Pith. sign in

REVIEW 4 major objections 6 minor 49 references

TomaMMU: A Comprehensive Multimodal Understanding Benchmark for Tomato Leaf Diseases

T0 review · 4 major / 6 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 The paper claims that current vision-language models, tested zero-shot on a new tomato-disease benchmark, fall short on fine-grained symptom recognition and diagnostic reasoning, while fine-tuning on the TomaMMU dataset lifts…

desk verdict Useful tomato VQA dataset and benchmark, but the 96.09% fine-tuning claim is unverifiable until the authors prove TomaBench is disjoint from the training split. read the letter →

arxiv 2608.08727 v1 pith:44UZ4NE6 submitted 2026-08-09 cs.CV cs.AI

Gia-Han Truong , Khang Nguyen Quoc , Luyl-Da Quach This is my paper · ORCID
classification cs.CVcs.AI
keywords Vision-LanguageModelsMultimodalDatasetVisualQuestionAnsweringBenchmarkAgriculturaldecisionsupportTomatoleafdiseaseFine-grainedrecognitionTomaBench
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

Vision-language models are being proposed for agricultural decision support, but the paper argues that no benchmark has measured whether they can actually diagnose tomato leaf disease beyond simple classification. TomaMMU supplies that measurement: 28,808 field and laboratory images, 213,119 human-annotated question-answer pairs, and a held-out TomaBench of 42,626 multiple-choice and 37,816 open-ended questions spanning seven tasks from basic perception to expert diagnosis. Testing 14 current VLMs zero-shot, the paper finds they underperform precisely where diagnosis is hardest—symptoms, pathogens, scientific names—and that simple fine-tuning on TomaMMU raises MCQ accuracy to 96.09%, far above every zero-shot baseline. If the benchmark is sound, it gives the field both a target to beat and evidence that domain-specific data, not model scale, is what moves agricultural VLM performance.

What carries the argument

The load-bearing object is TomaBench's hierarchical task structure: seven question types—Healthy/Diseased Classification, Leaf Count, Crop Species Identification, Symptom Identification, Pathogen Classification, Disease Classification, Scientific Name Classification—stacked into Basic Perception, Pathology Understanding, and Expert Diagnosis levels. The argument runs through this ladder: models climb from binary screening to fine-grained morphological reasoning, and the benchmark's score profile (strong on Perception, weak on Pathology/Expert) is what exposes the recognition-and-reasoning gap. The fine-tuned TomaLLaVA, built from SCOLD as image encoder and LFM2.5-1.2B-Instruct as base LLM with LoRA training, is the vehicle that demonstrates the gap closes with domain data.

What would settle it

Take a random sample of TomaBench images and have independent plant pathologists re-answer the questions, then compute inter-annotator agreement. Separately, check for image or question-template overlap between the fine-tuning set and TomaBench. If agreement is low or overlap is nonzero, the benchmark's central comparisons do not hold.

Watch

Extended reading notes

Core claim

TomaMMU is a curated multimodal dataset of 28,808 tomato-leaf images (15 categories) with 213,119 human-annotated VQA pairs, and TomaBench is its held-out evaluation core of 42,626 MCQs and 37,816 OEQs organized into seven tasks under three levels of diagnostic depth. On TomaBench, the paper reports that 14 state-of-the-art vision-language models, evaluated zero-shot, suffer pronounced drops on fine-grained tasks (pathogen classification, symptom identification, scientific-name classification), with even the strongest proprietary model averaging only 70.3% on MCQs and roughly 2.5/5 on open-ended questions. Fine-tuning a compact model (SCOLD encoder + LFM2.5-1.2B base, LoRA) on the 124,367 training QA pairs yields 96.09% average MCQ accuracy, +34.8 points over the best zero-shot baseline, which the paper attributes to domain-specific supervision rather than scale.

Load-bearing premise

The entire benchmark rests on the accuracy of its human-curated answers: if the ground-truth labels are wrong, or if any of the 42,626 evaluation questions overlap the 124,367 fine-tuning pairs, then both the measured performance gap and the fine-tuning gain are artifacts rather than facts about the models.

Editorial extensions

If this is right

  • Agricultural decision-support systems built on general VLMs should not be deployed for tomato disease diagnosis without domain adaptation, since the best zero-shot model averages only 70.3% on MCQs and near-random scores on symptom, pathogen, and scientific-name tasks.
  • TomaBench gives the field a reusable evaluation protocol with seven tasks across three reasoning levels, so future agri-VLM claims can be compared on the same ground.
  • Because fine-tuning a small model on TomaMMU surpasses much larger zero-shot models, domain-specific data collection may matter more than parameter count for agricultural expertise.
  • The documented error categories—perceptual error and refusal, core miss, and granularity omission—point developers to specific failure modes to target.

