pith. sign in

arxiv: 2606.18946 · v1 · pith:SKOPQOQQnew · submitted 2026-06-17 · 💻 cs.CL

SenFlow: Inter-Sentence Flow Modeling for AI-Generated Text Detection in Hybrid Documents

Jingkun Luo , Yifan Sun , Da-Tian Peng , Guanxiong Pei This is my paper

Pith reviewed 2026-06-26 20:46 UTC · model grok-4.3

classification 💻 cs.CL
keywords AI-generated text detectionhybrid documentssentence-level detectiongraph propagationconditional random fieldMOSAIC benchmarkcross-domain transferstructured prediction
0
0 comments X

The pith

Inter-sentence graph modeling improves detection of AI text in hybrid documents over independent sentence classification.

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

Existing methods classify each sentence in isolation when spotting AI-generated text inside mixed human-LLM documents, ignoring how one sentence relates to the next. This paper builds the MOSAIC benchmark of 16,000 hybrid documents using recent generators and a perplexity-consistency filter, then introduces SenFlow to treat the task as structured prediction over the full sentence sequence. SenFlow combines graph propagation of information between sentences with linear-chain CRF decoding in one document-level pass. A sympathetic reader would care because hybrid documents are now common and isolated classifiers miss subtle, context-dependent insertions. If the claim holds, detection systems gain accuracy by explicitly modeling sentence-to-sentence flow.

Core claim

The paper claims that recasting sentence-level AI-generated text detection as structured prediction over the document sentence sequence, instantiated as SenFlow with graph-based inter-sentence propagation and linear-chain CRF decoding, reaches state-of-the-art performance on the MOSAIC benchmark, with a 4.15 percentage point average Macro-F1 margin on cross-domain transfer.

What carries the argument

SenFlow, which integrates graph-based propagation of inter-sentence information with linear-chain CRF decoding over a sentence graph to capture dependencies across the document.

If this is right

  • Modeling dependencies between sentences supplies additional signal that isolated classifiers miss.
  • Even after the perplexity filter removes some overt cues, generator-dependent sentence-length patterns remain and aid detection.
  • The largest gains appear on cross-domain transfer, the hardest of the three evaluation protocols.
  • The full model runs in a single document-level pass rather than sentence-by-sentence.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same graph-plus-CRF structure could apply to other sequence-labeling problems where units are edited or anomalous relative to their neighbors.
  • The persistent length gap finding suggests length normalization as a cheap baseline feature worth testing on future detectors.
  • Evaluating the method on documents that humans have lightly edited after AI generation would test robustness to realistic hybrid use.

Load-bearing premise

Inter-sentence dependencies supply detection signal beyond what independent per-sentence classifiers already capture.

What would settle it

Compare SenFlow against a strong independent sentence classifier on the same test documents after randomly shuffling sentence order inside each document; if the performance gap disappears, the inter-sentence modeling claim is falsified.

Figures

Figures reproduced from arXiv: 2606.18946 by Da-Tian Peng, Guanxiong Pei, Jingkun Luo, Yifan Sun.

