{"id":"bf0834f6-9813-492c-9fc6-13b28b61e010","arxiv_id":"2412.16188","paper_version":1,"verdict":"REJECT","confidence":"HIGH","novelty_score":2.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"This is a review of seven influential deep learning models that does not present new research results and suffers from methodological and integrity issues.","lead":"This preprint is a survey of seven well-known deep learning architectures: ResNets, Transformers, GANs, VAEs, GNNs, CLIP, and diffusion models. It presents a selection methodology based on a small expert survey and Google Scholar citation counts, but the execution contains serious errors and unverifiable references.","discovery_kind":"review","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The paper's own chronological filter contradicts the inclusion of GNNs; the 'Magnificent Seven' selection is therefore internally unsupported.","rationale":"The reader's REJECT verdict is supportable, but the most decisive flaw is narrower and more internal than 'survey respondents may be biased.' The paper's own Section 3.1 eliminates algorithms whose chronological beginnings predate 2013, yet Section 4.5 acknowledges that GNNs were introduced in the early 2000s. Because the selection of exactly seven algorithms is the paper's central contribution, this inconsistency alone invalidates the claim as written, independent of citation-count volatility or survey representativeness. The proposed test is a literature-date check that requires no new experiments: if the original GNN papers predate 2013, the stated selection rule does not produce the listed seven. The manuscript does contain useful expository material on each architecture, but the 'Magnificent Seven' framing and the 'first to survey the top seven' claim fail at the first methodological step. I therefore agree with rejection, with the caveat that the reason is stronger than a merely biased sample: the stated selection rule, applied consistently, does not yield the set presented.","tokens_in":32462,"tokens_out":5695,"duration_ms":55410,"concrete_test":"Take the original GNN publications (e.g., Gori et al. 2005; Scarselli et al. 2009) and apply the same chronological-beginnings rule used in Section 3.1 to exclude CNN and LSTM. If GNNs predate 2013, the paper must either drop GNN from the seven or explicitly redefine what 'created in the decade' means; under either correction, the current title and Section 3.1 conclusion change. Also check whether the survey's twelve pre-filter algorithms are listed with their introduction dates; without such a list, the filtering step cannot be audited.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that the seven algorithms were selected as the most influential of the last decade via a survey plus a chronological filter. In Section 3.1, the authors state that they checked the 'chronological beginnings' of candidate algorithms and eliminated CNN and LSTM because they were not introduced in the considered timeline, leaving exactly seven. Footnote 1 explicitly dates GNNs to 2016. However, Section 4.5's Overview states that GNNs were 'introduced in the early 2000s.' Applying the same chronological-beginnings test used to remove CNN and LSTM, GNNs should also be removed, which would leave only six algorithms. Instead, the paper uses GCN (Kipf & Welling 2016) as the foundational GNN paper for citation ranking, conflating a specific modern architecture with the algorithm family as a whole. This is an internal contradiction in the selection methodology, not a matter of external taste. If GNNs are excluded, the set must be rebuilt or the decade criterion must be redefined; as written, the headline selection claim is unsupported.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript proposes a survey of what it calls the \"Magnificent Seven\" deep learning algorithms — ResNets, Transformers, GANs, VAEs, GNNs, CLIP, and Diffusion Models — selected through a questionnaire completed by 100 respondents and ranked by Google Scholar citation counts of foundational papers. For each algorithm, the paper provides an overview, core architecture, mathematical foundations, algorithmic procedure, training and optimization, extensions, applications, challenges, and future directions, plus a discussion of common building blocks. The stated aim is to serve as a practical manual for newcomers and transitioning researchers.","tokens_in":32612,"tokens_out":4754,"duration_ms":40406,"significance":"The paper is clearly written in a tutorial style and covers a broad set of important material, including the main equations for each architecture (e.g., Eqs. (1), (3), (5), (6), (7), (8), (9)) and practical training advice. If the selection were robust, the survey would be a convenient entry point. However, the central selection claim is undermined by an internal chronological contradiction and by a survey base that the authors themselves acknowledge is institutionally