Reading between the lines

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

  • A formal leakage check between TomaBench and the fine-tuning split is the immediate next test; the paper does not report one, so the 96.09% gain should be read as provisional.
  • Because answers were verified at 224x224 resolution, the benchmark may underestimate what higher-resolution models could see; re-annotation at native resolution could change difficulty.
  • The OEQ scores (~2.3 to 2.7 on a 5-point scale) suggest the MCQ results overstate real-world readiness; open-ended diagnosis, the format closer to farmer use, remains substantially harder.
  • The same three-stage pipeline could transfer to other crops, making TomaBench a template for crop-specific diagnostic benchmarks rather than a tomato-only resource.
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 / 6 minor

Summary. The paper introduces TomaMMU, a tomato-leaf disease multimodal understanding dataset with 28,808 images across 15 categories and 213,119 human-annotated visual question-answer pairs, together with TomaBench, an evaluation benchmark of multiple-choice and open-ended questions organized into a seven-task, three-level taxonomy. The authors evaluate 14 vision-language models and CLIP-based models in a zero-shot setting, report generally low accuracy on the benchmark, and introduce a fine-tuned model, TomaLLaVA, which reaches 96.09% accuracy on the MCQ portion. The central claim is that TomaBench exposes pronounced gaps in fine-grained recognition and factually grounded reasoning in current VLMs, and that simple fine-tuning on TomaMMU substantially narrows this gap.

Significance. If the benchmark's validity and the held-out status of TomaBench are established, TomaMMU would be a useful, large-scale resource for agricultural VQA: it is publicly released, combines field and in-house images, covers 14 diseases plus a healthy class with seven task types, and provides a systematic zero-shot comparison across multiple model families. The hierarchical taxonomy from Basic Perception to Expert Diagnosis is a reasonable structure for probing diagnostic reasoning, and the external zero-shot evaluations give the benchmark independent grounding. However, the manuscript's central quantitative claims currently rest on unresolved internal inconsistencies in the dataset counts and on the absence of a demonstrated split between fine-tuning data and TomaBench, so the resource's value cannot yet be assessed at face value.

major comments (4)
  1. [§3.1, §3.2, Fig. 1, Introduction] The reported dataset sizes are mutually inconsistent. Section 3.1 states a training set of 124,367 VQA pairs and an evaluation set of 42,626 MCQs plus 37,816 OEQs called TomaBench; the Introduction and Conclusion describe TomaBench as 42,626 question-answer pairs; the dataset total is 213,119; and Fig. 1 shows a 60/20/20 split. The sum 124,367 + 42,626 + 37,816 = 204,809 differs from 213,119, and 60% of 213,119 is 127,871.4, not 124,367. Please provide a single consistent breakdown of how the 213,119 QA pairs are divided into training, validation, and TomaBench, and reconcile the MCQ and OEQ counts.
  2. [§3.2 (Stage 3), §4.3, Table 1] The central fine-tuning claim cannot be interpreted without an explicit leakage analysis. Stage 3 says only that 'a small, balanced subset, TomaBench, is selected for evaluation, while the remaining data is used for fine-tuning and training,' which is an assertion of intent rather than a demonstrated property. Because each MCQ and merged OEQ is templated from per-image metadata, a single shared or near-duplicate image between the training set and TomaBench would let TomaLLaVA memorize the associated metadata rather than perform diagnostic reasoning. The paper reports no image-level deduplication, no near-duplicate analysis, and no question/answer overlap check between the 124,367 fine-tuning pairs and TomaBench. Until this is established, the 96.09% MCQ result in §4.3 and Table 1 cannot support the claim that simple fine-tuning closes the gap.
  3. [§4.1, §4.4, Table 2] The OEQ evaluation pipeline is internally inconsistent and is not validated. Section 4.1 says the LLM-as-judge uses 'the GPT score,' while Section 4.4 says 'Our prompt for Gemini 2.0 Flash is presented in Table S2'; Table 2 labels the column 'GPT'. The paper should name the judge model, state whether GPT or Gemini was used, and report human-agreement or correlation results for the 1-5 score. The approximate values (e.g., '∼2.3') are also not exact scores, and no confidence intervals are given for any of the Table 1 or Table 2 numbers, so small accuracy differences among models (e.g., 59.2 vs. 58.7) may not be meaningful.
  4. [§3.2 (Stage 2), §3.3, §3.1] The benchmark's ground truth rests on expert annotation, but the manuscript provides no quantitative evidence of annotation quality. Section 3.2 states that 'a human domain expert strictly executes the entire metadata extraction and annotation pipeline' and that 'every data point undergoes a detailed manual verification process,' yet no inter-annotator agreement, error rate, or sample verification statistics are reported. Given that images are resized to 224×224 px, the claim that symptoms are 'visually identifiable' needs support; fine-grained symptoms such as small necrotic spots or mosaic patterns may not be reliably judged at this resolution. Please report annotation-quality measurements and discuss the resolution choice.
minor comments (6)
  1. [Abstract] The sentence 'Our results pronounced gaps...' is missing a verb; it should read 'Our results reveal pronounced gaps...'.
  2. [§3.2, Fig. 3a] The image count differs between text and figure: the text says 28,808 images, while Fig. 3a shows 28,799; also, Fig. 3a lists 'Bacterial Floundering' while Table S1 uses 'Bacterial Wilt'.
  3. [Table 1] Table 1 reports TomaLLaVA's average accuracy as 96.1 while the CSI column is omitted ('-'). The averaging convention is not defined; if the average is over the six reported tasks, state this explicitly, and if the CSI column is excluded from the average, say so.
  4. [§4.3] The text cites 'Fine-tuned VLMs (AgriCLIP, 93.73% Acc, 0.948 F1)' but AgriCLIP appears neither in Table 1 nor in the model list in Section 4; either include the full AgriCLIP evaluation or remove the comparison.
  5. [Tables 1 and 2] Model names are inconsistent between tables: 'LLaVA-1.5-7B' in Table 1 versus 'LLaVA-OV' in Table 2 for reference [16]; clarify which checkpoint was evaluated.
  6. [§4.1] No prompt template is provided for MCQ answering or for zero-shot evaluation of non-generative CLIP-based models; please include the exact prompts and the protocol used for models that do not natively generate answer options.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: TomaBench is asserted as a held-out evaluation subset and the zero-shot baselines are external, so the central claims do not reduce to their inputs.

