Try:droughtFusariumnitrogen fertilizerArabidopsisseed germination

“parameter count”

576 matches

Claim types

Direction of the findings

plus 106 non-evaluative

Over time

1960 ’70 ’80 ’90 2000 ’10 ’20

Refine

More filters

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

OW-YOLO parameter count decreased by 49.2%.

Outcome: parameter counte0
Direction: decreasede0
Value: 49.2%e0

Not reportedTaxon

Verbatim

e0
“model’s parameter count decreased by 49.2%”
✓ located verbatim in the source · abstract, characters 1294–1336

Title

OW-YOLO: An Improved YOLOv8s Lightweight Detection Method for Obstructed Walnuts

Haoyu Wang; Lijun Yun; Chenggui Yang; et al. Agriculture 2025. doi:10.3390/agriculture15020159

Other claims from this paper (5)

Method / resource OW-YOLO achieved 83.6% mAP@0.5.
Method / resource OW-YOLO achieved 53.7% mAP@[0.5:0.95].
Method / resource OW-YOLO achieved an F1 score of 77.9%.
Method / resource OW-YOLO weight file size was reduced by 48.1%.
Method / resource OW-YOLO computational load dropped by 37.3%.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

YOLO-MSNet reduced parameter count by 21.5%.

Taxon: PomegranatePunica granatumt0
Outcome: parameter counte0
Direction: reductione0
Value: 21.5%e0

Verbatim

e0
“a 21.5% reduction in parameter count”
✓ located verbatim in the source · abstract, characters 1454–1490

t0
“Pomegranate Fruit”
✓ located verbatim in the source · title, characters 46–63

Title

YOLO-MSNet: Real-Time Detection Algorithm for Pomegranate Fruit Improved by YOLOv11n

Liang Xu; Bing Li; Xue Fu; et al. Agriculture 2025. doi:10.3390/agriculture15101028

Other claims from this paper (3)

Method / resource YOLO-MSNet raised mAP50 by 1.7% against comparison models.
Method / resource YOLO-MSNet decreased model size by 21.8%.
Method / resource YOLO-MSNet had superior detection accuracy to mainstream YOLO models.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

Modified-DCNet has a parameter count of only 5.2 M, a smaller parameter scale.

Taxon: carrotDaucus carotat0
Outcome: parameter counte3
Direction: smallere2
Value: 5.2 Me3

Verbatim

t0
“carrot appearance quality”
✓ located verbatim in the source · title, characters 12–37

e2
“has higher recognition accuracy and smaller parameter scales”
✓ located verbatim in the source · abstract, characters 1251–1311

e3
“the parameter count is the only 5.2 M”
✓ located verbatim in the source · abstract, characters 1119–1156

Title

Identifying carrot appearance quality by an improved dense CapNet

Hongfei Zhu; Lianhe Yang; Yuxin Sun; et al. Journal of Food Process Engineering 2020. doi:10.1111/jfpe.13586

Other claims from this paper (1)

Method / resource Modified-DCNet recognized carrot appearance quality with 97.50% accuracy at 227 × 227 pixel input, with higher recognition accuracy.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

The model reduced parameter count by 89.36%.

Taxon: AppleMalus domesticat0
Outcome: parameter counte0
Direction: reducese0
Value: 89.36%e0

Verbatim

e0
“reduces the parameter count by 89.36%”
✓ located verbatim in the source · abstract, characters 1241–1278

t0
“Apple Detection”
✓ located verbatim in the source · title, characters 19–34

Title

Real-Time Accurate Apple Detection Based on Improved YOLOv8n in Complex Natural Environments

Mingjie Wang; Fuzhong Li Plants 2025. doi:10.3390/plants14030365

Other claims from this paper (5)

Method / resource The improved model raised apple detection mAP by 2.90% to 88.90% versus YOLOv8n.
Method / resource Detection speed reached 220 FPS, 30.55% faster than YOLOv8n.
Method / resource The model reduced FLOPs by 2.47%.
Method / resource The model outperformed the listed object detection algorithms in mAP and speed.
Method / resource Smartphone deployment achieved 40 FPS, a 26.93% gain over YOLOv8n deployment.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

The model reduced parameter count by 48.89%.

Outcome: parameter counte0
Direction: reducede0
Value: 48.89%e0

Verbatim

e0
“the parameter count is reduced by 48.89%”
✓ located verbatim in the source · abstract, characters 1283–1323

Title

The Development of a Lightweight DE-YOLO Model for Detecting Impurities and Broken Rice Grains

Zhenwei Liang; Xingyue Xu; Deyong Yang; et al. Agriculture 2025. doi:10.3390/agriculture15080848

Other claims from this paper (4)

Method / resource DE-YOLO achieved 97.55% mAP, 2.9% higher than YOLOX.
Method / resource The model had a recall rate of 94.46%.
Method / resource The model had an F1 value of 0.96.
Method / resource The model reduced GFLOPS by 46.33%.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

CSPPC reduced parameter count and computational complexity while maintaining mAP.