concentrated. The paper is also not a systematic review; it is a curated tutorial, and the \"top seven\" framing overstates the evidence. The manuscript ships no machine-checked proofs or reproducible code, which is not required for a survey, but it does offer a useful consolidated reference for the included architectures.","major_comments":[{"comment":"The selection criterion in Section 3.1 removes CNN and LSTM because their chronological beginnings lie outside 2013–2024, and Footnote 1 explicitly dates GNNs to 2016. Yet Section 4.5 (Overview) states that GNNs were \"introduced in the early 2000s,\" and the citation-based ranking uses the 2016 GCN paper as the foundational GNN reference. If the chronological-beginnings test is applied uniformly, GNNs should be removed by the same reasoning used to remove CNN and LSTM. The paper either needs to define the criterion as \"the breakthrough paper appeared in the window,\" which would also affect other choices, or it needs a separate argument for why GNNs are included. As written, the selection of the seven is internally inconsistent.","section":"Section 3.1 / Footnote 1 / Section 4.5"},{"comment":"Figure 3 is not an original figure. It contains text from another paper on ResNet training heuristics, including \"Model Tweaks,\" \"Table 4,\" references to ResNet-B/C/D, and a citation to \"He et al. [9]\" that are not part of the present manuscript. This is a copy-paste error that must be corrected; as published it undermines the reliability of the other figures and the survey as a whole.","section":"Section 4.1, Figure 3"},{"comment":"The survey-based selection in Section 3.1 is not as robust as the paper claims. The sample is 100 respondents, and Figure 2a shows a majority from MBZUAI and Sun Yat-sen University, with most respondents being M.Sc./Ph.D. students with 1–5 years of experience. The paper provides no response rate, no sampling frame, no independent validation, and the authors' own \"Bias\" section concedes institutional bias. The Google Scholar citation ranking is also not independent of popularity and includes self-citations. Since the \"top seven\" claim rests on this methodology, the paper should either report a more diverse and documented survey or reframe the selection as the authors' curated choice informed by the survey.","section":"Section 3.1 and Bias section"},{"comment":"Figure 1's caption states that the answers to the survey questions were \"randomly generated by the authors.\" If these are illustrative examples, the caption should say so explicitly and should not present them as actual survey responses; if they are actual responses, the caption is false. In either case, the figure as presented confuses the evidentiary status of the survey. Additionally, the questionnaire itself is not described in the body text beyond this figure, so the reader cannot evaluate the wording or the response format.","section":"Figure 1 caption"}],"minor_comments":[{"comment":"The references \"Smith & Lee (2020)\" and \"Doe & Row (2021)\" appear to be placeholder citations; they are not identifiable real works and should be replaced or removed.","section":"Section 2"},{"comment":"The text contains typos \"GPT-4o14\" and \"GPT4o1\" for GPT-4o; these should be corrected for consistency.","section":"Section 4.2"},{"comment":"The citation to \"Dziugaite (2020)\" for layer normalization in VAEs is not a standard reference for this technique; consider citing the original layer normalization work instead.","section":"Section 4.4"},{"comment":"The statement that GANs \"Employ RMSProp, ADAM, and rate scheduling\" is vague and unsupported by a specific reference; the sentence should be clarified or removed.","section":"Section 5.2"},{"comment":"The algorithmic procedure for ResNets has a numbering error: item \"2. Residual Block Processing:\" is immediately followed by \"3. Convolutional Layers,\" leaving an empty step; the numbering should be fixed.","section":"Section 4.1"},{"comment":"The three panels of Figure 2 do not report sample sizes per category; adding exact numbers would help the reader assess the representativeness of the survey.","section":"Figure 2"}],"recommendation":"major_revision","confidential_remarks":"The manuscript contains several nonstandard citation practices, including placeholder citations and a high number of self-citations by the corresponding author and co-authors. The survey appears to have been circulated largely within the authors' own institutions. These factors, combined with the Figure 3 copy-paste error, suggest the manuscript needs thorough editorial vetting. I do not see evidence of deliberate misconduct, but the current version is not ready for publication as a rigorous survey."