Pith. sign in

REVIEW 4 major objections 5 minor 4 cited by

Generating Personalized Recipes from Historical User Preferences

T0 review · 4 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read Conditioning recipe generation on a user's past reviews produces recipes judged more plausible and more personal than generation without user history.

desk verdict A useful new-task paper with a reusable dataset, but the personalization claim is softer than the abstract suggests: the UMA metric is a likelihood-ranking proxy, and the human evaluation shows plausibility, not personalized fit. read the letter →

arxiv 1909.00105 v1 pith:NQ5WJQWH submitted 2019-08-31 cs.CL cs.AIcs.LG

Bodhisattwa Prasad Majumder , Shuyang Li , Jianmo Ni , Julian McAuley This is my paper · ORCID
classification cs.CLcs.AIcs.LG
keywords personalizedrecipegenerationdata-to-textuserpreferencemodelingattentionfusionencoder-decodercoherencematchingaccuracyincompletespecifications
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 proposes personalized recipe generation: turning a dish name, a few key ingredients, and a calorie level into full cooking instructions adapted to a specific user. Its central claim is that attending to a user's history of reviewed recipes is enough to make the output more plausible, more diverse, and more aligned with that user than a generic encoder-decoder recipe generator. To support this claim, the paper contributes a new dataset of over 180K recipes and 700K user reviews, automatic metrics for recipe coherence and personalization, and a pairwise human evaluation in which the personalized outputs were preferred over the non-personalized baseline 63% of the time.

What carries the argument

The central mechanism is the attention fusion layer, which at each decoding step concatenates the ingredient context, the prior-recipe or prior-technique context, the decoder hidden state, and the previous token embedding, then applies a ReLU projection and softmax to produce the next-token distribution. Prior-recipe attention averages over the user's k=20 most recent reviewed recipes, represented either by recipe id or by the mean of the recipe name's token embeddings, while prior-technique attention adds a user-technique co-occurrence prior to the attention score. This is what injects user history into the generation process and what distinguishes the three personalized variants: Prior Recipe, Prior Name, and Prior Tech.

What would settle it

Run a blind comparison in which the same dish is generated once from the user's true review history and once from the history of ten other users with similar culinary tastes, then ask the user which recipe better fits their preference; if the true-history recipe is not chosen significantly above chance, the apparent personalization is mostly genre or cuisine matching.

Watch

Extended reading notes

Core claim

The paper's central discovery is that conditioning recipe generation on a user's historical recipe interactions improves the plausibility, diversity, and personalization of the generated instructions relative to a non-personalized encoder-decoder. A user is represented by the set of recipes they have reviewed, by the names of those recipes, or by the cooking techniques occurring in them; at each decoding step the model attends over the user's k most recent reviewed recipes and fuses that context with ingredient attention and decoder state. In user-matching tests, the variant attending over prior recipe names ranks the correct user first in about half the cases and best in mean reciprocal rank, and human evaluators preferred the personalized outputs to the non-personalized baseline in 63% of pairwise comparisons.

Load-bearing premise

The load-bearing assumption is that what a person reviewed in the past reliably predicts what they want in a new dish, even when their review history contains as few as four items.

Editorial extensions

If this is right

  • Users with as little as four prior reviews can receive a plausible recipe from a dish name and a few ingredients; the model works even though half of the users have six or fewer reviews.
  • The most effective personalization signal is the names of previously reviewed recipes, not the recipes' ids or their cooking techniques.
  • Personalized generation increases recipe diversity and explicit ingredient mentions without sacrificing coherence, as measured by step-order scoring and step entailment.
  • The dataset and evaluation protocol, including sequential leave-one-out splits, user-matching accuracy, and coherence scorers, can serve as a benchmark for incomplete-specification recipe generation.
  • Human preference for personalized recipes, 63% over baseline, indicates that user history can steer instructional text generation beyond simple template filling.

