Truace
Subscribe
Good Space a ranking of the record

TheGood surrounding AI

101 gains match this view. 94% rest on at least one peer-reviewed study, and 17% clear the high-evidence bar. The latest was filed Oct 16, 2026.

Every gain in this view, one square each, shaded by the strength of its evidence. High 17 Moderate 84 Emerging 0
101 results
Show filters and sorting
Reset all
Sector: science

Matching records · page 2 of 4

31
Evidence-backed gainPeer-reviewedScience

Pairing amino acid and 3Di structural alignments in generative landscapes recovered known subfamilies and revealed evolutionary relationships in viral glycoproteins that sequence alone did not show.

Source article: A structure-aware generative AI framework for revealing functional relationships in protein families

Proceedings of the National Academy of Sciences
Reader signal

How should this claim be treated?

32
Evidence-backed gainPeer-reviewedScience

In a pilot, ChatGPT-4o produced 55 hypotheses on fungal factors and cervical carcinogenesis, and expert reviewers rated 63.6% as high quality, suggesting it can help seed research directions.

Source article: A pilot evaluation of ChatGPT-4o-assisted hypothesis generation for exploring potential links between fungal biology and cervical cancer

Antonie van Leeuwenhoek
Reader signal

How should this claim be treated?

33
Evidence-backed gainPeer-reviewedScience

Genomic selection models including machine learning algorithms predicted PFR resistance in watermelon, achieving high cross-validated accuracy in the RIL population and capturing top-performing lines.

Source article: Machine Learning and Statistical Models for Genomic Selection of Phytophthora Fruit Rot Resistance in Watermelon

Phytopathology®
Reader signal

How should this claim be treated?

34
Evidence-backed gainPeer-reviewedScience

Random Forest predicted atmospheric attenuation for free-space optical links more accurately than deep learning models on a 1500-sample synthetic Iraqi weather dataset, with dust concentration and visibility as top drivers and faster training and inference.

Source article: Machine Learning Outperforms Deep Learning for Atmospheric Attenuation Prediction in Free-Space Optical Communications under Iraqi Weather Conditions

Journal of Visualized Experiments
Reader signal

How should this claim be treated?

36
Evidence-backed gainPeer-reviewedScience

Interpretable CatBoost-based machine learning framework predicted GenX-induced biphasic dose-responses in zebrafish with high fidelity, enabling mechanistic ecological risk assessment of PFAS alternatives.

Source article: Predicting Hormesis Effects of GenX in Zebrafish via Interpretable Machine Learning: Insights From SHAP Analysis

IUBMB Life
Reader signal

How should this claim be treated?

37
Evidence-backed gainPeer-reviewedScience

AlphaFold DB scaled from 300k to over 214 million predicted structures and was integrated into primary resources like PDB and UniProt, expanding structure coverage for research use.

Source article: AlphaFold Protein Structure Database in 2024: providing structure coverage for over 214 million protein sequences

Nucleic Acids Research
Reader signal

How should this claim be treated?

38
Evidence-backed gainPeer-reviewedScience

By October 2023 testing, six generative chatbots including GPT-4 produced ideas judged equal in quality to human ideas, indicating they can function as assistants in creative work.

Source article: Artificial muses: Generative artificial intelligence chatbots have risen to human-level creativity

Journal of Creativity
Reader signal

How should this claim be treated?

39
Evidence-backed gainPeer-reviewedScience

We unfold the successive-interference cancellation (SIC) algorithm into a deep-learning network, where repeated stages of a single multitask network implement iterative signal estimation and interference cancellation that reflect the structural prior of SIC.

Source article: Code-multiplexed multi-frequency impedance cytometry with a unified deep-unfolding network

Microsystems & Nanoengineering
Reader signal

How should this claim be treated?

40
Evidence-backed gainPeer-reviewedScience

Genolator integrates DNA, amino acid and protein structure embeddings with natural language queries and answers questions about protein localization and function with high accuracy, outperforming general LLMs.