full rationale

The paper's central claims are empirical benchmark results, not derived quantities. Section 3.2 Stage 3 states that 'a small, balanced subset, TomaBench, is selected for evaluation, while the remaining data is used for fine-tuning and training,' which describes a conventional held-out split rather than a circular reuse of the evaluation data. The 96.09% MCQ result in Table 1 is a measured accuracy of a LoRA-fine-tuned model evaluated on that asserted held-out TomaBench, not a fitted parameter renamed as a prediction, and no equation in the paper forces TomaBench performance to equal training-set statistics by construction. Self-citations to LeafNet (image source) and SCOLD (image encoder) are present but are not load-bearing for the headline result: the zero-shot comparisons include external VLM baselines, and the fine-tuned model's improvement is not justified by those citations. The absence of an explicit leakage or near-duplicate analysis is a validity risk that should be addressed, but the manuscript's own split description is a contrary assertion of disjointness, so leakage is not demonstrated and cannot be scored as circularity under the requirement to exhibit a specific reduction.

Assumptions & free parameters 5 free parameters · 5 assumptions · 3 invented entities

The central claims rest on hand-chosen design choices (seven-task taxonomy, three-level hierarchy, 60/20/20 split, 224x224 resolution, 1-5 judge scale) and on unverified domain assumptions about label correctness, source dataset accuracy, and the reliability of an LLM judge. These choices are not derived from external principles and are not calibrated against independent expert agreement, so the reader pays for them upstream.

free parameters (5)
  • Train/Val/Test split (60/20/20)
    Arbitrary split adopted without justification or leakage analysis; all fine-tuning and evaluation claims depend on it.
  • Image resolution (224x224)
    Chosen for consistency; fine-grained disease symptoms may be unresolvable at this size, which affects the visual verifiability of labels.
  • Seven-task taxonomy and three-level hierarchy
    Hand-designed evaluation structure (Basic Perception, Pathology Understanding, Expert Diagnosis) not derived from agronomic or psychometric principles.
  • GPT-score scale with five ordinal levels
    The semantic grading rubric is a hand-set ordinal scale; no calibration against human expert ratings is reported.
  • Number of MCQ options (4) and one OEQ per image merging all symptoms
    These choices shape the difficulty of the benchmark and the number of QA pairs but are not justified or varied experimentally.
assumptions (5)
  • domain assumption The human annotation pipeline yields correct ground-truth labels for disease, symptom, pathogen, and scientific name.
    Invoked in §3.2 Stage 2; no inter-annotator agreement statistics or expert validation audits are provided.
  • domain assumption Symptoms are visually identifiable from 224x224 RGB images.
    Stage 2 says mislabeled data were removed only if symptoms are not visually identifiable; this conflates image resolvability with label correctness.
  • domain assumption Gemini 2.0 Flash (or GPT) as LLM judge reliably maps semantic correctness to the 1-5 GPT score.
    Invoked in §4.1 and §4.4; no validation of judge agreement with human ratings is reported.
  • domain assumption LeafNet and TOM2024 source image labels are accurate enough to serve as a benchmark backbone.
    Stage 1 adopts 89.69% of images from LeafNet (the authors' own prior dataset) without independently re-verifying source labels.
  • domain assumption Zero-shot failures reflect model capability rather than benchmark artifacts such as ambiguous questions or option leakage.
    The paper interprets low zero-shot scores as diagnostic gaps (§4.2), but no sanity-check baselines such as random guess or language-prior-only are reported.
invented entities (3)
  • TomaMMU dataset independent evidence
    purpose: Provides 28,808 tomato leaf images with 213,119 claimed VQA pairs for fine-tuning and evaluation.
    Released on HuggingFace; a falsifiable artifact, though its exact contents could not be independently verified during review.
  • TomaBench benchmark independent evidence
    purpose: Defines the three-level, seven-task evaluation protocol for measuring VLM tomato disease understanding.
    Published as a subset of TomaMMU; reproducible if the dataset URL and split definitions are correct.
  • TomaLLaVA fine-tuned model independent evidence
    purpose: Demonstrates that fine-tuning on TomaMMU boosts MCQ accuracy to 96.09%.
    It is a trained model artifact, not a postulated physical entity, and its reported performance depends on the benchmark quality.