Reading between the lines

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

  • If review history is a reliable taste signal, the same attention-fusion idea should transfer to other content-to-text settings where the "user" is defined by feedback, such as shopping lists, workout plans, or medication instructions.
  • The user-matching experiment could be sharpened by testing users whose histories overlap in cuisine: if a model conditioned on a different but cuisine-similar user scores almost as well, the personalization is largely genre matching rather than per-person taste.
  • A simple ablation anyone can run is to pass the k recent recipes in shuffled or reversed order; if performance holds, the model is using the set of prior recipes rather than their temporal drift, which would simplify the memory mechanism.
  • The coherence metrics introduced here, step-order scoring and step entailment, could be applied to other instructional prose, since they do not depend on recipe-specific knowledge.
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 introduces the task of personalized recipe generation: given a recipe name, a few ingredients, a calorie level, and a user's historical recipe interactions, the model generates full recipe instructions tailored to that user. The proposed architecture encodes the recipe name and ingredients with BiGRUs, applies ingredient attention during decoding, and incorporates user information either by attending over the k most recent prior recipes, prior recipe names, or a technique-preference vector; an attention fusion layer combines these contexts before the final output softmax. The authors release a Food.com dataset of 180K recipes and 700K interactions and evaluate against a nearest-neighbor baseline and an encoder-decoder baseline, reporting perplexity, BLEU/ROUGE, distinctness, user matching accuracy (UMA), mean reciprocal rank (MRR), automatic coherence and entailment scores, and pairwise human preference. They report that personalized models beat the baselines on perplexity, diversity, UMA/MRR, coherence, and a 63% human pairwise preference rate.

Significance. If the claims hold, this is a useful new task connecting natural language generation and recommender systems, with a novel dataset and a clean decomposition of user modeling into recipe-level, name-level, and technique-level signals. The public release of code and data is a strength, as are the automatic coherence metrics and the UMA/MRR personalization probes. However, the paper's central claim has two parts, plausibility and personalization, and the evidence for the second half is indirect: the human judges were not the target users, and UMA measures likelihood ranking under different conditioning profiles rather than actual user preference. The absence of error bars and significance tests further weakens the quantitative comparisons. With additional validation, the task and model would be a solid contribution.

major comments (4)
  1. [Section 5, Personalization] The UMA/MRR result demonstrates that the model's output distribution changes with the conditioning user profile, but it does not establish that the generated recipe is more appropriate for the target user than for another user. The pairwise human evaluation uses anonymous judges who see only the recipe name and ingredients, not the target user's history, so it tests plausibility, not personalized fit. Add a target-user study (for example, each user chooses between the personalized and baseline recipe for their own held-out review) or validate UMA against a behavioral signal such as the user's subsequent review or rating of the generated recipe.
  2. [Section 5, Tables 2 and 4] All quantitative results are reported as point estimates without error bars, confidence intervals, or significance tests, and several margins are small (for example, BPE PPL 9.516 vs. 9.551 for Prior Name vs. Enc-Dec, and coherence 1.82 vs. 1.77). Report multiple random seeds or bootstrapped intervals and paired significance tests, or explicitly state that the differences are not statistically tested. This is essential for the comparative claims made throughout the paper.
  3. [Section 4, Dataset and task framing] The paper treats users' Food.com reviews as evidence of recipes they 'consumed,' but a review is not necessarily a record of cooking or liking a recipe. The personalization signal and the UMA evaluation both depend on this assumption, and the paper provides no analysis of review text or ratings to validate it. Please justify or soften this assumption and discuss how review behavior might differ from consumption behavior.
  4. [Section 5, Personalization] The description of the nine 'randomly generated user profiles' used in UMA evaluation is underspecified. If these profiles are drawn from a different activity distribution (for example, shorter or less typical histories), the gold user could rank highest for reasons unrelated to personalization. Describe the sampling procedure and, ideally, include a control condition with random histories matched on length and recipe popularity.
minor comments (5)
  1. [Section 5, Baselines] The Neural Checklist model is mentioned as adapted and then dropped because Enc-Dec has 'comparable performance,' but no Neural Checklist numbers are shown; either report them in Table 2 or move them to the appendix for transparency.
  2. [Section 4, Techniques] The manually constructed list of 58 techniques and the string-match detection method are described with no precision about matching rules or false positives; a sentence on how inflected forms and multiword expressions are handled would help.
  3. [Section 5] Calorie-level conditioning is part of the model but is never ablated or analyzed; a small experiment or a statement on its effect would clarify whether it contributes to the reported gains.
  4. [Table 2] The NN row reports a dash for BPE PPL with no explanation; please add a note on why that value is omitted.
  5. [Section 5, Recipe Level Coherence] The recipe-level coherence metric is said to have a maximum score of 2, but the reported scores (1.77-1.82) are all close to each other and to the maximum; consider reporting the distribution or the score of gold recipes as a reference point.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the personalized-generation claim is supported by held-out likelihood, external coherence metrics, and human evaluation rather than by a fitted constant or self-citation chain.