Figure 1
Figure 1. Figure 1: S-AGTD is hard in isolation but tractable with context. Sentence s2 is ambiguous alone; its sur￾rounding context reveals authorship through the local clustering of AI-generated sentences. are interleaved within a single piece of text. Hybrid writing is increasingly prevalent in practice (Lee et al., 2022; Zhang et al., 2024; Su et al., 2025): a student may polish paragraphs, a researcher may insert AI-gene… view at source ↗
Figure 2
Figure 2. Figure 2: SenFlow architecture. Four per-sentence streams from a frozen LoRA-aligned Llama-3.1-8B-Instruct are fused by dual cross-attention, refined by the Hybrid Adjacency GCN, and decoded by a linear-chain CRF. Solid edges denote positional adjacency Apos, dashed edges semantic Asem; blue and amber mark human and AI predictions. above make MOSAIC substantially harder than prior S-AGTD benchmarks. Multi-block repl… view at source ↗
Figure 3
Figure 3. Figure 3: The semantic branch encodes class-aligned long-range structure that local positional adjacency cannot capture. Hybrid adjacency for a Pub→XSum cross-domain test document (N = 13); ground-truth labels below each panel, blue human, amber AI. Positional adjacency Apos (left) is banded and local; semantic adjacency Asem (middle) is near-perfectly class-aligned with long-range edges; the learned hybrid (right) … view at source ↗
Figure 4
Figure 4. Figure 4: Counter to intuition, DeepSeek-V3.2 inser￾tions are easier to detect than Kimi K2 insertions, consistently across all four training-based methods. Macro-F1 on the DeepSeek-V3.2 (reasoning) and Kimi K2 (chat) subsets under the Unified protocol, averaged across PubMed and XSum. Parenthetical values report the pp drop from DeepSeek-V3.2 to Kimi K2. A consistent pattern across all training-based methods is tha… view at source ↗
read the original abstract

Sentence-level AI-generated text detection (S-AGTD) for hybrid documents, where humans and LLMs co-author one text, faces two gaps: existing methods classify each sentence in isolation, discarding inter-sentence dependencies, and existing benchmarks omit the newest generation of generators. We construct MOSAIC, a benchmark of 16,000 hybrid documents over PubMed and XSum, generated by DeepSeek-V3.2 and Kimi K2 under stringent quality controls including a perplexity-consistency filter absent from prior benchmarks. We recast S-AGTD as structured prediction over the document sentence sequence and instantiate it as SenFlow, integrating graph-based inter-sentence propagation with linear-chain CRF decoding in a single document-level pass over a sentence graph. SenFlow reaches state-of-the-art performance on MOSAIC, with a +4.15 pp average Macro-F1 margin on cross-domain transfer, the hardest of three protocols of increasing difficulty. We further find that even after the perplexity filter equalizes overt cues, AI insertions retain a generator-dependent sentence-length gap that sentence-level detectors still exploit. Code and data: https://github.com/luojingkun22/SenFlow

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The manuscript introduces MOSAIC, a benchmark of 16,000 hybrid documents from PubMed and XSum generated by DeepSeek-V3.2 and Kimi K2 under quality controls including a perplexity-consistency filter. It recasts sentence-level AI-generated text detection as structured prediction and proposes SenFlow, which integrates graph-based inter-sentence propagation with linear-chain CRF decoding. The paper reports state-of-the-art results on MOSAIC, including a +4.15 pp average Macro-F1 margin on the cross-domain transfer protocol, the hardest of three evaluation settings.

Significance. If the reported gains are shown to arise from the inter-sentence modeling rather than benchmark construction artifacts, the work would establish the utility of graph propagation and CRF decoding for hybrid-document detection tasks. The release of code and data, together with the new benchmark addressing gaps in prior evaluations, would constitute a concrete contribution to the field.

major comments (2)
  1. [Abstract] The headline claim that inter-sentence flow modeling supplies the observed +4.15 pp Macro-F1 margin on the cross-domain protocol is not supported by any ablation that removes the graph edges while retaining the identical sentence encoder, CRF, and MOSAIC splits. Without this isolation, it remains possible that the gains trace to sentence-level cues (including the residual length gap noted in the abstract) rather than the graph propagation component.
  2. [Abstract] The assertion that the perplexity-consistency filter produces an unbiased benchmark representative of real hybrid documents is undercut by the paper's own observation of a generator-dependent sentence-length gap that any sentence-level detector can exploit; this directly affects whether the cross-domain gains can be attributed to flow modeling.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments highlighting the need for clearer isolation of the inter-sentence modeling contribution and for more precise language on benchmark construction. We address each major comment below.

read point-by-point responses

  1. Referee: [Abstract] The headline claim that inter-sentence flow modeling supplies the observed +4.15 pp Macro-F1 margin on the cross-domain protocol is not supported by any ablation that removes the graph edges while retaining the identical sentence encoder, CRF, and MOSAIC splits. Without this isolation, it remains possible that the gains trace to sentence-level cues (including the residual length gap noted in the abstract) rather than the graph propagation component.