how reviews work

0 comments
Cite this review

Pith. "Pith review of TomaMMU: A Comprehensive Multimodal Understanding Benchmark for Tomato Leaf Diseases." pith.science (2026). https://pith.science/paper/44UZ4NE6

@misc{pith2026260808727,
  author       = {Pith},
  title        = {Pith review of: TomaMMU: A Comprehensive Multimodal Understanding Benchmark for Tomato Leaf Diseases},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/44UZ4NE6}},
  note         = {Machine review of arXiv:2608.08727}
}
read the original abstract

To address this gap, we introduce TomaMMU, a large-scale Tomato leaf disease MultiModal Understanding dataset, alongside TomaBench, a benchmark for evaluating VLMs on tomato disease understanding. TomaMMU comprises 28,808 high-quality images spanning 15 categories and 213,119 human-annotated visual question-answer pairs, generated through a three-stage pipeline comprising Data Collection, Human Annotation, and Question-Answer Generation. Building on this foundation, TomaBench organizes seven agricultural tasks into a hierarchical three-level taxonomy spanning Basic Perception, Pathology Understanding, and Expert Diagnosis, which together enable systematic evaluation from low-level visual recognition to high-level diagnostic reasoning. The tasks assess visual symptom recognition, taxonomic relationships, and diagnostic reasoning, offering a comprehensive view of how well models grasp plant pathology. Our results pronounced gaps in fine-grained recognition and factually grounded reasoning with 14 state-of-the-art VLMs, consistently underperforming on both challenging MCQs and open-ended questions. These results suggest that current VLMs struggle to translate visual perception into reliable diagnostic knowledge, motivating the need for targeted domain adaptation. Simple fine-tuning on TomaMMU substantially narrows this gap, boosting accuracy on challenging MCQs to 96.09%, outperforming recent VLMs, and pointing toward promising directions for future work. All data and code is available in https://huggingface.co/datasets/enalis/TomaMMU.

Figures

Figures reproduced from arXiv: 2608.08727 by the authors.

Figure 1
Figure 1. The TomaMMU Curation and Benchmarking Pipeline. a, Raw image data of tomato leaf types were collected from in-the-wild sources and in-house collections. b, Humans participated in manual labeling and classification of disease symptoms. A metadata set with Train/Val/Test sets proportionally divided into 60% / 20% / 20% portions. c, Question and Answer Dataset. From metadata-image pairs, we developer TomaBench, a curat… view at source ↗
Figure 2
Figure 2. Visualization of the question tasks distribution in VQA benchmark dataset. Each colored segment represents one of seven question types. The dataset includes HDC, SI, DC, LC, CSI, SNC and PC. Stage 3: QA Generation For rigorous evaluation, TomaMMU adopts MCQs and OEQs. Using the dataset, a small, balanced subset, TomaBench, is selected for evaluation, while the remaining data is used for fine-tuning and training. All… view at source ↗
Figure 3
Figure 3. Visualization of key components of TomaMMU. a, Distribution of number classes of disease in the TomaMMU. b, TomaMMU data sources. c, Number of VQA on different question types. model to predict. With OEQs, the models are set up to offer short answers and uses Gemini 2.0 Flash [4] as a judge to validate generated answers and ground truth labels to specify semantic accuracy and correctness of VLMs. 3.3 Additional Prope… view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Summary of model results on MCQs and OEQs. a, Performance of various foundation models across seven tomato disease understanding tasks (pink indicates Acc, and blue denotes F1). Performance of Top-5 SOTA models on MCQs, showing that TomaLLaVA maintains the highest over…
Figure 5
Figure 5. Figure 5: Average evaluation scores on MCQs. Fine-tuning TomaLLaVA significantly en￾hances its understanding of agricultural knowledge, demonstrating the effectiveness of our development set compared with four state-of-the-art models. Fine-tuning Performance [PITH_FULL_IMAGE:fi…
Figure 6
Figure 6. Figure 6: The effectiveness of TomaLLaVA fine-tuning on OEQ examples. After simple fine-tuning, TomaLLaVA can accurately identify issues that SCOLD [28] and Gemini 2.0 Flash [4] failed to recognize in zero-shot scenarios. focus more on the background than on the main object, as …

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

49 extracted references · 25 canonical work pages

  1. [1]

    48550/ARXIV.2502.02737,https://arxiv.org/abs/2502.027379, 10, 11

    Allal, L.B., Lozhkov, A., Bakouch, E., Blázquez, G.M., Penedo, G., Tunstall, L., Marafioti, A., Kydlíček, H., Lajarín, A.P., Srivastav, V., Lochner, J., Fahlgren, C., Nguyen, X.S., Fourrier, C., Burtenshaw, B., Larcher, H., Zhao, H., Zakka, C., Morlon, M., Raffel, C., von Werra, L., Wolf, T.: Smollm2: When smol goes big TomaMMU Benchmark for Tomato Leaf D...

    arXiv doi:10.48550/arxiv.2502.02737 2025

  2. [2]

