Engineering Sciences

140 items

arXiv Sep 29, 2026
The work proposes Recursive Harness Distillation (RHD): a strong agent (GPT-6 Astra) first interacts with a frozen vision-language-action model and distills effective intervention strategies into a reusable playbook, then recursively revises that playbook using the light agent's (GPT-5.6 Luna) execution feedback, improving manipulation without updating any model parameters—real-world success rises from 37.3% to 64.0%, the light agent with the playbook reaches 66.7% on SimplerEnv Bridge versus 41.7% for GR00T alone, and the same playbook also lifts the strong agent to 79.2%.
The work proposes Recursive Harness Distillation (RHD): a strong agent (GPT-6 Astra) first interacts with a frozen vision-language-action model and distills effective intervention strategies into a reusable playbook, then recursively revises that playbook using the light agent's (GPT-5.6 Luna) execution feedback, improving manipulation without updating any model parameters—real-world success rises from 37.3% to 64.0%, the light agent with the playbook reaches 66.7% on SimplerEnv Bridge versus 41.7% for GR00T alone, and the same playbook also lifts the strong agent to 79.2%.
A strong agent writes its intervention experience into a playbook for a light agent, lifting real-world manipulation success from 37.3% to 64.0%
The work proposes Recursive Harness Distillation (RHD): a strong agent (GPT-6 Astra) first interacts with a frozen vision-language-action model and distills effective intervention strategies into a reusable playbook, then recursively revises that playbook using the light agent's (GPT-5.6 Luna) execution feedback, improving manipulation without updating any model parameters—real-world success rises from 37.3% to 64.0%, the light agent with the playbook reaches 66.7% on SimplerEnv Bridge versus 41.7% for GR00T alone, and the same playbook also lifts the strong agent to 79.2%.
The work proposes Recursive Harness Distillation (RHD): a strong agent (GPT-6 Astra) first interacts with a frozen vision-language-action model and distills effective intervention strategies into a reusable playbook, then recursively revises that playbook using the light agent's (GPT-5.6 Luna) execution feedback, improving manipulation without updating any model parameters—real-world success rises from 37.3% to 64.0%, the light agent with the playbook reaches 66.7% on SimplerEnv Bridge versus 41.7% for GR00T alone, and the same playbook also lifts the strong agent to 79.2%.
arXiv Sep 29, 2026
REALM is a coarse-to-fine framework for audio-driven reactive listening that fuses listener motion history with speaker audio through a Reactive Gated Speaker–Listener Fusion module using a delay-centered attention prior and adaptive gating, then predicts a base motion trajectory with a coarse decoder and augments it with audio-conditioned stochastic residuals in the expression subspace, improving over the evaluated baselines on multiple motion-quality metrics on ViCo and L2L and deploying to an Ameca humanoid robot with a perceptual user study.
REALM is a coarse-to-fine framework for audio-driven reactive listening that fuses listener motion history with speaker audio through a Reactive Gated Speaker–Listener Fusion module using a delay-centered attention prior and adaptive gating, then predicts a base motion trajectory with a coarse decoder and augments it with audio-conditioned stochastic residuals in the expression subspace, improving over the evaluated baselines on multiple motion-quality metrics on ViCo and L2L and deploying to an Ameca humanoid robot with a perceptual user study.
REALM adds a delay prior and stochastic expression refinement to turn listening from a frozen stare into blinks and smiles on a robot
REALM is a coarse-to-fine framework for audio-driven reactive listening that fuses listener motion history with speaker audio through a Reactive Gated Speaker–Listener Fusion module using a delay-centered attention prior and adaptive gating, then predicts a base motion trajectory with a coarse decoder and augments it with audio-conditioned stochastic residuals in the expression subspace, improving over the evaluated baselines on multiple motion-quality metrics on ViCo and L2L and deploying to an Ameca humanoid robot with a perceptual user study.
REALM is a coarse-to-fine framework for audio-driven reactive listening that fuses listener motion history with speaker audio through a Reactive Gated Speaker–Listener Fusion module using a delay-centered attention prior and adaptive gating, then predicts a base motion trajectory with a coarse decoder and augments it with audio-conditioned stochastic residuals in the expression subspace, improving over the evaluated baselines on multiple motion-quality metrics on ViCo and L2L and deploying to an Ameca humanoid robot with a perceptual user study.