Source article: Genolator enables protein function interpretation using a multimodal large language model fusing genomic and structural interpretation with natural language interaction

Genome Biology
Reader signal

How should this claim be treated?

41
Evidence-backed gainPeer-reviewedScience

An interpretable SISSO symbolic-regression AI approach identified materials genes, including d-band center and particle diameter, that statistically describe selectivity of supported palladium-based alloy nanoparticles during hydrogenation of concentrated acetylene streams under elevated pressure and temperature.

Source article: Rethinking catalysis: interpretable AI and description of real-world conditions <i>via</i> materials genes

Faraday Discussions
Reader signal

How should this claim be treated?

43
Evidence-backed gainPeer-reviewedClimate

Conformal Sampling-derived MLIPs predicted reaction energies and barriers for CO2 hydrogenation on Cu(100) and Ni(100) with maximum errors of 0.05 eV and 0.03 eV, enabling rapid transfer from copper to nickel.

Source article: Carbon dioxide hydrogenation on copper and nickel catalysts <i>via</i> a conformal sampling approach

Faraday Discussions
Reader signal

How should this claim be treated?

44
Evidence-backed gainPeer-reviewedScience

Deep-learning model automatically segmented 52 anatomical regions on brain FDG PET alone and achieved MRI-comparable pons-based SUVR quantification across cognitively normal and neurodegenerative patients.

Source article: Brain fluorodeoxyglucose PET anatomical segmentation via AI: extensive validation in the neurodegenerative spectrum

Brain
Reader signal

How should this claim be treated?

45
Evidence-backed gainPeer-reviewedScience

Life-science physics students engaged in prompting, refining, and validating generative AI-constructed simulations as a new method for virtual learning about physical phenomena.

Source article: Leveraging generative artificial intelligence for simulation-based physics experiments: A new approach to virtual learning about the real world

Semantic Scholar–indexed venue
Reader signal

How should this claim be treated?

46
Evidence-backed gainPeer-reviewedClimate

Machine learning-based risk zoning using integrated biological and abiotic indexes improved fit and classification, with CFBRI raising R2 to 0.62 and Random Forest models reaching cross-validation accuracies of 0.888 for agricultural land and 0.905 for construction land.

Source article: Ecological risk zoning of soil heavy metal pollution using biomarker-enhanced indexing and machine learning

Environmental Geochemistry and Health
Reader signal

How should this claim be treated?

47
Evidence-backed gainPeer-reviewedClimate

Physics-informed categorical chain differential model improved coupled prediction of four ash fusion temperatures, reducing deformation temperature error and eliminating physically impossible temperature inversions to support boiler safety and slagging-risk management.

Source article: Physics-informed machine learning for universal ash fusion temperatures prediction: A novel categorical chain differential framework

Bioresource Technology
Reader signal

How should this claim be treated?

49
Evidence-backed gainPeer-reviewedScience

Ensemble tree-based machine learning models predicted band gap energies across multiple halide perovskite families with strong accuracy and identified B-site and X-site properties as key descriptors to guide design of new low-toxicity materials for optoelectronics.

Source article: Machine learning insights into band gap properties in halide-based perovskites

Physical Chemistry Chemical Physics
Reader signal

How should this claim be treated?

50
Evidence-backed gainPeer-reviewedClimate

Explainable AI-Supported Cyber-Physical Collaboration for Sustainable Manufacturing in Industry 5.0: Experimental findings show that the proposed CNN-Transformer architecture achieves 98.5% accuracy and outperforms existing CPHS systems while maintaining low latency.

Source article: Explainable AI-Supported Cyber-Physical Collaboration for Sustainable Manufacturing in Industry 5.0

Journal of Visualized Experiments
Reader signal

How should this claim be treated?

51
Evidence-backed gainPeer-reviewedClimate

To enhance prediction accuracy across a broad concentration range, a machine learning model was introduced to develop a high-precision quantitative analysis method for Fe 3+ .

Source article: Machine learning-assisted nitrogen-doped carbon dots for Fe<sup>3+</sup> detection in aqueous environments

Analytical Methods
Reader signal

How should this claim be treated?