    23404,https://arxiv.org/abs/2511.234049, 10, 11, 13

    Amini, A., Banaszak, A., Benoit, H., Böök, A., Dakhran, T., Duong, S., Eng, A., Fernandes, F., Härkönen, M., Harrington, A., Hasani, R., Karwa, S., Khrustalev, Y., et al.: Lfm2 technical report (2025).https://doi.org/10.48550/ARXIV.2511. 23404,https://arxiv.org/abs/2511.234049, 10, 11, 13

    doi:10.48550/arxiv.2511.23404 2025

  3. [3]

    Appiah, O., Hackman, K.O., Diallo, B.A.A., Ogunjobi, K.O., Ouedraogo, V., Bebe, M., SON, D.: Tom2024 (2024).https://doi.org/10.17632/3D4YG89RTR.1, https://data.mendeley.com/datasets/3d4yg89rtr/16

    work page doi:10.17632/3d4yg89rtr.1 2024

  4. [4]

    Frontiers in Artificial Intelligence8(Mar 2025)

    Balestri, R.: Gender and content bias in large language models: a case study on google gemini 2.0 flash experimental. Frontiers in Artificial Intelligence8(Mar 2025). https://doi.org/10.3389/frai.2025.1558696 , http://dx.doi.org/10. 3389/frai.2025.15586968, 9, 10, 11, 13, 14

    work page arXiv 2025

  5. [5]

    Archives of Virology142(7), 1391–1406 (1997)

    Czosnek, H., Laterrot, H.: A worldwide survey of tomato yellow leaf curl viruses. Archives of Virology142(7), 1391–1406 (1997). https://doi.org/10.1007/ s007050050168,http://dx.doi.org/10.1007/s0070500501681

    work page doi:10.1007/s0070500501681 1997

  6. [6]

    Computers and Electronics in Agriculture234, 110128 (2025)

    George, R., Thuseethan, S., Ragel, R.G., Mahendrakumaran, K., Nimishan, S., Wimalasooriya, C., Alazab, M.: Past, present and future of deep plant leaf disease recognition: A survey. Computers and Electronics in Agriculture234, 110128 (2025). https://doi.org/10.1016/j.compag.2025.110128, http://dx.doi.org/ 10.1016/j.compag.2025.1101282

    work page arXiv 2025

  7. [7]

    GLASSCOCK,W.H.,WARE,W.M.:Alternariasolaniontomato.Nature154(3916), 642–642 (Nov 1944).https://doi.org/10.1038/154642a0, http://dx.doi.org/ 10.1038/154642a01

    work page doi:10.1038/154642a0 1944

  8. [8]

    https://doi.org/10.48550/ARXIV.1612.00837, https://arxiv.org/abs/ 1612.008373

    Goyal, Y., Khot, T., Summers-Stay, D., Batra, D., Parikh, D.: Making the v in vqa matter: Elevating the role of image understanding in visual question answering (2016). https://doi.org/10.48550/ARXIV.1612.00837, https://arxiv.org/abs/ 1612.008373

    arXiv doi:10.48550/arxiv.1612.00837 2016

Show all 49 references
  1. [9]

    Nature479(7374), 487–492 (Nov 2011).https://doi.org/10.1038/ nature10640,http://dx.doi.org/10.1038/nature106401

    Grbić, M., Van Leeuwen, T., Clark, R.M., Rombauts, S., Rouzé, P., Grbić, V., Osborne, E.J., et al.: The genome of tetranychus urticae reveals herbivorous pest adaptations. Nature479(7374), 487–492 (Nov 2011).https://doi.org/10.1038/ nature10640,http://dx.doi.org/10.1038/nature106401

    2011 doi

  2. [10]

    73–82 (2018)

    Gupta, A., Kamalachandran, D., Longchar, B., Senthil-Kumar, M.: Impact of soil moisture regimes on wilt disease in tomatoes: Current understanding p. 73–82 (2018). https://doi.org/10.1016/b978-0-12-813066-7.00005-x , http://dx. doi.org/10.1016/B978-0-12-813066-7.00005-X1

    2018 doi

  3. [11]

    48550/ARXIV.2106.09685,https://arxiv.org/abs/2106.0968513

    Hu, E.J., Shen, Y., Wallis, P., Allen-Zhu, Z., Li, Y., Wang, S., Wang, L., Chen, W.: Lora: Low-rank adaptation of large language models (2021).https://doi.org/10. 48550/ARXIV.2106.09685,https://arxiv.org/abs/2106.0968513

    2021 doi arXiv

  4. [12]

    1902.09506,https://arxiv.org/abs/1902.095063

    Hudson, D.A., Manning, C.D.: Gqa: A new dataset for real-world visual reasoning and compositional question answering (2019).https://doi.org/10.48550/ARXIV. 1902.09506,https://arxiv.org/abs/1902.095063

    2019 doi arXiv

  5. [13]

    Frontiers in Plant Science15(Mar 2024).https://doi.org/ 10.3389/fpls.2024.1356260, http://dx.doi.org/10.3389/fpls.2024.1356260 2

    Jafar, A., Bibi, N., Naqvi, R.A., Sadeghi-Niaraki, A., Jeong, D.: Revolutionizing agriculture with artificial intelligence: plant disease detection methods, applications, and their limitations. Frontiers in Plant Science15(Mar 2024).https://doi.org/ 10.3389/fpls.2024.1356260, ...