full rationale

The paper contains no derivation step that reduces to its own inputs by construction. The model is trained to maximize the log-likelihood of the gold recipe text conditioned on a user's prior reviews, and the UMA/MRR personalization metric then asks whether the same model assigns the highest likelihood to the gold user among ten profiles. This is a self-consistency check of whether user conditioning affects the output distribution, not a tautology: random-user likelihoods are not forced to be lower, and the gold recipe itself is held out from the user history by sequential leave-one-out. The coherence and entailment evaluations use externally pretrained scorers (Bosselut et al. 2018a; BERT), and the human pairwise evaluation is an external judge-based benchmark, albeit one that tests general plausibility more than per-user fit. Self-citations (Ni and McAuley 2018; Kang and McAuley 2018) are methodological or related-work references and are not load-bearing to the central claim. Any concerns about whether UMA or anonymous judges adequately capture true personalization are validity questions, not circularity.

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

The model's central claim rests on the neural network's learned parameters and on domain assumptions about how reviews signal preference. The free parameters listed are hand-chosen hyperparameters that affect performance but are not fitted to a target prediction. The axioms capture the data and task assumptions that are load-bearing for the evaluation.

free parameters (4)
  • k (prior recipe attention window) = 20
    The number of most recent prior recipes attended over; set to the 80th percentile of user interaction counts. Affects how much history is used.
  • top-k sampling k = 3
    Decoding parameter selected because it produces satisfactory results (Section 5); a tuned value, not a predicted quantity.
  • Hidden size dh = 256
    Model capacity for encoder and decoder GRUs; standard choice, no sensitivity analysis.
  • Embedding dimensions (V/I/R/X/C) = 300/10/50/50/5
    Hand-set embedding sizes for vocabulary, ingredients, recipes, techniques, and calorie level.
assumptions (4)
  • standard math GRU, BiGRU, and attention mechanisms provide differentiable learning signal for the encoder-decoder.
    The model relies on standard neural components described in Section 3.
  • domain assumption Each review implies the user consumed and preferred the recipe.
    Section 4 treats all 700K reviews as positive actions for building user profiles; noisy signals could weaken personalization.
  • domain assumption The first 3-5 ingredients in a recipe are a sufficient partial input to generate the full instructions.
    Section 5 gives the model the first 3-5 ingredients of each recipe during training and evaluation.
  • ad hoc to paper The manually constructed list of 58 techniques, with string-match detection, adequately represents cooking techniques.
    Section 4: list built from 384 actions in Bosselut et al. (2018b); string matching is a coarse approximation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Generating Personalized Recipes from Historical User Preferences." pith.science (2026). https://pith.science/paper/NQ5WJQWH

@misc{pith2026190900105,
  author       = {Pith},
  title        = {Pith review of: Generating Personalized Recipes from Historical User Preferences},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NQ5WJQWH}},
  note         = {Machine review of arXiv:1909.00105}
}
read the original abstract

Existing approaches to recipe generation are unable to create recipes for users with culinary preferences but incomplete knowledge of ingredients in specific dishes. We propose a new task of personalized recipe generation to help these users: expanding a name and incomplete ingredient details into complete natural-text instructions aligned with the user's historical preferences. We attend on technique- and recipe-level representations of a user's previously consumed recipes, fusing these 'user-aware' representations in an attention fusion layer to control recipe text generation. Experiments on a new dataset of 180K recipes and 700K interactions show our model's ability to generate plausible and personalized recipes compared to non-personalized baselines.

