Try:droughtFusariumnitrogen fertilizerArabidopsisseed germination

“overall model performance”

380 matches

Claim types

Direction of the findings

plus 65 non-evaluative

Over time

1980 ’90 2000 ’10 ’20

Refine

More filters

Method / resourceobservedoriginal resultnegativemeasured

Claim · model-written summary

Overall LUE model performance was limited.

Outcome: overall performancee0
Qualifier: structural constraintse0

Not reportedComparatorValue

Verbatim

e0
“its overall performance was limited by structural constraints”
✓ located verbatim in the source · abstract, characters 1680–1741

Title

Simulation of Winter Wheat Gross Primary Productivity Incorporating Solar-Induced Chlorophyll Fluorescence

Xuegui Zhang; Yao Li; Xiaoya Wang; et al. Agronomy 2025. doi:10.3390/agronomy15092187

Other claims from this paper (5)

Method / resource LSM outperformed other machine learning models in both scenarios.
Method / resource Without SIF, LSM averaged R2 0.87 and exceeded individual models.
Method / resource Adding SIF improved the accuracy of all machine learning models.
Method / resource With SIF, LSM average R2 rose to 0.91.
Method / resource LUE model accuracy approached machine learning accuracy in some growth stages.

Method / resourceobservedoriginal resultpositivesimulated

Claim · model-written summary

Direction-specific biomass density improved overall model performance.

Outcome: overall model performancee0
Direction: improvede0

Not reportedTaxonComparatorValue

Verbatim

e0
“Accounting for changes in biomass density as a function of direction upwind of the tower improved the overall model performance”
✓ located verbatim in the source · abstract, characters 1374–1501

Title

Evaluation of forest canopy models for estimating isoprene emissions

Brian Lamb; Thomas Pierce; Dennis Baldocchi; et al. Journal of Geophysical Research Atmospheres 1996. doi:10.1029/96jd00056

Other claims from this paper (5)

Method / resource Predicted isoprene fluxes agreed among canopy models within about 20%.
Method / resource Models overestimated mean observed flux by about twofold.
Method / resource Models overestimated maximum observed flux by 30 to 50%.
Method / resource Reducing the emission factor and biomass density by 20% gave predicted means about 20% above observations.
Method / resource That adjustment gave predicted maxima about 25% below observations.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

The support vector machine model achieved over 97% overall performance.

Outcome: overall performancee3
Value: >97%e3

Not reportedTaxon

Verbatim

e3
“The overall performance of the model was >97%”
✓ located verbatim in the source · abstract, characters 1119–1164

Title

The use of infrared spectroscopy and machine learning tools for detection of Meloidogyne infestations

Ernesto San‐Blas; Gabriel Paba; Néstor Cubillán; et al. Plant Pathology 2020. doi:10.1111/ppa.13246

Other claims from this paper (8)

Observation Meloidogyne-infested tomato plants had different spectral patterns than uninfested plants.
Observation Meloidogyne-infested guava plants had different spectral patterns than uninfested plants.
Method / resource Principal component analysis diagnosed symptomless infested tomato plants after five weeks.
Method / resource Principal component analysis diagnosed symptomless infested guava plants after eight weeks.
Method / resource The support vector machine model had significantly higher total accuracy.
Method / resource The best infestation prediction occurred in the second week for tomatoes, halving diagnostic time.
Method / resource The best infestation prediction occurred in the fourth week for guavas, halving diagnostic time.
Method / resource The combined techniques reduced processing time from field to laboratory and avoided root and soil sampling.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

The model had better overall performance than other mainstream models.

Outcome: overall performancee0
Direction: much bettere0
Comparator: other mainstream modelse0

Not reportedTaxon

Verbatim

e0
“Our model’s overall performance is much better than other mainstream models.”
✓ located verbatim in the source · abstract, characters 1607–1683

Title

HR-YOLOv8: A Crop Growth Status Object Detection Method Based on YOLOv8

Jin Zhang; Wenzhong Yang; Zhifeng Lu; et al. Electronics 2024. doi:10.3390/electronics13091620

Other claims from this paper (4)

Method / resource HR-YOLOv8 improved average detection accuracy on the oilpalmuav dataset by 5.2% against baseline.
Method / resource HR-YOLOv8 improved average detection accuracy on strawberry ripeness data by 0.6%.
Method / resource HR-YOLOv8 used fewer parameters than baseline on the two crop datasets.
Method / resource The method improved small object detection ability.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

Parallel attention and parallel loss improved overall model performance.

Taxon: Jujube Forestt0
Outcome: overall performance of the modele0
Value: enhancinge0

Verbatim

e0
“parallel attention and parallel loss function in enhancing the overall performance of the model”
✓ located verbatim in the source · abstract, characters 1285–1380

t0
“Jujube Forest”
✓ located verbatim in the source · title, characters 10–23

Title

Enhancing Jujube Forest Growth Estimation and Disease Detection Using a Novel Diffusion-Transformer Architecture

Xiangyi Hu; Zhihao Zhang; Liping Zheng; et al. Plants 2024. doi:10.3390/plants13172348

Other claims from this paper (5)

Method / resource The model reached 95% precision for jujube forest disease detection.
Method / resource The model reached 92% recall for jujube forest disease detection.
Method / resource The model reached 93% accuracy for jujube forest disease detection.
Method / resource The model reached 94% F1-score for jujube forest disease detection.
Method / resource The model also outperformed comparators in jujube growth estimation.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

Model 2.0 had the best overall performance among three variants.