    Authors: We agree that the manuscript lacks a direct ablation removing only the graph edges while retaining the identical sentence encoder, CRF decoder, and MOSAIC data splits. The reported comparisons are against prior sentence-level baselines rather than an internal SenFlow variant without propagation. In the revision we will add this ablation to quantify the incremental benefit of the graph component on the cross-domain protocol. revision: yes

  2. Referee: [Abstract] The assertion that the perplexity-consistency filter produces an unbiased benchmark representative of real hybrid documents is undercut by the paper's own observation of a generator-dependent sentence-length gap that any sentence-level detector can exploit; this directly affects whether the cross-domain gains can be attributed to flow modeling.

    Authors: The manuscript already notes the residual generator-dependent length gap after the perplexity filter. We accept that this observation weakens the claim of an unbiased benchmark. In the revision we will revise the abstract to state that the filter equalizes overt perplexity cues but leaves other distributional differences, and we will qualify the attribution of cross-domain gains accordingly while retaining the description of the filter's purpose. revision: yes

Circularity Check

0 steps flagged

No circularity; empirical application of standard graph/CRF components to new benchmark.

full rationale

The paper constructs MOSAIC benchmark under explicit quality controls (perplexity-consistency filter) and applies established graph propagation plus linear-chain CRF to recast sentence-level detection as structured prediction. The +4.15 pp Macro-F1 claim is an empirical measurement on held-out cross-domain splits, not a quantity derived by construction from fitted parameters or self-referential definitions. No equations, self-citations, or ansatzes are presented that reduce the central result to its inputs. The derivation chain is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Only the abstract is available, so specific free parameters, ad-hoc axioms, or invented entities cannot be enumerated; the approach appears to rest on standard graph neural network and CRF assumptions from the broader literature.

axioms (1)
  • standard math Graph-based message passing and linear-chain CRF can be combined for structured sequence labeling
    The method description invokes these established components without deriving them.

pith-pipeline@v0.9.1-grok · 5748 in / 1276 out tokens · 30115 ms · 2026-06-26T20:46:15.824744+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

75 extracted references · 14 canonical work pages · 11 internal anchors

  1. [2]

    Guangsheng Bao, Yanbin Zhao, Zhiyang Teng, Linyi Yang, and Yue Zhang. 2024. Fast- D etect GPT : Efficient zero-shot detection of machine-generated text via conditional probability curvature. In International Conference on Learning Representations

    2024

  2. [3]

    Steven Bird, Ewan Klein, and Edward Loper. 2009. Natural language processing with P ython: analyzing text with the natural language toolkit . O'Reilly Media, Inc

    2009

  3. [4]

    Arman Cohan, Franck Dernoncourt, Doo Soon Kim, Trung Bui, Seokhwan Kim, Walter Chang, and Nazli Goharian. 2018. 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: Human Language Technologies, volume 2, pages...

    2018

  4. [5]

    Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Alex Vaughan, and 1 others. 2024. The L lama 3 herd of models. arXiv preprint arXiv:2407.21783

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  5. [6]

    Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Peiyi Wang, Qihao Zhu, Runxin Xu, Ruoyu Zhang, Shirong Ma, Xiao Bi, and 1 others. 2025. D eep S eek- R 1: Incentivizing reasoning capability in LLM s via reinforcement learning. arXiv preprint arXiv:2501.12948

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  6. [7]