Figures

Figures reproduced from arXiv: 1909.00105 by the authors.

Figure 1
Figure 1. Sample data flow through model architec [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. A sample question for pairwise evaluation survey. [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. A sample question for coherence evaluation survey. [PITH_FULL_IMAGE:figures/full_fig_p012_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 4 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Symmetric Behavior Regularized Policy Optimization

    cs.LG 2025-08 unverdicted novelty 6.0 of 10

    Symmetric behavior regularization for offline RL becomes tractable by expanding any f-divergence into a truncated Pearson-Vajda series, yielding a closed-form policy and bounded approximation error.

  2. MemBench: Towards More Comprehensive Evaluation on the Memory of LLM-based Agents

    cs.CL 2025-06 conditional novelty 5.0 of 10

    MemBench introduces a multi-scenario, multi-level memory benchmark for LLM agents, evaluating factual and reflective memory across participation and observation settings.

  3. Fine-tuning Language Models for Recipe Generation: A Comparative Analysis and Benchmark Study

    cs.CL 2025-02 conditional novelty 4.0 of 10

    Fine-tuning small language models for recipe generation produces mixed results: Phi-2 degrades on the authors' custom quality scores while SmolLM-360M and 1.7B perform similarly.

  4. Retrieval Augmented Recipe Generation

    cs.CV 2024-11 conditional novelty 4.0 of 10

    A retrieval-augmented LMM with stochastic retrieval sampling and self-consistency voting improves recipe generation from food images on Recipe1M.

Reference graph

Works this paper leans on

37 extracted references · 24 canonical work pages · cited by 4 Pith papers

  1. [1]

    URL: " 'urlintro :=

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

  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]

    Lawrence Zitnick, Devi Parikh, and Dhruv Batra

    Aishwarya Agrawal, Jiasen Lu, Stanislaw Antol, Margaret Mitchell, C. Lawrence Zitnick, Devi Parikh, and Dhruv Batra. 2017. https://doi.org/10.1007/s11263-016-0966-6 VQA: visual question answering . IJCV, 123(1):4--31

    doi:10.1007/s11263-016-0966-6 2017

  4. [4]

    Dzmitry Bahdanau, Kyunghyun Cho, and Yoshua Bengio. 2015. http://arxiv.org/abs/1409.0473 Neural machine translation by jointly learning to align and translate . In ICLR

    arXiv 2015

  5. [5]

    Ashutosh Baheti, Alan Ritter, Jiwei Li, and Bill Dolan. 2018. https://aclanthology.info/papers/D18-1431/d18-1431 Generating more interesting responses in neural conversation models with distributional constraints . In EMNLP

    work page 2018

  6. [6]

    Antoine Bosselut, Asli C elikyilmaz, Xiaodong He, Jianfeng Gao, Po - Sen Huang, and Yejin Choi. 2018 a . https://aclanthology.info/papers/N18-1016/n18-1016 Discourse-aware neural rewards for coherent text generation . In NAACL-HLT

    work page 2018

  7. [7]

    Antoine Bosselut, Omer Levy, Ari Holtzman, Corin Ennis, Dieter Fox, and Yejin Choi. 2018 b . https://openreview.net/forum?id=rJYFzMZC- Simulating action dynamics with neural process networks . In ICLR

    work page 2018

  8. [8]

