGE scores against the human reference summary. Although ROUGE-L is embedded in both the training-pair construction and the SPIN 3 selection heuristic, the final generated summary is not derived from the reference summary, and the reported ROUGE score against the reference is not algebraically forced by the selection objective. The selection criterion is a plausible proxy, not a definitional equivalence. There are no load-bearing self-citations, no imported uniqueness theorems, and no ansatz smuggled in via citation. The paper's principal weakness is evidentiary rather than circular: no train/test split is described, so Table 3 may reflect in-sample fit or memorization. That is a correctness risk, not a circularity, and it does not make the derivation equivalent to its inputs by construction.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

The central contribution rests on two hand-made algorithmic choices: the 20k-token filter and ROUGE-L-based pairing/selection. No new physical or mathematical entities are introduced, but the method embeds the evaluation metric family in both training data construction and inference selection.

free parameters (1)

20,000-token minimum document length = 20,000
Section 4.1: 'We filter out all documents which length less than 20,000 to focus on very long documents.' This threshold is chosen by hand, not derived from data or theory, and it defines the evaluation subset, so the central result depends on it.

assumptions (4)

domain assumption A very long document can be split into independent 4,096-token chunks and summarized chunk-by-chunk without loss of whole-document coherence.
Section 3 states the SPIN training and generation procedure; no evidence is provided that chunk independence holds for the 20k+ token documents in the filtered subsets.
ad hoc to paper ROUGE-L between a document part and a generated summary is a valid proxy for selecting the best part summary in SPIN 3.
Section 3.2: 'we calculated ROUGE-L score between {D_i} and their respective generated summary {S_i}. Then we choose S_i that has greatest score as final generated summary.' The paper does not validate this proxy against gold summaries.
domain assumption Large filtered subsets of arXiv and BigPatent with documents over 20,000 tokens are representative of very long documents in general.
Section 4.1 selects only these two datasets and filters by length; no argument or cross-dataset check supports generalization.
ad hoc to paper Splitting the target summary into the same number of parts as the document and pairing by ROUGE-L produces valid training signal (SPIN 1).
Algorithm 1 and Section 3.1.1 define the pairing; the poor SPIN 1 results in Table 3 show the assumption is questionable and the paper itself attributes it to 'unrelevant' pairs.

how reviews work

0 comments

Cite this review

Pith. "Pith review of A Split-then-Join Approach to Abstractive Summarization for Very Long Documents in a Low Resource Setting." pith.science (2026). https://pith.science/paper/6VHIEC3H

@misc{pith2026250506862,
  author       = {Pith},
  title        = {Pith review of: A Split-then-Join Approach to Abstractive Summarization for Very Long Documents in a Low Resource Setting},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6VHIEC3H}},
  note         = {Machine review of arXiv:2505.06862}
}

abstract

$\texttt{BIGBIRD-PEGASUS}$ model achieves $\textit{state-of-the-art}$ on abstractive text summarization for long documents. However it's capacity still limited to maximum of $4,096$ tokens, thus caused performance degradation on summarization for very long documents. Common method to deal with the issue is to truncate the documents. In this reasearch, we'll use different approach. We'll use the pretrained $\texttt{BIGBIRD-PEGASUS}$ model by fine tuned the model on other domain dataset. First, we filter out all documents which length less than $20,000$ tokens to focus on very long documents. To prevent domain shifting problem and overfitting on transfer learning due to small dataset, we augment the dataset by splitting document-summary training pair into parts, to fit the document into $4,096$ tokens. Source code available on $\href{https://github.com/lhfazry/SPIN-summ}{https://github.com/lhfazry/SPIN-summ}$.

Figures

Figures reproduced from arXiv: 2505.06862 by the authors.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

27 extracted references · 6 canonical work pages

[1]

Yi-Syuan Chen and Hong-Han Shuai. 2021. http://arxiv.org/abs/2102.09397 Meta- Transfer Learning for Low - Resource Abstractive Summarization . arXiv:2102.09397 [cs]. ArXiv: 2102.09397

arXiv 2021
[2]

Rewon Child, Scott Gray, Alec Radford, and Ilya Sutskever. 2019. http://arxiv.org/abs/1904.10509 Generating Long Sequences with Sparse Transformers . arXiv:1904.10509 [cs, stat]. ArXiv: 1904.10509

arXiv 2019
[3]

Sumit Chopra, Michael Auli, and Alexander M. Rush. 2016. https://doi.org/10.18653/v1/N16-1012 Abstractive Sentence Summarization with Attentive Recurrent Neural Networks . In Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics : Human Language Technologies , pages 93--98, San Diego, California....

work page doi:10.18653/v1/n16-1012 2016
[4]

Arman Cohan, Franck Dernoncourt, Doo Soon Kim, Trung Bui, Seokhwan Kim, Walter Chang, and Nazli Goharian. 2018. https://doi.org/10.18653/v1/N18-2097 A Discourse - Aware Attention Model for Abstractive Summarization of Long Documents . In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics : Hum...