    Abhimanyu Hans, Avi Schwarzschild, Valeriia Cherepanova, Hamid Kazemi, Aniruddha Saha, Micah Goldblum, Jonas Geiping, and Tom Goldstein. 2024. Spotting LLM s with binoculars: Zero-shot detection of machine-generated text. In Proceedings of the 41st International Conference on Machine Learning, pages 17519--17537

    2024

  7. [8]

    Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Liang Wang, Weizhu Chen, and 1 others. 2022. Lo RA : Low-rank adaptation of large language models. In International Conference on Learning Representations

    2022

  8. [9]

    Xiaomeng Hu, Pin-Yu Chen, and Tsung-Yi Ho. 2023. RADAR : Robust AI -text detection via adversarial learning. Advances in Neural Information Processing Systems, 36:15077--15095

    2023

  9. [10]

    Guanhua Huang, Yuchen Zhang, Zhe Li, Yongjian You, Mingze Wang, and Zhouwang Yang. 2024. Are AI -generated text detectors robust to adversarial perturbations? In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics, volume 1, pages 6005--6024

    2024

  10. [11]

    Aaron Jaech, Adam Kalai, Adam Lerer, Adam Richardson, Ahmed El-Kishky, Aiden Low, Alec Helyar, Aleksander Madry, Alex Beutel, Alex Carney, and 1 others. 2024. O pen AI o1 system card. arXiv preprint arXiv:2412.16720

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  11. [12]

    Lei Jiang, Desheng Wu, and Xiaolong Zheng. 2025. SenDetEX : Sentence-level AI -generated text detection for human- AI hybrid content via style and context fusion. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 5287--5302

    2025

  12. [14]

    John Kirchenbauer, Jonas Geiping, Yuxin Wen, Jonathan Katz, Ian Miers, and Tom Goldstein. 2023. A watermark for large language models. In International Conference on Machine Learning, pages 17061--17084. PMLR

    2023

  13. [15]

    John Lafferty, Andrew McCallum, and Fernando CN Pereira. 2001. Conditional random fields: Probabilistic models for segmenting and labeling sequence data. In Proceedings of the Eighteenth International Conference on Machine Learning, pages 282--289

    2001

  14. [16]

    Mina Lee, Percy Liang, and Qian Yang. 2022. C o A uthor: Designing a human- AI collaborative writing dataset for exploring language model capabilities. In Proceedings of the 2022 CHI Conference on Human Factors in Computing Systems, pages 1--19

    2022

  15. [17]

    Yafu Li, Qintong Li, Leyang Cui, Wei Bi, Zhilin Wang, Longyue Wang, Linyi Yang, Shuming Shi, and Yue Zhang. 2024. MAGE : Machine-generated text detection in the wild. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics, volume 1, pages 36--53

    2024

  16. [18]

    Tsung-Yi Lin, Priya Goyal, Ross Girshick, Kaiming He, and Piotr Doll \'a r. 2017. Focal loss for dense object detection. In Proceedings of the IEEE International Conference on Computer Vision, pages 2980--2988

    2017

  17. [19]

    Ilya Loshchilov and Frank Hutter. 2016. SGDR : Stochastic gradient descent with warm restarts. arXiv preprint arXiv:1608.03983

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  18. [21]

    Chengzhi Mao, Carl Vondrick, Hao Wang, and Junfeng Yang. 2024. RAIDAR : generative AI detection via rewriting. arXiv preprint arXiv:2401.12970

    work page arXiv 2024

  19. [22]

    Eric Mitchell, Yoonho Lee, Alexander Khazatsky, Christopher D Manning, and Chelsea Finn. 2023. D etect GPT : Zero-shot machine-generated text detection using probability curvature. In International Conference on Machine Learning, pages 24950--24962. PMLR

    2023

  20. [23]

    Shashi Narayan, Shay B Cohen, and Mirella Lapata. 2018. Don't give me the details, just the summary! topic-aware convolutional neural networks for extreme summarization. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 1797--1807

    2018

  21. [24]