    Kyunghyun Cho, Bart van Merrienboer, C aglar G \" u l c ehre, Dzmitry Bahdanau, Fethi Bougares, Holger Schwenk, and Yoshua Bengio. 2014. http://aclweb.org/anthology/D/D14/D14-1179.pdf Learning phrase representations using RNN encoder-decoder for statistical machine translation . In EMNLP

    work page 2014

Show all 37 references
  1. [9]

    Jacob Devlin, Ming - Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. https://aclweb.org/anthology/papers/N/N19/N19-1423/ BERT: pre-training of deep bidirectional transformers for language understanding . In NAACL-HLT 2019

    2019

  2. [10]

    Angela Fan, Mike Lewis, and Yann Dauphin. 2018. https://aclanthology.info/papers/P18-1082/p18-1082 Hierarchical neural story generation . In ACL

    2018

  3. [11]

    Albert Gatt and Emiel Krahmer. 2018. https://doi.org/10.1613/jair.5477 Survey of the state of the art in natural language generation: Core tasks, applications and evaluation . J. Artif. Intell. Res., 61:65--170

    2018 doi

  4. [12]

    Jiatao Gu, Zhengdong Lu, Hang Li, and Victor O. K. Li. 2016. http://aclweb.org/anthology/P/P16/P16-1154.pdf Incorporating copying mechanism in sequence-to-sequence learning . In ACL

    2016

  5. [13]

    Jeremy Howard and Sebastian Ruder. 2018. https://doi.org/10.18653/v1/P18-1031 Universal language model fine-tuning for text classification . In ACL

    2018 doi

  6. [14]

    Wang - Cheng Kang and Julian McAuley. 2018. https://doi.org/10.1109/ICDM.2018.00035 Self-attentive sequential recommendation . In ICDM

    2018

  7. [15]

    Chlo \' e Kiddon, Ganesa Thandavam Ponnuraj, Luke Zettlemoyer, and Yejin Choi. 2015. http://aclweb.org/anthology/D/D15/D15-1114.pdf Mise en place: Unsupervised interpretation of instructional recipes . In EMNLP

    2015

  8. [16]

    Chlo \' e Kiddon, Luke Zettlemoyer, and Yejin Choi. 2016. http://aclweb.org/anthology/D/D16/D16-1032.pdf Globally coherent text generation with neural checklist models . In EMNLP

    2016

  9. [17]

    Kingma and Jimmy Ba

    Diederik P. Kingma and Jimmy Ba. 2015. http://arxiv.org/abs/1412.6980 Adam: A method for stochastic optimization . In ICLR

    2015 arXiv

  10. [18]

    Leo Lepp \" a nen, Myriam Munezero, Mark Granroth - Wilding, and Hannu Toivonen. 2017. https://aclanthology.info/papers/W17-3528/w17-3528 Data-driven news generation for automated journalism . In INLG

    2017

  11. [19]

    Moore, Shuo Chen, Douglas Turnbull, and Thorsten Joachims

    Joshua L. Moore, Shuo Chen, Douglas Turnbull, and Thorsten Joachims. 2013. http://www.ppgia.pucpr.br/ismir2013/wp-content/uploads/2013/09/220\_Paper.pdf Taste over time: The temporal dynamics of user preferences . In ISMIR

    2013

  12. [20]

    Lipton, Sharad Vikram, and Julian McAuley

    Jianmo Ni, Zachary C. Lipton, Sharad Vikram, and Julian McAuley. 2017. https://aclanthology.info/papers/I17-1079/i17-1079 Estimating reactions and recommending products with generative models of reviews . In IJCNLP

    2017

  13. [21]

    Jianmo Ni and Julian McAuley. 2018. https://aclanthology.info/papers/P18-2112/p18-2112 Personalized review generation by expanding phrases and attending on aspect-aware representations . In ACL

    2018

  14. [22]

    Romain Paulus, Caiming Xiong, and Richard Socher. 2018. https://openreview.net/forum?id=HkAClQgA- A deep reinforced model for abstractive summarization . In ICLR