    2024

  6. [14]

    Journal of King Saud 16 Truong et al

    Khan, S.U., Alsuhaibani, A., Alabduljabbar, A., Almarshad, F., Altherwy, Y.N., Akram, T.: A review on automated plant disease detection: motivation, limitations, challenges, and recent advancements for future research. Journal of King Saud 16 Truong et al. University Computer ...

    2025 doi

  7. [15]

    In: Text Summarization Branches Out

    Lin, C.Y.: ROUGE: A package for automatic evaluation of summaries. In: Text Summarization Branches Out. pp. 74–81. Association for Computational Linguistics, Barcelona, Spain (Jul 2004),https://aclanthology.org/W04-1013/9

    2004

  8. [16]

    org/10.48550/ARXIV.2304.08485, https://arxiv.org/abs/2304.08485 9, 10, 11

    Liu, H., Li, C., Wu, Q., Lee, Y.J.: Visual instruction tuning (2023).https://doi. org/10.48550/ARXIV.2304.08485, https://arxiv.org/abs/2304.08485 9, 10, 11

    2023 doi arXiv

  9. [17]

    Liu, X., Liu, Z., Hu, H., Chen, Z., Wang, K., Wang, K., Lian, S.: A Multimodal Benchmark Dataset and Model for Crop Disease Diagnosis, p. 157–170. Springer Nature Switzerland (Oct 2024). https://doi.org/10.1007/978-3-031-73016- 0_10,http://dx.doi.org/10.1007/978-3-031-73016-0_104, 8

    2024 doi

  10. [18]

    Manik, M.M.H., Wang, G.: Unified deployment-aware evaluation of open reasoning language models (2026).https://doi.org/10.48550/ARXIV.2604.07035, https: //arxiv.org/abs/2604.070359, 10, 11, 12

    2026 doi arXiv

  11. [19]

    Frontiers in Plant Science7(2016).https://doi.org/10

    Mohanty, S.P., Hughes, D.P., Salathé, M.: Using deep learning for image-based plant disease detection. Frontiers in Plant Science7(2016).https://doi.org/10. 3389/fpls.2016.01419,http://dx.doi.org/10.3389/fpls.2016.014192

    2016

  12. [20]

    IEEE Access11, 35398–35410 (2023)

    Moupojou, E., Tagne, A., Retraint, F., Tadonkemwa, A., Wilfried, D., Tapamo, H., Nkenlifack, M.: Fieldplant: A dataset of field plant images for plant disease detection and classification with deep learning. IEEE Access11, 35398–35410 (2023). https://doi.org/10.1109/access.202...

    2023

  13. [21]

    Naik, A., Chaudhary, A.: Tomato Mosaic Virus, p. 329–347. Springer Nature Switzerland (2025). https://doi.org/10.1007/978-3-031-81884-4_21 , http: //dx.doi.org/10.1007/978-3-031-81884-4_211

    2025 doi

  14. [22]

    Nawaz, U., Awais, M., Gani, H., Naseer, M., Khan, F., Khan, S., Anwer, R.M.: Agriclip: Adapting clip for agriculture and livestock via domain-specialized cross- model alignment (2024).https://doi.org/10.48550/ARXIV.2410.01407, https: //arxiv.org/abs/2410.014075

    2024 doi arXiv

  15. [23]

    https://doi.org/10.1002/9781118728475.ch13 , http://dx.doi.org/10.1002/ 9781118728475.ch131

    Nowicki, M., Kozik, E.U., Foolad, M.R.: Late blight of tomato (Oct 2013). https://doi.org/10.1002/9781118728475.ch13 , http://dx.doi.org/10.1002/ 9781118728475.ch131

    2013 doi

  16. [24]

    https://doi.org/10.1079/cabicompendium.56981 , http://dx.doi.org/ 10.1079/cabicompendium.569811

    Osdaghi, E.: Xanthomonas vesicatoria (bacterial spot of tomato and pepper) (Dec 2020). https://doi.org/10.1079/cabicompendium.56981 , http://dx.doi.org/ 10.1079/cabicompendium.569811

    2020 doi

  17. [25]

    Horticulturae10(12), 1299 (Dec 2024)

    Pandey, A., Paudel, R., Adhikari, T.B., Panthee, D.R., Louws, F.J.: Septo- ria leaf spot of tomatoes: Historical insights, present challenges, and future prospects. Horticulturae10(12), 1299 (Dec 2024). https://doi.org/10.3390/ horticulturae10121299, http://dx.doi.org/10.3390/...

    2024 doi

  18. [26]

    Journal of Information and Telecommunication9(3), 335–356 (Jan 2025)

    Quach, L.D., Quoc Khang, N., Thai-Nghe, N., Nguyen, C.N.: Xai-bo: an architecture using grad-cam technique to evaluate bayesian optimization algorithms on deep learning models. Journal of Information and Telecommunication9(3), 335–356 (Jan 2025). https://doi.org/10.1080/247518...