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The central claim of this paper is that it identifies the top seven deep learning algorithms of the last decade via a survey plus chronological filter. That claim does not survive contact with the paper itself. The selection method is biased and internally inconsistent. The survey is 100 respondents, mostly from two institutions, and the citation counts are non-reproducible snapshots. More damning, the chronological filter is applied differently to GNNs than to CNN and LSTM: the paper's own footnote dates GNNs to 2016, but Section 4.5 says GNNs were introduced in the early 2000s. Under the same test that eliminated CNN and LSTM, GNNs should also be removed. Using GCN (2016) as the foundational GNN paper is a category error. That is a load-bearing contradiction, not a matter of taste.\n\nThere is some value here. The individual sections on each architecture are readable and mostly accurate, with standard math and a reasonable structure. A newcomer could learn the basics of Transformers or VAEs from these pages. The discussion of normalization, optimization, and rate scheduling in Section 5 is a useful addition.\n\nThe soft spots go beyond methodology. Figure 3 is an obvious copy-paste from a different paper about ResNet tweaks; it contains references to Table 4 and batch size experiments that have nothing to do with this survey. That kind of error destroys confidence in the whole manuscript. There are also references that look fabricated: \"Smith & Lee 2020, Quantum machine learning, Nature\" and \"Doe & Row 2021, Ethical considerations in AI deployment\" are placeholder names, not real citations. The paper also overstates its novelty; being the first to survey these seven is trivial and not a contribution.\n\nWho is this for? A patient beginner might get some value from isolated sections, but the reliability problems mean I would not point students to it. The central argument is unsupported as written.\n\nMy recommendation: desk reject. This needs a complete rewrite of the methodology, removal of the copied figure, and a clean reference list. As it stands, it is not a serious research contribution.","headline":"A survey with a broken selection method and integrity problems that undercut its central claim.","tokens_in":33197,"tokens_out":2785,"would_cite":false,"duration_ms":27822,"reading_group":"no","serious_thinker":"no","would_accept_peer_review":false},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["68T07"],"pacs":[],"model":"deepseek-v4-flash","headline":"This survey argues that seven algorithms—ResNets, Transformers, GANs, VAEs, GNNs, CLIP, and diffusion models—are the most influential deep learning developments of 2013–2024.","keywords":["deep learning survey","ResNets","Transformers","GANs","VAEs","GNNs","CLIP","diffusion models"],"falsifier":"Run the same four-question survey with a much larger and more globally diverse respondent pool and re-collect citation counts at a later date; if a different set of algorithms reaches the top seven, or the ordering changes materially, the paper's central ranking fails as a stable description. A simpler check is to compute citation counts for the excluded algorithms (CNN, LSTM) within the same window; if those counts dwarf several of the seven, the decade filter rather than the citation proxy is doing the selecting.","tokens_in":32268,"feed_emoji":"🤖","tokens_out":6949,"duration_ms":51130,"temperature":0.7,"pith_summary":"This survey argues that seven algorithms—ResNets, Transformers, GANs, VAEs, GNNs, CLIP, and diffusion models—are the most influential deep learning developments of 2013–2024. It selects them through a questionnaire answered by 100 academics and professionals, keeps only algorithms born within the decade, and orders them by the citation counts of their foundational papers as of December 10, 2024. The authors claim this is the first survey to treat these seven as a group and intend it as a practical manual for newcomers and researchers transitioning into deep learning. If the selection is accepted, the paper provides a compact entry point to the architectures that now dominate vision, language, and generative modeling.","feed_headline":"Survey names seven algorithms that defined deep learning's decade","feed_subtitle":"ResNets, Transformers, GANs, VAEs, GNNs, CLIP, and diffusion models top the influence ranking.","key_machinery":"The load-bearing mechanism is the two-stage selection protocol: a four-question survey distributed to students, research assistants, postdocs, and faculty yields an initial candidate set; a chronological filter restricts it to algorithms introduced within 2013–2024; and citation counts of each algorithm's foundational paper, taken on December 10, 2024, set the final order. This protocol turns subjective judgments of influence into an ordered, reproducible list, and the composition of the respondent pool determines whether that list generalizes.","core_discovery":"On the paper's own terms, the central result is a grouping and a ranking: ResNets, Transformers, GANs, VAEs, GNNs, CLIP, and diffusion models are 'The Magnificent Seven' of the past decade. The ranking starts