doi:10.18653/v1/n18-2097 2018
[5]

Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. http://arxiv.org/abs/1810.04805 BERT : Pre -training of Deep Bidirectional Transformers for Language Understanding . arXiv:1810.04805 [cs]. ArXiv: 1810.04805

arXiv 2019
[6]

Alexander Gaskell, Dr Pedro Baiz, Lucia Specia, Hugo Barbaroux, and Dr Eric Topham. 2020. https://www.imperial.ac.uk/media/imperial-college/faculty-of-engineering/computing/public/1920-pg-projects/Technical-Writer-Assistant.pdf On the Summarization and Evaluation of Long Documents . page 98

work page 2020
[7]

Alexios Gidiotis and Grigorios Tsoumakas. 2020. https://doi.org/10.1109/TASLP.2020.3037401 A Divide -and- Conquer Approach to the Summarization of Long Documents . IEEE/ACM Transactions on Audio, Speech, and Language Processing, 28:3029--3040

work page arXiv 2020
[8]

Luyang Huang, Shuyang Cao, Nikolaus Parulian, Heng Ji, and Lu Wang. 2021. http://arxiv.org/abs/2104.02112 Efficient Attentions for Long Document Summarization . arXiv:2104.02112 [cs]. ArXiv: 2104.02112

arXiv 2021

Show all 27 references

[9]

Oleksandra Klymenko, Daniel Braun, and Florian Matthes. 2020. https://doi.org/10.5220/0009723306480655 Automatic Text Summarization : A State -of-the- Art Review : . In Proceedings of the 22nd International Conference on Enterprise Information Systems , pages 648--655, Prague,...

2020 doi
[10]

Julian Kupiec, Jan Pedersen, and Francine Chen. 1995. https://doi.org/10.1145/215206.215333 A trainable document summarizer . In Proceedings of the 18th annual international ACM SIGIR conference on Research and development in information retrieval - SIGIR '95 , pages 68--73, S...

1995
[11]

Mike Lewis, Yinhan Liu, Naman Goyal, Marjan Ghazvininejad, Abdelrahman Mohamed, Omer Levy, Ves Stoyanov, and Luke Zettlemoyer. 2019. http://arxiv.org/abs/1910.13461 BART : Denoising Sequence -to- Sequence Pre -training for Natural Language Generation , Translation , and Compre...

2019 arXiv
[12]

Ramesh Nallapati, Bowen Zhou, Cicero Nogueira dos santos, Caglar Gulcehre, and Bing Xiang. 2016. http://arxiv.org/abs/1602.06023 Abstractive Text Summarization Using Sequence -to- Sequence RNNs and Beyond . arXiv:1602.06023 [cs]. ArXiv: 1602.06023

2016 arXiv
[13]

Ani Nenkova. 2011. https://doi.org/10.1561/1500000015 Automatic Summarization . Foundations and Trends® in Information Retrieval, 5(2):103--233

2011 doi
[14]

Improving Language Understanding by Generative Pre - Training

Alec Radford, Karthik Narasimhan, Tim Salimans, and Ilya Sutskever. Improving Language Understanding by Generative Pre - Training . page 12
[15]

Pranav Rajpurkar, Jian Zhang, Konstantin Lopyrev, and Percy Liang. 2016. http://arxiv.org/abs/1606.05250 SQuAD : 100,000+ Questions for Machine Comprehension of Text . arXiv:1606.05250 [cs]. ArXiv: 1606.05250

2016 arXiv
[16]

Rush, Sumit Chopra, and Jason Weston

Alexander M. Rush, Sumit Chopra, and Jason Weston. 2015. https://doi.org/10.18653/v1/D15-1044 A Neural Attention Model for Abstractive Sentence Summarization . In Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing , pages 379--389, Lisbon, P...

2015 doi
[17]

Horacio Saggion and Thierry Poibeau. 2013. https://doi.org/10.1007/978-3-642-28569-1_1 Automatic Text Summarization : Past , Present and Future . In Thierry Poibeau, Horacio Saggion, Jakub Piskorski, and Roman Yangarber, editors, Multi-source, Multilingual Information Extracti...

2013 doi
[18]

Liu, and Christopher D

Abigail See, Peter J. Liu, and Christopher D. Manning. 2017. https://doi.org/10.18653/v1/P17-1099 Get To The Point : Summarization with Pointer - Generator Networks . In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics ( Volume 1: Long Pa...