IEEE Spectrum Sep 28, 2026
Written by a Delhi power-engineering professor, this article traces how the city's distribution grid went from losses above 50% and a reliability index of about 70% in 2002 to 5-6% losses and a reliability index above 99.9% in 2026, and distills the path into a combination of technical upgrades, organizational and billing reform, enforcement, and community engagement.
Written by a Delhi power-engineering professor, this article traces how the city's distribution grid went from losses above 50% and a reliability index of about 70% in 2002 to 5-6% losses and a reliability index above 99.9% in 2026, and distills the path into a combination of technical upgrades, organizational and billing reform, enforcement, and community engagement.
Delhi's grid cut electricity losses from over 50% to 5-6% and lifted its reliability index from about 70% to above 99.9%
Written by a Delhi power-engineering professor, this article traces how the city's distribution grid went from losses above 50% and a reliability index of about 70% in 2002 to 5-6% losses and a reliability index above 99.9% in 2026, and distills the path into a combination of technical upgrades, organizational and billing reform, enforcement, and community engagement.
Written by a Delhi power-engineering professor, this article traces how the city's distribution grid went from losses above 50% and a reliability index of about 70% in 2002 to 5-6% losses and a reliability index above 99.9% in 2026, and distills the path into a combination of technical upgrades, organizational and billing reform, enforcement, and community engagement.
NVIDIA 开发者技术博客 Sep 28, 2026
NVIDIA and Nscale jointly evaluated DSX MaxLPS policy-governed dynamic power allocation with Kimi K2.5 (FP4) inference workloads on NVIDIA GB300 NVL72 systems at Nscale's data center at the Verne campus in Keflavík, Iceland: under the same 264.4 kW approved power budget, managed GPUs rose from 140 to 192 (+37.1%), aggregate throughput rose from 1,084,503 to 1,618,443 tokens/s (+49.2%), throughput per provisioned watt rose from 4.10 to 6.12 tokens/s/W (+49.2%), median and P75 latency stayed within 5% of baseline, while P99 time to first token increased 17% from the 15.7-second baseline.
NVIDIA and Nscale jointly evaluated DSX MaxLPS policy-governed dynamic power allocation with Kimi K2.5 (FP4) inference workloads on NVIDIA GB300 NVL72 systems at Nscale's data center at the Verne campus in Keflavík, Iceland: under the same 264.4 kW approved power budget, managed GPUs rose from 140 to 192 (+37.1%), aggregate throughput rose from 1,084,503 to 1,618,443 tokens/s (+49.2%), throughput per provisioned watt rose from 4.10 to 6.12 tokens/s/W (+49.2%), median and P75 latency stayed within 5% of baseline, while P99 time to first token increased 17% from the 15.7-second baseline.
NVIDIA and Nscale test DSX MaxLPS in Iceland: GPUs rise from 140 to 192 and aggregate throughput gains 49.2% under the same 264.4 kW budget
NVIDIA and Nscale jointly evaluated DSX MaxLPS policy-governed dynamic power allocation with Kimi K2.5 (FP4) inference workloads on NVIDIA GB300 NVL72 systems at Nscale's data center at the Verne campus in Keflavík, Iceland: under the same 264.4 kW approved power budget, managed GPUs rose from 140 to 192 (+37.1%), aggregate throughput rose from 1,084,503 to 1,618,443 tokens/s (+49.2%), throughput per provisioned watt rose from 4.10 to 6.12 tokens/s/W (+49.2%), median and P75 latency stayed within 5% of baseline, while P99 time to first token increased 17% from the 15.7-second baseline.
NVIDIA and Nscale jointly evaluated DSX MaxLPS policy-governed dynamic power allocation with Kimi K2.5 (FP4) inference workloads on NVIDIA GB300 NVL72 systems at Nscale's data center at the Verne campus in Keflavík, Iceland: under the same 264.4 kW approved power budget, managed GPUs rose from 140 to 192 (+37.1%), aggregate throughput rose from 1,084,503 to 1,618,443 tokens/s (+49.2%), throughput per provisioned watt rose from 4.10 to 6.12 tokens/s/W (+49.2%), median and P75 latency stayed within 5% of baseline, while P99 time to first token increased 17% from the 15.7-second baseline.