    2025

  19. [27]

    https://doi.org/10.48550/ARXIV.2602.13662, https://arxiv.org/abs/ 2602.136623, 4, 6 TomaMMU Benchmark for Tomato Leaf Diseases 17

    Quoc, K.N., Dao, P.D., Quach, L.D.: Leafnet: A large-scale dataset and comprehen- sive benchmark for foundational vision-language understanding of plant diseases (2026). https://doi.org/10.48550/ARXIV.2602.13662, https://arxiv.org/abs/ 2602.136623, 4, 6 TomaMMU Benchmark for T...

    2026 doi

  20. [28]

    Quoc, K.N., Thu, L.L.T., Quach, L.D.: A vision-language foundation model for leaf disease identification (2025).https://doi.org/10.48550/ARXIV.2505.07019, https://arxiv.org/abs/2505.070195, 9, 10, 13, 14

    2025 doi

  21. [29]

    https://doi.org/10

    Radford, A., Kim, J.W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., Sastry, G., Askell, A., Mishkin, P., Clark, J., Krueger, G., Sutskever, I.: Learning transferable visual models from natural language supervision (2021). https://doi.org/10. 48550/ARXIV.2103.00020,https://a...

    2021 doi arXiv

  22. [30]

    International Journal on Digital Libraries23(3), 289–301 (2022)

    Saikh, T., Ghosal, T., Mittal, A., Ekbal, A., Bhattacharyya, P.: Scienceqa: a novel resource for question answering on scholarly articles. International Journal on Digital Libraries23(3), 289–301 (2022). https://doi.org/10.1007/s00799-022-00329-y, http://dx.doi.org/10.1007/s00...

    2022 doi

  23. [31]

    Journal of Industrial Information Integration38, 100572 (Mar 2024)

    Sajitha, P., Andrushia, A.D., Anand, N., Naser, M.: A review on machine learning and deep learning image-based plant disease classification for industrial farming systems. Journal of Industrial Information Integration38, 100572 (Mar 2024). https://doi.org/10.1016/j.jii.2024.10...

    2024

  24. [32]

    https://doi.org/10.48550/ARXIV.2508.17117 , https://arxiv

    Sakib, S.N., Haque, N., Hossain, M.Z., Arman, S.E.: Plantvillagevqa: A visual question answering dataset for benchmarking vision-language models in plant science (2025). https://doi.org/10.48550/ARXIV.2508.17117 , https://arxiv. org/abs/2508.171176

    2025 doi arXiv

  25. [33]

    Acta Horticulturae (808), 25–28 (Jan 2009)

    Schlub, R., Smith, L., Datnoff, L., Pernezny, K.: An overview of target spot of tomato caused by corynespora cassiicola. Acta Horticulturae (808), 25–28 (Jan 2009). https://doi.org/10.17660/actahortic.2009.808.1 , http://dx.doi.org/10. 17660/ActaHortic.2009.808.11

    2009 doi

  26. [34]

    Ecological Modelling474, 110166 (Dec 2022)

    Senthil Pandi, S., Senthilselvi, A., Gitanjali, J., ArivuSelvan, K., Gopal, J., Vellingiri, J.: Rice plant disease classification using dilated convolutional neural network with global average pooling. Ecological Modelling474, 110166 (Dec 2022). https: //doi.org/10.1016/j.ecol...

    2022

  27. [35]

    In: Proceedings of the 7th ACM IKDD CoDS and 25th COMAD

    Singh, D., Jain, N., Jain, P., Kayal, P., Kumawat, S., Batra, N.: Plantdoc: A dataset for visual plant disease detection. In: Proceedings of the 7th ACM IKDD CoDS and 25th COMAD. p. 249–253. CoDS COMAD 2020, ACM (Jan 2020).https://doi. org/10.1145/3371158.3371196, http://dx.do...

    2020

  28. [36]

    Annals of Applied Biology62(2), 319–327 (Oct 1968)

    SINHA, A.K., WOOD, R.K.S.: Studies on the nature of resistance in tomato plants to verticillium albo-atrum. Annals of Applied Biology62(2), 319–327 (Oct 1968). https://doi.org/10.1111/j.1744-7348.1968.tb02827.x , http://dx.doi.org/ 10.1111/j.1744-7348.1968.tb02827.x1

    1968

  29. [37]

    Srinivas, C., Nirmala Devi, D., Narasimha Murthy, K., Mohan, C.D., Laksh- meesha, T., Singh, B., Kalagatur, N.K., Niranjana, S., Hashem, A., Alqarawi, A.A., Tabassum, B., Abd_Allah, E.F., Chandra Nayaka, S., Srivastava, R.K.: Fusarium oxysporum f. sp. lycopersici causal agent ...

    2019 doi

  30. [38]

    Stevens, S., Wu, J., Thompson, M.J., Campolongo, E.G., Song, C.H., Carlyn, D.E., Dong, L., Dahdul, W.M., Stewart, C., Berger-Wolf, T., Chao, W.L., Su, Y.: Bioclip: A vision foundation model for the tree of life (2023).https://doi.org/10.48550/ ARXIV.2311.18803,https://arxiv.or...