52
Evidence-backed gainPeer-reviewedScience

Data-efficient exploration of atomic clusters via uncertainty quantification on complex potential energy surfaces: Furthermore, active learning driven by MCD significantly reduces the computational overhead of first-principles calculations while maintaining high predictive accuracy.

Source article: Data-efficient exploration of atomic clusters via uncertainty quantification on complex potential energy surfaces

The Journal of Chemical Physics
Reader signal

How should this claim be treated?

53
Evidence-backed gainPeer-reviewedScience

Forest kernel balancing that builds kernels from random forest and BART leaf co-occurrence improves computational and statistical performance for balancing covariates in observational causal inference by prioritizing outcome-relevant nonlinearities and interactions.

Source article: Forest Kernel Balancing Weights: Outcome-Guided Features for Causal Inference

Statistics in Medicine
Reader signal

How should this claim be treated?

54
Evidence-backed gainPeer-reviewedClimate

An integrated source-pathway-receptor framework using XGBoost and SHAP improved differentiation of industrial versus agricultural PTE risks in farmland soils and supported targeted zoned control to avoid excessive remediation.

Source article: Quantitative decoupling of source-pathway-receptor driving mechanisms for integrated soil risk via interpretable machine learning

Environmental Pollution
Reader signal

How should this claim be treated?

55
Evidence-backed gainPeer-reviewedScience

Logistic regression screening model using routine indicators like total protein and hemoglobin achieved AUC 0.843 and provides an accessible tool for early identification of individuals at high risk of M-protein in resource-limited primary care.

Source article: Development and validation of a generalizable M-protein screening model using routine laboratory indicators: a multicenter retrospective study

Clinica Chimica Acta
Reader signal

How should this claim be treated?

56
Evidence-backed gainPeer-reviewedScience

Adaptive boosting support vector regression trained on GC parameters predicts retention times with high accuracy, enabling optimization of capillary gas chromatography to separate coeluted C1-C12 hydrocarbon isomers.

Source article: Separation of coeluted compounds in capillary gas chromatography: customization via machine learning retention time prediction

Analytical and Bioanalytical Chemistry
Reader signal

How should this claim be treated?

57
Evidence-backed gainPeer-reviewedClimate

Domain-specific EfficientNet-B0 classifier and Claude Vision API with human-in-the-loop guidance automated morphological characterization of microplastics from optical microscope images, achieving high F1-scores for shape/type, color and texture.

Source article: Developing and evaluating automated deep learning and human-in-the-loop vision-language systems for microplastic characterization

Scientific Reports
Reader signal

How should this claim be treated?

58
Evidence-backed gainPeer-reviewedScience

Optimal Initialization Scale for Neural Networks With Locally Quadratic Loss Landscapes: An SGD Dynamics Perspective: The results show that, for the simple network model, if the variance of the initialization distribution satisfies our theoretical optimal condition, then the corresponding network achieves lower final training loss and higher test accuracy than the conventional He-normal initialization.

Source article: Optimal Initialization Scale for Neural Networks With Locally Quadratic Loss Landscapes: An SGD Dynamics Perspective

IEEE Transactions on Neural Networks and Learning Systems
Reader signal

How should this claim be treated?

59
Evidence-backed gainPeer-reviewedClimate

Integrating an NDVI-based Productivity Index with regional prediction uncertainty to select four field samples reduced mean field-scale SOC prediction error from 0.24% to 0.18% RMSE, achieving accuracy comparable to using all available field observations.

Source article: Productivity-based sampling for field-scale soil organic carbon mapping: from regional models to carbon farming needs

Environmental Monitoring and Assessment
Reader signal

How should this claim be treated?

60
Evidence-backed gainPeer-reviewedScience

A hybrid quantum-classical model combining variational quantum circuits with ESM2 achieved statistically significant improvements over classical baselines for multi-class IDR binding partner prediction.

Source article: Quantum-Enhanced Transfer Learning for IDR Binding Partner Prediction

IEEE Transactions on Computational Biology and Bioinformatics
Reader signal

How should this claim be treated?