Outcome: overall performancee0
Direction: beste0
Comparator: Model 1.0, Model 2.0, and Model 3.0e0

Verbatim

e0
“By comparing Model 1.0, Model 2.0, and Model 3.0, it was found that the overall performance of Model 2.0 was best”
✓ located verbatim in the source · abstract, characters 1717–1830

Title

Precise Selection and Visualization of Maize Kernels Based on Electromagnetic Vibration and Deep Learning

Chengshun Zhao; Longzhe Quan; Hailong Li; et al. Transactions of the ASABE 2020. doi:10.13031/trans.13389

Other claims from this paper (8)

Method / resource Model 1.0 had the highest recall, 98.42%.
Method / resource Model 1.0 reached 97.27% mAP for excellent kernels only.
Method / resource Model 1.0 required fewer computer resources.
Method / resource Model 2.0 precision exceeded Faster R-CNN by 3.73%.
Method / resource Model 2.0 recall exceeded Faster R-CNN by 3.55%.
Method / resource Model 2.0 F1 exceeded Faster R-CNN by 3.79%.
Method / resource Model 2.0 false positive rate was reduced by 1.31% versus Faster R-CNN.
Method / resource Moderate model size gave the best kernel selection.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

The model had superior overall performance to mainstream object detection models.

Taxon: soybeanGlycine maxt0
Outcome: superior overall performancee0
Direction: superiore0

Not reportedComparator

Verbatim

e0
“YOLOv8-DML delivers superior overall performance”
✓ located verbatim in the source · abstract, characters 1485–1533

t0
“soybean leaf disease”
✓ located verbatim in the source · title, characters 12–32

Title

Research on soybean leaf disease recognition in natural environment based on improved Yolov8

Chen Chen; Xiaolei Lu; Lei He; et al. Frontiers in Plant Science 2025. doi:10.3389/fpls.2025.1523633

Other claims from this paper (3)

Method / resource YOLOv8-DML achieved mAP50 of 96.9%.
Method / resource YOLOv8-DML improved mAP50 by 1.8% over original YOLOv8.
Method / resource YOLOv8-DML reduced parameters by 18.6%.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

Model 1 had the best overall performance.

Taxon: cornZea mays subsp. mayst0
Outcome: overall performancee0
Direction: beste0
Comparator: Model 1e0

Verbatim

e0
“Although Model 1 had the best overall performance, it may not be feasible for growers to utilize a model with multiple terms.”
✓ located verbatim in the source · abstract, characters 1309–1434

t0
“corn grain yield prediction in Florida”
✓ located verbatim in the source · title, characters 10–48

Title

Irrigated corn grain yield prediction in Florida using active sensors and plant height

Diego A. H. de S． Leitão; Sudeep S． Sidhu; Winniefred D． Griffin; et al. Smart Agricultural Technology 2023. doi:10.1016/j.atech.2023.100276

Other claims from this paper (6)

Association N rate, plant height, NDVI and SPAD correlated with corn grain yield at r = 0.93, 0.91, 0.81 and 0.93, respectively.
Method / resource Models with multiple variables fit corn grain yield better than single-variable models.
Method / resource Model 15 explained 92.8% of corn grain yield variance.
Method / resource Model 15 had an RMSE of 1,315.685 kg ha−1.
Association N rate contributed the most to corn grain yield in every model.
Method / resource Model 15 could estimate corn grain yield in Florida using height at 60 DAP and NDVI at 32 DAP.

Method / resourceobservedoriginal resultno effectnot significant (as stated)simulated

Claim · model-written summary

Plant hydraulic modules alone do not significantly improve overall model performance under observed climate variability.

Outcome: overall model performancee0
Direction: cannot significantly improvee0

Not reportedComparator

Verbatim

e0
“when considering ecosystem responses to the observed climate variability, plant hydraulic modules alone cannot significantly improve the overall model performance”
✓ located verbatim in the source · abstract, characters 1567–1729

Title

When do plant hydraulics matter in terrestrial biosphere modelling?

Athanasios Paschalis; Martin G. De Kauwe; Manon Sabot; et al. Global Change Biology 2023. doi:10.1111/gcb.17022

Other claims from this paper (5)

Method / resource Models with plant hydraulics predict a slower onset of plant water stress as drought develops.
Method / resource Models with plant hydraulics produce diurnal water and carbon flux variability closer to observations.
Method / resource Including plant water capacitance improves modeled diurnal water and carbon flux dynamics.
Method / resource Representing permanent xylem damage produces a significant drought legacy effect that limits plant productivity during recovery.
Method / resource Plant hydraulic modules reproduce more realistic water and carbon dynamics.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

On the larger AMG HS dataset the model had the best overall detection performance.

Outcome: detection performancee0
Direction: best overalle0

Not reportedComparator

Verbatim

e0
“The proposed paradigm achieves the best overall detection performance while maintaining superior computational efficiency”
✓ located verbatim in the source · abstract, characters 1428–1549

Title

Precise leaf damage detection across diverse species and environments via a large-scale vision model

Wei Chen; Hao Ruan; Peng Zhou; et al. Frontiers in Plant Science 2026. doi:10.3389/fpls.2026.1788926

Other claims from this paper (7)

Method / resource The proposed model outperformed Unet, Unet++, and SwinUnet on coffee and black gram datasets.
Method / resource On coffee leaves the model achieved 78.31% IoU.
Method / resource On coffee leaves the model achieved 88.00% pixel accuracy.
Method / resource The model exceeded baseline Unet by over 10.5% in IoU.
Method / resource The architecture reduced inference time by approximately 93.6%, from 63.41s to 4.07s.
Method / resource On the larger dataset the model maintained superior computational efficiency.
Method / resource Interpretability analysis indicated that the backbone captures high-level lesion features.

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.

“overall model performance”

Claim types

Direction of the findings

Over time

Claim type

Research area

Polarity

Statistics