    Hoang-Quoc Nguyen-Son, Minh-Son Dao, and Koji Zettsu. 2024. S im LLM : Detecting sentences generated by large language models using similarity between the generation and its re-generation. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 22340--22352

    2024

  22. [26]

    Nils Reimers and Iryna Gurevych. 2019. Sentence- BERT : Sentence embeddings using siamese BERT -networks. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, pages 3982--3992

    2019

  23. [27]

    Vinu Sankar Sadasivan, Aounon Kumar, Sriram Balasubramanian, Wenxiao Wang, and Soheil Feizi. 2023. Can AI -generated text be reliably detected? arXiv preprint arXiv:2303.11156

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  24. [28]

    Sunil Kumar Sahu, Fenia Christopoulou, Makoto Miwa, and Sophia Ananiadou. 2019. Inter-sentence relation extraction with document-level graph convolutional neural network. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 4309--4316

    2019

  25. [29]

    Yuhui Shi, Qiang Sheng, Juan Cao, Hao Mi, Beizhe Hu, and Danding Wang. 2024. Ten words only still help: Improving black-box AI -generated text detection via proxy-guided efficient re-sampling. In Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence, pages 494--502

    2024

  26. [31]

    Zhixiong Su, Yichen Wang, Herun Wan, Zhaohan Zhang, and Minnan Luo. 2025. HACo - D et: A study towards fine-grained machine-generated text detection under human- AI coauthoring. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics, volume 1, pages 22015--22036

    2025

  27. [32]

    Christian Szegedy, Vincent Vanhoucke, Sergey Ioffe, Jon Shlens, and Zbigniew Wojna. 2016. Rethinking the inception architecture for computer vision. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 2818--2826

    2016

  28. [33]

    Pengyu Wang, Linyang Li, Ke Ren, Botian Jiang, Dong Zhang, and Xipeng Qiu. 2023. SeqXGPT : Sentence-level AI -generated text detection. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 1144--1156

    2023

  29. [34]

    Debora Weber-Wulff, Alla Anohina-Naumeca, Sonja Bjelobaba, Tom \'a s Folt \`y nek, Jean Guerrero-Dib, Olumide Popoola, Petr S igut, and Lorna Waddington. 2023. Testing of detection tools for AI -generated text. International Journal for Educational Integrity, 19(1):1--39

    2023

  30. [35]

    Junchao Wu, Shu Yang, Runzhe Zhan, Yulin Yuan, Lidia Sam Chao, and Derek Fai Wong. 2025. A survey on LLM -generated text detection: Necessity, methods, and future directions. Computational Linguistics, 51(1):275--338

    2025

  31. [36]

    Xianjun Yang, Wei Cheng, Yue Wu, Linda Petzold, William Wang, and Haifeng Chen. 2024. DNA - GPT : Divergent n-gram analysis for training-free detection of GPT -generated text. In International Conference on Learning Representations

    2024

  32. [38]

    Cong Zeng, Shengkun Tang, Xianjun Yang, Yuanzhou Chen, Yiyou Sun, Zhiqiang Xu, Yao Li, Haifeng Chen, Wei Cheng, and Dongkuan Xu. 2024. DALD : Improving logits-based detector without logits from black-box LLM s. Advances in Neural Information Processing Systems, 37:54947--54973

    2024

  33. [39]

    Qihui Zhang, Chujie Gao, Dongping Chen, Yue Huang, Yixin Huang, Zhenyang Sun, Shilin Zhang, Weiye Li, Zhengyan Fu, Yao Wan, and 1 others. 2024. LLM -as-a-coauthor: Can mixed human-written and machine-generated text be detected? In Findings of the Association for Computational Linguistics: NAACL 2024, pages 409--436

    2024

  34. [41]

    Xuandong Zhao, Prabhanjan Ananth, Lei Li, and Yu-Xiang Wang. 2023 b . Provable robust watermarking for AI -generated text. arXiv preprint arXiv:2306.17439

    work page arXiv 2023

  35. [42]