2017 doi
[19]

Eva Sharma, Chen Li, and Lu Wang. 2019. http://arxiv.org/abs/1906.03741 BIGPATENT : A Large - Scale Dataset for Abstractive and Coherent Summarization . arXiv:1906.03741 [cs]. ArXiv: 1906.03741

2019 arXiv
[20]

Ilya Sutskever, Oriol Vinyals, and Quoc V. Le. 2014. http://arxiv.org/abs/1409.3215 Sequence to Sequence Learning with Neural Networks . arXiv:1409.3215 [cs]. ArXiv: 1409.3215

2014 arXiv
[21]

Gomez, Lukasz Kaiser, and Illia Polosukhin

Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. 2017. http://arxiv.org/abs/1706.03762 Attention Is All You Need . arXiv:1706.03762 [cs]. ArXiv: 1706.03762

2017 arXiv
[22]

Alex Wang, Amanpreet Singh, Julian Michael, Felix Hill, Omer Levy, and Samuel R. Bowman. 2019. http://arxiv.org/abs/1804.07461 GLUE : A Multi - Task Benchmark and Analysis Platform for Natural Language Understanding . arXiv:1804.07461 [cs]. ArXiv: 1804.07461

2019 arXiv
[23]

Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Clement Delangue, Anthony Moi, Pierric Cistac, Tim Rault, Rémi 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 arXiv
[24]

Manzil Zaheer, Guru Guruganesh, Avinava Dubey, Joshua Ainslie, Chris Alberti, Santiago Ontanon, Philip Pham, Anirudh Ravula, Qifan Wang, Li Yang, and Amr Ahmed. 2021. http://arxiv.org/abs/2007.14062 Big Bird : Transformers for Longer Sequences . arXiv:2007.14062 [cs, stat]. Ar...

2021 arXiv
[25]

Wenyuan Zeng, Wenjie Luo, Sanja Fidler, and Raquel Urtasun. 2016. http://arxiv.org/abs/1611.03382 Efficient Summarization with Read - Again and Copy Mechanism . arXiv:1611.03382 [cs]. ArXiv: 1611.03382

2016 arXiv
[26]

Jingqing Zhang, Yao Zhao, Mohammad Saleh, and Peter J. Liu. 2020. http://arxiv.org/abs/1912.08777 PEGASUS : Pre -training with Extracted Gap -sentences for Abstractive Summarization . arXiv:1912.08777 [cs]. ArXiv: 1912.08777

2020 arXiv
[27]

Chenguang Zhu, William Hinthorn, Ruochen Xu, Qingkai Zeng, Michael Zeng, Xuedong Huang, and Meng Jiang. 2021. https://doi.org/10.18653/v1/2021.naacl-main.58 Enhancing Factual Consistency of Abstractive Summarization . In Proceedings of the 2021 Conference of the North American...

2021 doi

Pith tools

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

[1] [1]

Yi-Syuan Chen and Hong-Han Shuai. 2021. http://arxiv.org/abs/2102.09397 Meta- Transfer Learning for Low - Resource Abstractive Summarization . arXiv:2102.09397 [cs]. ArXiv: 2102.09397

arXiv 2021

[2] [2]

Rewon Child, Scott Gray, Alec Radford, and Ilya Sutskever. 2019. http://arxiv.org/abs/1904.10509 Generating Long Sequences with Sparse Transformers . arXiv:1904.10509 [cs, stat]. ArXiv: 1904.10509

arXiv 2019

[3] [3]

Sumit Chopra, Michael Auli, and Alexander M. Rush. 2016. https://doi.org/10.18653/v1/N16-1012 Abstractive Sentence Summarization with Attentive Recurrent Neural Networks . In Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics : Human Language Technologies , pages 93--98, San Diego, California....

work page doi:10.18653/v1/n16-1012 2016

[4] [4]

Arman Cohan, Franck Dernoncourt, Doo Soon Kim, Trung Bui, Seokhwan Kim, Walter Chang, and Nazli Goharian. 2018. https://doi.org/10.18653/v1/N18-2097 A Discourse - Aware Attention Model for Abstractive Summarization of Long Documents . In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics : Hum...

doi:10.18653/v1/n18-2097 2018

[5] [5]

Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. http://arxiv.org/abs/1810.04805 BERT : Pre -training of Deep Bidirectional Transformers for Language Understanding . arXiv:1810.04805 [cs]. ArXiv: 1810.04805

arXiv 2019

[6] [6]

Alexander Gaskell, Dr Pedro Baiz, Lucia Specia, Hugo Barbaroux, and Dr Eric Topham. 2020. https://www.imperial.ac.uk/media/imperial-college/faculty-of-engineering/computing/public/1920-pg-projects/Technical-Writer-Assistant.pdf On the Summarization and Evaluation of Long Documents . page 98

work page 2020

[7] [7]

Alexios Gidiotis and Grigorios Tsoumakas. 2020. https://doi.org/10.1109/TASLP.2020.3037401 A Divide -and- Conquer Approach to the Summarization of Long Documents . IEEE/ACM Transactions on Audio, Speech, and Language Processing, 28:3029--3040

work page arXiv 2020

[8] [8]

Luyang Huang, Shuyang Cao, Nikolaus Parulian, Heng Ji, and Lu Wang. 2021. http://arxiv.org/abs/2104.02112 Efficient Attentions for Long Document Summarization . arXiv:2104.02112 [cs]. ArXiv: 2104.02112

arXiv 2021