    2023 doi arXiv

  31. [39]

    Tong, S., Liu, Z., Zhai, Y., Ma, Y., LeCun, Y., Xie, S.: Eyes wide shut? exploring the visual shortcomings of multimodal llms (2024).https://doi.org/10.48550/ ARXIV.2401.06209,https://arxiv.org/abs/2401.062094

    2024 doi arXiv

  32. [40]

    Tschannen, M., Gritsenko, A., Wang, X., Naeem, M.F., Alabdulmohsin, I., Parthasarathy, N., Evans, T., Beyer, L., Xia, Y., Mustafa, B., Hénaff, O., Harm- sen, J., Steiner, A., Zhai, X.: Siglip 2: Multilingual vision-language encoders with improved semantic understanding, locali...

    2025 doi arXiv

  33. [41]

    Nature Food 4(11), 941–948 (Nov 2023)

    Tzachor, A., Devare, M., Richards, C., Pypers, P., Ghosh, A., Koo, J., Johal, S., King, B.: Large language models and agricultural extension services. Nature Food 4(11), 941–948 (Nov 2023). https://doi.org/10.1038/s43016- 023- 00867- x , http://dx.doi.org/10.1038/s43016-023-00867-x3

    2023 doi

  34. [42]

    Wang, W., Gao, Z., Gu, L., Pu, H., Cui, L., Wei, X., Liu, Z., Jing, L., Ye, S., Shao, J., et al.: Internvl3: Exploring advanced training and test-time recipes for open- source multimodal models (2025).https://doi.org/10.48550/ARXIV.2504.10479, https://arxiv.org/abs/2504.104799, 10, 11

    2025 doi arXiv

  35. [43]

    https://doi.org/10.48550/ARXIV.2505.09388 , https://arxiv

    Yang, A., Li, A., Yang, B., Zhang, B., Hui, B., Zheng, B., Yu, B., Gao, C., Huang, C., Lv, C., Zheng, C., Liu, D., Zhou, F., Huang, F., Hu, F., et al.: Qwen3 technical report (2025). https://doi.org/10.48550/ARXIV.2505.09388 , https://arxiv. org/abs/2505.093889, 10, 11

    2025 doi arXiv

  36. [44]

    https://doi.org/10.48550/ARXIV.2412.15115 , https://arxiv.org/abs/2412

    Yang, A., Yang, B., Zhang, B., Hui, B., Zheng, B., Yu, B., Li, C., Liu, D., Huang, F., Wei, H., Lin, H., Yang, J., Tu, J., Zhang, J., et al.: Qwen2.5 technical report (2024). https://doi.org/10.48550/ARXIV.2412.15115 , https://arxiv.org/abs/2412. 151159, 10, 11

    2024 doi arXiv

  37. [45]

    Yue, X., Ni, Y., Zhang, K., Zheng, T., Liu, R., Zhang, G., Stevens, S., Jiang, D., Ren, W., Sun, Y., Wei, C., Yu, B., Yuan, R., Sun, R., Yin, M., Zheng, B., Yang, Z., Liu, Y., Huang, W., Sun, H., Su, Y., Chen, W.: Mmmu: A massive multi-discipline multimodal understanding and r...

    2023 doi arXiv

  38. [46]

    In: The Thirty-eight Conference on Neural Information Processing Systems Datasets and Benchmarks Track (2024),https://openreview.net/forum?id=hMj6jZ6JWU8

    Zhang, H., Sun, J., Chen, R., Liu, W., Yuan, Z., Zheng, X., Wang, Z., Yang, Z., Yan, H., Zhong, H.S., Wang, X., Ouyang, W., Yang, F., Dong, N.: Empowering and assessing the utility of large language models in crop science. In: The Thirty-eight Conference on Neural Information ...

    2024

  39. [47]

    Horticultural Plant Journal8(6), 667–675 (Nov 2022).https://doi.org/10.1016/j.hpj.2022.04.008, http://dx.doi.org/10.1016/j.hpj.2022.04.0081

    Zhao, T., Pei, T., Jiang, J., Yang, H., Zhang, H., Li, J., Xu, X.: Understanding the mechanisms of resistance to tomato leaf mold: A review. Horticultural Plant Journal8(6), 667–675 (Nov 2022).https://doi.org/10.1016/j.hpj.2022.04.008, http://dx.doi.org/10.1016/j.hpj.2022.04.0081

    2022 doi

  40. [48]

    Zhou, Y., Ryo, M.: AgriBench: A Hierarchical Agriculture Benchmark for Multi- modal Large Language Models, p. 207–223. Springer Nature Switzerland (2025). https://doi.org/10.1007/978- 3- 031- 91835- 3_14 , http://dx.doi.org/10. 1007/978-3-031-91835-3_144

    2025 doi

  41. [49]

    Frontiers in Plant Science16(2025)

    Zhu, H., Qin, S., Su, M., Lin, C., Li, A., Gao, J.: Harnessing large vision and language models in agriculture: a review. Frontiers in Plant Science16(2025). https://doi.org/10.3389/fpls.2025.1579355 , http://dx.doi.org/10.3389/ fpls.2025.15793552 TomaMMU Benchmark for Tomato ...

    2025

Pith tools

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