Cautions · 01Evaluation and Overfitting Risks
119 papers23.4% of this theme group
Works caution against biased benchmarks, over‑complex models, and unreliable statistical assessments.
Pattern recognition (7) · Benchmark (8) · Random forest (6) · Statistical analysis (5)
Double/debiased machine learning for treatment and structural parameters
Making better M <scp>axent</scp> models of species distributions: complexity, overfitting and evaluation
View research history
Cautions · 02Cloud Vulnerability Exposure
107 papers21.0% of this theme group
Authors warn of security threats, virtualization flaws, and scalability‑related risks in cloud platforms.
Cloud computing (98) · Operating system (14) · Virtualization (11) · Scalability (14)
On the Use of Cloud Computing for Scientific Workflows
A survey of risks, threats and vulnerabilities in cloud computing
View research history
Cautions · 03Atmospheric Model Biases
83 papers16.3% of this theme group
Studies caution about systematic errors in aerosol and cloud simulations affecting climate predictions.
Atmospheric sciences (29) · Aerosol (24) · Meteorology (23) · Climatology (15)
Intercomparison of model simulations of mixed‐phase clouds observed during the ARM Mixed‐Phase Arctic Cloud Experiment. I: single‐layer clou
Exposing Global Cloud Biases in the Community Atmosphere Model (CAM) Using Satellite Observations and Their Corresponding Instrument Simulat
View research history
Cautions · 043‑D IC Design and Yield
59 papers11.6% of this theme group
Papers highlight manufacturing, interconnect, and testing challenges that threaten high‑yield 3D chips.
Three-dimensional integrated circuit (12) · Integrated circuit (12) · Interconnection (11) · Electronic circuit (11)
Design automation and testing of monolithic 3D ICs: Opportunities, challenges, and solutions: (Invited paper)
Assembly challenges in developing 3D IC package with ultra high yield and high reliability
View research history
Cautions · 05Generative AI Transparency
58 papers11.4% of this theme group
Researchers stress ethical and interpretability concerns surrounding AI‑generated text and educational use.
MEDLINE (6) · Generative grammar (6) · Transparency (4) · Applications of artificial intelligence (4)
Opinion Paper: “So what if ChatGPT wrote it?” Multidisciplinary perspectives on opportunities, challenges and implications of generative con
Students’ perceptions and responsible adoption of artificial intelligence in education: Ethical considerations, impacts, and academic perfor
View research history
Cautions · 06Radiative Transfer Modeling Limits
53 papers10.4% of this theme group
Works point to uncertainties in cloud radiative forcing and satellite‑based measurement accuracy.
Radiative transfer (13) · Cloud computing (32) · Satellite (11) · Radiative forcing (5)
Effects of cloud horizontal inhomogeneity and drizzle on remote sensing of cloud droplet effective radius: Case studies based on large‐eddy
Experimental Investigation of Convective and Radiative Heat Transfer to a Dilute Particle Cloud in a Subatmospheric Environment
View research history
Cautions · 07Adversarial and Covert Threats
30 papers5.9% of this theme group
Papers warn of adversarial attacks and hidden channels that can compromise AI training pipelines.
Adversarial system (5) · Exploit (4) · Training set (3) · Covert (3)
Ensemble Adversarial Training: Attacks and Defenses
Invisible Leaks: Covert Channel Exploitation in In-Sensor Computing System
View research history