from a survey of 100 respondents; twelve candidate algorithms survive aggregation, and five are dropped because they predate the 2013–2024 window (CNN and LSTM are named). The final order follows the citation counts of each algorithm's foundational paper, with ResNets first (247,440), Transformers second (144,501), GANs third (75,445), VAEs fourth (40,549), GNNs fifth (40,130), CLIP sixth (25,847), and diffusion models seventh (15,632). Each chapter then presents the algorithm's core architecture, mathematical foundations, algorithmic procedure, training and optimization, extensions, applications, challenges, and future directions, all in service of the stated goal: a practical manual for entering deep learning.","pith_inferences":["Re-running the same survey with a larger, more geographically diverse respondent pool is the natural test of whether these seven, and not another seven, are the decade's leaders.","Because citation counts accumulate over time, the ranking may understate the current influence of recent algorithms such as CLIP and diffusion models; a usage-based or survey-based recency measure would likely shift the order.","The two-stage protocol itself could be reused as a standing method to rank algorithmic influence in future decades.","If the selection is right, textbooks and course syllabi for the decade could reasonably be organized around these seven architectures."],"forward_implications":["If the ranking is accepted, newcomers have a defensible shortlist for what to learn first: the seven architectures that define modern practice.","The citation ordering provides a rough influence hierarchy, placing ResNets highest and diffusion models lowest among the seven despite diffusion's recent dominance.","Each algorithm's chapter supplies a structured reference covering math, training, variants, applications, and challenges, usable independent of the ranking.","The decade boundary means older workhorses like CNN and LSTM are excluded by design, not because the authors consider them unimportant.","The paper's future-work agenda anticipates expanding the list and building new evaluation measures and benchmarks for deep learning algorithms."],"supporting_citations":[{"why":"Foundational ResNet paper; its citation count (247,440) anchors ResNets' number-one rank.","marker":"He et al. (2016a)"},{"why":"Foundational transformer paper; its citation count (144,501) anchors Transformers' second rank.","marker":"Vaswani et al. (2017)"},{"why":"Foundational GAN paper; its citation count (75,445) anchors GANs' third rank.","marker":"Goodfellow et al. (2014)"},{"why":"Foundational VAE paper; its citation count (40,549) anchors VAEs' fourth rank.","marker":"Kingma & Welling (2013)"},{"why":"Foundational GCN paper; its citation count (40,130) anchors GNNs' fifth rank.","marker":"Kipf & Welling (2016)"},{"why":"Foundational CLIP paper; its citation count (25,847) anchors CLIP's sixth rank.","marker":"Radford et al. (2021)"},{"why":"Foundational DDPM paper; its citation count (15,632) anchors diffusion models' seventh rank.","marker":"Ho et al. (2020)"}],"fun_headline_variants":["Seven algorithms that ruled deep learning's decade","The Magnificent Seven: deep learning's top algorithms","ResNets top the seven that defined deep learning","Survey ranks the seven most influential deep learning models"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The ranking is only as trustworthy as the respondent pool and the citation snapshot: if the 100 respondents, drawn mostly from two universities, are not representative of the global deep learning community, the 'Magnificent Seven' may not be the true top seven.","fun_headline_variants_meta":{"raw":{"variants":["Seven algorithms that ruled deep learning's decade","The Magnificent Seven: deep learning's top algorithms","ResNets top the seven that defined deep learning","Survey ranks the seven most influential deep learning models"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000165,"raw_usage":{"total_tokens":1251,"prompt_tokens":949,"completion_tokens":302,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":565,"completion_tokens_details":{"reasoning_tokens":241}},"tokens_in":565,"tokens_out":302,"duration_ms":592319,"temperature":1.0,"reasoning_tokens":241,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T15:56:57.270580+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the same four-question survey with a much larger and more globally diverse respondent pool and re-collect citation counts at a later date; if a different set of algorithms reaches the top seven, or the ordering changes materially, the paper's central ranking fails as a stable description. A simpler check is to compute citation counts for the excluded algorithms (CNN, LSTM) within the same window; if those counts dwarf several of the seven, the decade filter rather than the citation proxy is doing the selecting.","supporting_citations":[],"review_version":1}