Outcome: parameter count and computational complexitye0
Direction: reducinge0

Not reportedTaxonComparatorValue

Verbatim

e0
“effectively reducing both parameter count and computational complexity while maintaining mAP values”
✓ located verbatim in the source · abstract, characters 829–928

Title

YOLO-MECD: Citrus Detection Algorithm Based on YOLOv11

Yue Liao; Lerong Li; Huiqiang Xiao; et al. Agronomy 2025. doi:10.3390/agronomy15030687

Other claims from this paper (15)

Method / resource Replacing C2PSA with EMA improved citrus detection accuracy and reduced model parameters.
Method / resource MPDIoU improved bounding-box accuracy and convergence.
Method / resource Reducing backbone convolutions improved small-object detection and decreased parameters.
Method / resource YOLO-MECD precision was 84.4%, 0.2 points above baseline.
Method / resource YOLO-MECD recall was 73.3%.
Method / resource YOLO-MECD mAP was 81.6%.
Method / resource Against baseline, precision improved 0.2 percentage points.
Method / resource Against baseline, recall improved 4.1 percentage points.
Method / resource Against baseline, mAP improved 3.9 percentage points.
Method / resource Model parameters fell from 9,413,574 to 2,297,334, a 75.6% decrease.
Method / resource Model size fell from 18.2 MB to 4.66 MB, a 74.4% reduction.
Method / resource mAP exceeded YOLOv8s by 3.8%.
Method / resource mAP exceeded YOLOv9s by 3.2%.
Method / resource mAP exceeded YOLOv10s by 5.5%.
Method / resource The model detected pomelos and kumquats well.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

The model reduced parameter count by 12.9%.

Taxon: PeanutArachis hypogaeat0
Outcome: Parameter counte0
Direction: reducede0
Value: 12.9%e0

Not reportedComparator

Verbatim

e0
“Parameter count and computational complexity are reduced by 12.9% and 20.7%”
✓ located verbatim in the source · abstract, characters 1130–1205

t0
“Peanut”
✓ located verbatim in the source · title, characters 34–40

Title

A Lightweight Detection Model for Peanut Leaf Diseases

Zongle Xiao; Jie Zhou; Xiaoxiao Li; et al. Agronomy 2026. doi:10.3390/agronomy16090864

Other claims from this paper (7)

Method / resource YOLOv8-MSDH precision on the peanut disease dataset was 90.14%, 3.5 percentage points above YOLOv8.
Method / resource YOLOv8-MSDH recall was 82.16%, 0.7 percentage points above YOLOv8.
Method / resource YOLOv8-MSDH mAP50 was 90.71%, 1.5 percentage points above YOLOv8.
Method / resource YOLOv8-MSDH mAP50-95 was 72.14%, 2.89 percentage points above YOLOv8.
Method / resource The model reduced computational complexity by 20.7%.
Method / resource The model operated at 509.95 FPS.
Method / resource The model showed high robustness across varying light conditions.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

The model parameter count fell to 29M.

Taxon: sesame seedlingsSesamum indicumt0
Outcome: parameter counte0
Direction: reducede0
Value: 29Me0

Verbatim

t0
“sesame seedlings”
✓ located verbatim in the source · title, characters 54–70

e0
“The model's parameter count was reduced to 29M”
✓ located verbatim in the source · abstract, characters 1186–1232

Title

Optimized DINO model for accurate object detection of sesame seedlings and weeds

Yong Wang; ShunFa Xu; ZhenYuan Ye; et al. Scientific Reports 2025. doi:10.1038/s41598-025-96826-6

Other claims from this paper (5)

Method / resource The modified DINO improved overall AP by 5.1% over original DINO.
Method / resource The model inference time decreased by 60%.
Method / resource The model computational cost decreased by 43.72%.
Method / resource Improved DINO reached 81.8% maximum AP, 5.6% above YOLOv7.
Method / resource The improved model reached 24 frames per second.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

The improved detector reduced parameter count by 85.6%.

Taxon: appleMalus domesticat0
Outcome: parameter counte0
Direction: reducese0
Value: 85.6%e0

Verbatim

e0
“improved lightweight YOLOv5s network reduces the parameter count by 85.6%”
✓ located verbatim in the source · abstract, characters 618–691

t0
“Apple”
✓ located verbatim in the source · title, characters 21–26

Title

Improved Lightweight Apple Object Detection Method Based on YOLOv5s

Yunfei Jia; Chang Liu 2023 5th International Conference on Frontiers Technology of Information and Computer (ICFTIC) 2023. doi:10.1109/icftic59930.2023.10456316

Other claims from this paper (4)

Method / resource The improved detector reduced computational load by 87.3%.
Method / resource Apple detection accuracy reached mAP@0.5 of 97.0%.
Method / resource Detection accuracy was 0.4% higher than the original YOLOv5s.
Method / resource Detection frame rate improved by 15 FPS relative to the original model.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

Parameter count was reduced by 46.8%.

Taxon: cropt0
Outcome: parameter counte0
Direction: reducinge0
Value: 46.8%e0

Verbatim

e0
“reducing the parameter count by 46.8%”
✓ located verbatim in the source · abstract, characters 1989–2026

t0
“crop disease identification tasks”
✓ located verbatim in the source · abstract, characters 304–337

Title

ConvTransNet-S: A CNN-Transformer Hybrid Disease Recognition Model for Complex Field Environments

Shangyun Jia; Guanping Wang; Hongling Li; et al. Plants 2025. doi:10.3390/plants14152252

Other claims from this paper (9)

Method / resource ConvTransNet-S reached 98.85% recognition accuracy on PlantVillage.
Method / resource ConvTransNet-S used 25.14 million parameters.
Method / resource ConvTransNet-S had a computational load of 3.762 GFLOPs.
Method / resource ConvTransNet-S inference time was 7.56 ms.
Method / resource ConvTransNet-S reached 88.53% accuracy on field scenes.
Method / resource Field accuracy exceeded EfficientNetV2 by 14.22%.
Method / resource Field accuracy exceeded Vision Transformer by 2.75%.
Method / resource Field accuracy exceeded Swin Transformer by 0.34%.
Method / resource Complex-background disease accuracy was up to 14.22% higher.

Lexical search (BM25). All words must match; if nothing does, any word. Organism names also match every spelling of the same organism. No semantic search yet.

“parameter count”

Claim types

Direction of the findings

Over time

Claim type

Research area

Polarity

Statistics