    A Survey of Large Language Models

    A survey of large language models , author=. arXiv preprint arXiv:2303.18223 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  36. [43]

    A survey on

    Wu, Junchao and Yang, Shu and Zhan, Runzhe and Yuan, Yulin and Chao, Lidia Sam and Wong, Derek Fai , journal=. A survey on

  37. [44]

    2023 , organization=

    Mitchell, Eric and Lee, Yoonho and Khazatsky, Alexander and Manning, Christopher D and Finn, Chelsea , booktitle=. 2023 , organization=

    2023

  38. [45]

    Bao, Guangsheng and Zhao, Yanbin and Teng, Zhiyang and Yang, Linyi and Zhang, Yue , booktitle=. Fast-

  39. [46]

    Yang, Xianjun and Cheng, Wei and Wu, Yue and Petzold, Linda and Wang, William and Chen, Haifeng , booktitle=

  40. [47]

    International Conference on Machine Learning , pages=

    A watermark for large language models , author=. International Conference on Machine Learning , pages=. 2023 , organization=

    2023

  41. [48]

    Jiang, Lei and Wu, Desheng and Zheng, Xiaolong , booktitle=

  42. [49]

    Wang, Pengyu and Li, Linyang and Ren, Ke and Jiang, Botian and Zhang, Dong and Qiu, Xipeng , booktitle=

  43. [50]

    Spotting

    Hans, Abhimanyu and Schwarzschild, Avi and Cherepanova, Valeriia and Kazemi, Hamid and Saha, Aniruddha and Goldblum, Micah and Geiping, Jonas and Goldstein, Tom , booktitle=. Spotting

  44. [51]

    Release Strategies and the Social Impacts of Language Models

    Release strategies and the social impacts of language models , author=. arXiv preprint arXiv:1908.09203 , year=

    work page internal anchor Pith review Pith/arXiv arXiv 1908

  45. [52]

    Hu, Xiaomeng and Chen, Pin-Yu and Ho, Tsung-Yi , journal=

  46. [53]

    Zeng, Cong and Tang, Shengkun and Yang, Xianjun and Chen, Yuanzhou and Sun, Yiyou and Xu, Zhiqiang and Li, Yao and Chen, Haifeng and Cheng, Wei and Xu, Dongkuan , journal=

  47. [54]

    Ten words only still help: Improving black-box

    Shi, Yuhui and Sheng, Qiang and Cao, Juan and Mi, Hao and Hu, Beizhe and Wang, Danding , booktitle=. Ten words only still help: Improving black-box

  48. [55]

    Jaech, Aaron and Kalai, Adam and Lerer, Adam and Richardson, Adam and El-Kishky, Ahmed and Low, Aiden and Helyar, Alec and Madry, Aleksander and Beutel, Alex and Carney, Alex and others , journal=

  49. [56]

    Guo, Daya and Yang, Dejian and Zhang, Haowei and Song, Junxiao and Wang, Peiyi and Zhu, Qihao and Xu, Runxin and Zhang, Ruoyu and Ma, Shirong and Bi, Xiao and others , journal=

  50. [57]

    Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies , pages=

    A discourse-aware attention model for abstractive summarization of long documents , author=. Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies , pages=

    2018

  51. [58]

    Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing , pages=

    Don't give me the details, just the summary! topic-aware convolutional neural networks for extreme summarization , author=. Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing , pages=

    2018

  52. [59]

    Natural language processing with

    Bird, Steven and Klein, Ewan and Loper, Edward , year=. Natural language processing with

  53. [60]

    Grattafiori, Aaron and Dubey, Abhimanyu and Jauhri, Abhinav and Pandey, Abhinav and Kadian, Abhishek and Al-Dahle, Ahmad and Letman, Aiesha and Mathur, Akhil and Schelten, Alan and Vaughan, Alex and others , journal=. The