Frontiers in Pharmacology Sep 28, 2026
This narrative review traces TDDS evaluation from pre-1975 methodological fragmentation through Franz diffusion cell and IVPT standardization to RHE, full-thickness skin models, ex vivo human skin and iPSC-derived skin organoids (SkOs), reporting that SkOs self-organize stratified epidermis, dermal-like structures, hair follicles and sebaceous glands and that after roughly 4-5 months in culture their transcriptome resembles second-trimester human fetal skin, but that the authors' search identified no clear study using Lee-type hiPSC-derived SkOs as standardized barrier models in Franz diffusion cells or conventional IVPT with systematic measurement of cumulative permeation, steady-state flux, permeability coefficient or skin retention, concluding that SkOs are a frontier candidate rather t
This narrative review traces TDDS evaluation from pre-1975 methodological fragmentation through Franz diffusion cell and IVPT standardization to RHE, full-thickness skin models, ex vivo human skin and iPSC-derived skin organoids (SkOs), reporting that SkOs self-organize stratified epidermis, dermal-like structures, hair follicles and sebaceous glands and that after roughly 4-5 months in culture their transcriptome resembles second-trimester human fetal skin, but that the authors' search identified no clear study using Lee-type hiPSC-derived SkOs as standardized barrier models in Franz diffusion cells or conventional IVPT with systematic measurement of cumulative permeation, steady-state flux, permeability coefficient or skin retention, concluding that SkOs are a frontier candidate rather t
Review reports that iPSC-derived skin organoids self-assemble hair follicles and sebaceous glands, yet no study has used them for standardized Franz diffusion cell or IVPT permeation parameters
This narrative review traces TDDS evaluation from pre-1975 methodological fragmentation through Franz diffusion cell and IVPT standardization to RHE, full-thickness skin models, ex vivo human skin and iPSC-derived skin organoids (SkOs), reporting that SkOs self-organize stratified epidermis, dermal-like structures, hair follicles and sebaceous glands and that after roughly 4-5 months in culture their transcriptome resembles second-trimester human fetal skin, but that the authors' search identified no clear study using Lee-type hiPSC-derived SkOs as standardized barrier models in Franz diffusion cells or conventional IVPT with systematic measurement of cumulative permeation, steady-state flux, permeability coefficient or skin retention, concluding that SkOs are a frontier candidate rather t
This narrative review traces TDDS evaluation from pre-1975 methodological fragmentation through Franz diffusion cell and IVPT standardization to RHE, full-thickness skin models, ex vivo human skin and iPSC-derived skin organoids (SkOs), reporting that SkOs self-organize stratified epidermis, dermal-like structures, hair follicles and sebaceous glands and that after roughly 4-5 months in culture their transcriptome resembles second-trimester human fetal skin, but that the authors' search identified no clear study using Lee-type hiPSC-derived SkOs as standardized barrier models in Franz diffusion cells or conventional IVPT with systematic measurement of cumulative permeation, steady-state flux, permeability coefficient or skin retention, concluding that SkOs are a frontier candidate rather t
Journal of Strategic Innovation and Sustainability Sep 28, 2026
This study presents a smart gardening framework that integrates a mobile robot for image acquisition with a convolutional neural network (CNN) for flower recognition, where the robot captures garden images, the CNN classifies flower species and provides plant-specific information for future care decisions, achieving 92.0% training accuracy and 95.0% testing accuracy on 50 training images and 50 independent testing images, and providing a foundation for future irrigation, fertilization, and autonomous garden-management functions.
This study presents a smart gardening framework that integrates a mobile robot for image acquisition with a convolutional neural network (CNN) for flower recognition, where the robot captures garden images, the CNN classifies flower species and provides plant-specific information for future care decisions, achieving 92.0% training accuracy and 95.0% testing accuracy on 50 training images and 50 independent testing images, and providing a foundation for future irrigation, fertilization, and autonomous garden-management functions.
A mobile-robot-plus-CNN flower recognition framework reached 92.0% training and 95.0% testing accuracy on 50 training and 50 test images
This study presents a smart gardening framework that integrates a mobile robot for image acquisition with a convolutional neural network (CNN) for flower recognition, where the robot captures garden images, the CNN classifies flower species and provides plant-specific information for future care decisions, achieving 92.0% training accuracy and 95.0% testing accuracy on 50 training images and 50 independent testing images, and providing a foundation for future irrigation, fertilization, and autonomous garden-management functions.
This study presents a smart gardening framework that integrates a mobile robot for image acquisition with a convolutional neural network (CNN) for flower recognition, where the robot captures garden images, the CNN classifies flower species and provides plant-specific information for future care decisions, achieving 92.0% training accuracy and 95.0% testing accuracy on 50 training images and 50 independent testing images, and providing a foundation for future irrigation, fertilization, and autonomous garden-management functions.