    2018

  15. [23]

    Massimo Quadrana, Paolo Cremonesi, and Dietmar Jannach. 2018. https://doi.org/10.1145/3209219.3209270 Sequence-aware recommender systems . In UMAP

    2018

  16. [24]

    Radev, Hong Qi, Harris Wu, and Weiguo Fan

    Dragomir R. Radev, Hong Qi, Harris Wu, and Weiguo Fan. 2002. http://www.lrec-conf.org/proceedings/lrec2002/pdf/301.pdf Evaluating web-based question answering systems . In LREC

    2002

  17. [25]

    Alec Radford, Karthik Narasimhan, Tim Salimans, and Ilya Sutskever. 2018. https://s3-us-west-2.amazonaws.com/openai-assets/research-covers/language-unsupervised/language_understanding_paper.pdf Improving language understanding by generative pre-training

    2018

  18. [26]

    Alec Radford, Jeff Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever. 2019. https://openai.com/blog/better-language-models/ Language models are unsupervised multitask learners

    2019

  19. [27]

    Lam, Sean M

    Al Mamunur Rashid, Istvan Albert, Dan Cosley, Shyong K. Lam, Sean M. McNee, Joseph A. Konstan, and John Riedl. 2002. https://doi.org/10.1145/502716.502737 Getting to know you: learning new user preferences in recommender systems . In IUI

    2002

  20. [28]

    Steffen Rendle, Christoph Freudenthaler, Zeno Gantner, and Lars Schmidt - Thieme. 2009. https://dslpitt.org/uai/displayArticleDetails.jsp?mmnu=1&smnu=2&article\_id=1630&proceeding\_id=25 BPR: bayesian personalized ranking from implicit feedback . In UAI

    2009

  21. [29]

    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 ACL

    2017 doi

  22. [30]

    Rico Sennrich, Barry Haddow, and Alexandra Birch. 2016. http://aclweb.org/anthology/P/P16/P16-1162.pdf Neural machine translation of rare words with subword units . In ACL

    2016

  23. [31]

    Mayumi Ueda, Mari Takahata, and Shinsuke Nakajima. 2011. http://dl.acm.org/citation.cfm?id=2887675.2887686 User's food preference extraction for personalized cooking recipe recommendation . In SPIM

    2011

  24. [32]

    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://papers.nips.cc/paper/7181-attention-is-all-you-need Attention is all you need . In NIPS

    2017

  25. [33]

    Williams and David Zipser

    Ronald J. Williams and David Zipser. 1989. https://doi.org/10.1162/neco.1989.1.2.270 A learning algorithm for continually running fully recurrent neural networks . Neural Computation, 1(2):270--280

    1989 doi

  26. [34]

    Jingjing Xu, Xuancheng Ren, Yi Zhang, Qi Zeng, Xiaoyan Cai, and Xu Sun. 2018. https://aclanthology.info/papers/D18-1462/d18-1462 A skeleton-based model for promoting coherence among sentences in narrative story generation . In EMNLP

    2018

  27. [35]

    Zichao Yang, Phil Blunsom, Chris Dyer, and Wang Ling. 2017. https://aclanthology.info/papers/D17-1197/d17-1197 Reference-aware language models . In EMNLP

    2017

  28. [36]

    Weischedel, Kevin Knight, Dongyan Zhao, and Rui Yan

    Lili Yao, Nanyun Peng, Ralph M. Weischedel, Kevin Knight, Dongyan Zhao, and Rui Yan. 2018. http://arxiv.org/abs/1811.05701 Plan-and-write: Towards better automatic storytelling . CoRR, abs/1811.05701

    2018 arXiv

  29. [37]

    Saizheng Zhang, Emily Dinan, Jack Urbanek, Arthur Szlam, Douwe Kiela, and Jason Weston. 2018. https://aclanthology.info/papers/P18-1205/p18-1205 Personalizing dialogue agents: I have a dog, do you have pets too? In ACL

    2018

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.