  54. [61]

    Hu, Edward J and Shen, Yelong and Wallis, Phillip and Allen-Zhu, Zeyuan and Li, Yuanzhi and Wang, Shean and Wang, Liang and Chen, Weizhu and others , booktitle=. Lo

  55. [62]

    An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling

    An empirical evaluation of generic convolutional and recurrent networks for sequence modeling , author=. arXiv preprint arXiv:1803.01271 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  56. [63]

    Semi-Supervised Classification with Graph Convolutional Networks

    Semi-supervised classification with graph convolutional networks , author=. arXiv preprint arXiv:1609.02907 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  57. [64]

    Proceedings of the Eighteenth International Conference on Machine Learning , pages=

    Conditional random fields: Probabilistic models for segmenting and labeling sequence data , author=. Proceedings of the Eighteenth International Conference on Machine Learning , pages=

  58. [65]

    Proceedings of the IEEE International Conference on Computer Vision , pages=

    Focal loss for dense object detection , author=. Proceedings of the IEEE International Conference on Computer Vision , pages=

  59. [66]

    Representation Learning with Contrastive Predictive Coding

    Representation learning with contrastive predictive coding , author=. arXiv preprint arXiv:1807.03748 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  60. [67]

    Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , pages=

    Rethinking the inception architecture for computer vision , author=. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , pages=

  61. [68]

    Decoupled Weight Decay Regularization

    Decoupled weight decay regularization , author=. arXiv preprint arXiv:1711.05101 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  62. [69]

    Loshchilov, Ilya and Hutter, Frank , journal=

  63. [70]

    Lee, Mina and Liang, Percy and Yang, Qian , booktitle=

  64. [71]

    Sadasivan, Vinu Sankar and Kumar, Aounon and Balasubramanian, Sriram and Wang, Wenxiao and Feizi, Soheil , journal=. Can

  65. [72]

    Testing of detection tools for

    Weber-Wulff, Debora and Anohina-Naumeca, Alla and Bjelobaba, Sonja and Folt. Testing of detection tools for. International Journal for Educational Integrity , volume=. 2023 , publisher=

    2023

  66. [73]

    Huang, Guanhua and Zhang, Yuchen and Li, Zhe and You, Yongjian and Wang, Mingze and Yang, Zhouwang , booktitle=. Are

  67. [74]

    Li, Yafu and Li, Qintong and Cui, Leyang and Bi, Wei and Wang, Zhilin and Wang, Longyue and Yang, Linyi and Shi, Shuming and Zhang, Yue , booktitle=

  68. [75]

    Mao, Chengzhi and Vondrick, Carl and Wang, Hao and Yang, Junfeng , journal=

  69. [76]

    Nguyen-Son, Hoang-Quoc and Dao, Minh-Son and Zettsu, Koji , booktitle=

  70. [77]

    Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics , pages=

    Inter-sentence relation extraction with document-level graph convolutional neural network , author=. Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics , pages=

  71. [78]

    Su, Zhixiong and Wang, Yichen and Wan, Herun and Zhang, Zhaohan and Luo, Minnan , booktitle=

  72. [79]

    arXiv preprint arXiv:1912.07225 , year=

    Graph-based neural sentence ordering , author=. arXiv preprint arXiv:1912.07225 , year=

    work page arXiv 1912

  73. [80]

    Zhang, Qihui and Gao, Chujie and Chen, Dongping and Huang, Yue and Huang, Yixin and Sun, Zhenyang and Zhang, Shilin and Li, Weiye and Fu, Zhengyan and Wan, Yao and others , booktitle=

  74. [81]

    Provable robust watermarking for

    Zhao, Xuandong and Ananth, Prabhanjan and Li, Lei and Wang, Yu-Xiang , journal=. Provable robust watermarking for

  75. [82]

    Sentence-

    Reimers, Nils and Gurevych, Iryna , booktitle=. Sentence-