Journal of Data Science and Intelligent Systems Sep 28, 2026
Using a Multilayer Perceptron Regressor (MLPR) as the base model and a Random Committee hybrid (RC-MLPR), the study predicted the particle Froude number (PFr) in auto-washout drainage systems from five heterogeneous datasets collected from existing literature covering a wide range of hydraulic and sediment conditions, evaluated the models with several performance measures including the agreement index, found that RC-MLPR outperformed other proposed ML models, state-of-the-art ML models, and existing empirical equations, and reported from sensitivity analysis that volumetric sediment concentration (Csed) is the most sensitive variable for PFr prediction by the hybrid RC-MLPR model.
Using a Multilayer Perceptron Regressor (MLPR) as the base model and a Random Committee hybrid (RC-MLPR), the study predicted the particle Froude number (PFr) in auto-washout drainage systems from five heterogeneous datasets collected from existing literature covering a wide range of hydraulic and sediment conditions, evaluated the models with several performance measures including the agreement index, found that RC-MLPR outperformed other proposed ML models, state-of-the-art ML models, and existing empirical equations, and reported from sensitivity analysis that volumetric sediment concentration (Csed) is the most sensitive variable for PFr prediction by the hybrid RC-MLPR model.
A hybrid Random Committee and Multilayer Perceptron Regressor model predicts particle Froude number in auto-washout drainage systems with sedimented beds and identifies volumetric sediment concentration as the most sensitive variable
Using a Multilayer Perceptron Regressor (MLPR) as the base model and a Random Committee hybrid (RC-MLPR), the study predicted the particle Froude number (PFr) in auto-washout drainage systems from five heterogeneous datasets collected from existing literature covering a wide range of hydraulic and sediment conditions, evaluated the models with several performance measures including the agreement index, found that RC-MLPR outperformed other proposed ML models, state-of-the-art ML models, and existing empirical equations, and reported from sensitivity analysis that volumetric sediment concentration (Csed) is the most sensitive variable for PFr prediction by the hybrid RC-MLPR model.
Using a Multilayer Perceptron Regressor (MLPR) as the base model and a Random Committee hybrid (RC-MLPR), the study predicted the particle Froude number (PFr) in auto-washout drainage systems from five heterogeneous datasets collected from existing literature covering a wide range of hydraulic and sediment conditions, evaluated the models with several performance measures including the agreement index, found that RC-MLPR outperformed other proposed ML models, state-of-the-art ML models, and existing empirical equations, and reported from sensitivity analysis that volumetric sediment concentration (Csed) is the most sensitive variable for PFr prediction by the hybrid RC-MLPR model.
Journal of Intelligent Media Computing Sep 28, 2026
This study used Mask R-CNN to segment goats from two-dimensional images and extract heart girth and body length, combined these with age and sex metadata, trained XGBoost, Random Forest and CatBoost and integrated them with weights of 0.5/0.3/0.2; on a validation set split by Animal ID (37 goats) XGBoost reached R²=0.8514 (RMSE=4.26 kg, MAE=3.47 kg, MAPE=16.27%), and on a completely unseen test set (37 goats) XGBoost performed best with R²=0.7955 (MSE=29.15 kg², RMSE=5.40 kg, MAE=4.01 kg, MAPE=16.38%) while the weighted ensemble recorded R²=0.7713 (MSE=32.60 kg², RMSE=5.71 kg, MAE=4.13 kg, MAPE=16.60%), indicating that combining visual and morphometric information under strict group-based evaluation provides reliable non-contact body weight estimation on unseen livestock.
This study used Mask R-CNN to segment goats from two-dimensional images and extract heart girth and body length, combined these with age and sex metadata, trained XGBoost, Random Forest and CatBoost and integrated them with weights of 0.5/0.3/0.2; on a validation set split by Animal ID (37 goats) XGBoost reached R²=0.8514 (RMSE=4.26 kg, MAE=3.47 kg, MAPE=16.27%), and on a completely unseen test set (37 goats) XGBoost performed best with R²=0.7955 (MSE=29.15 kg², RMSE=5.40 kg, MAE=4.01 kg, MAPE=16.38%) while the weighted ensemble recorded R²=0.7713 (MSE=32.60 kg², RMSE=5.71 kg, MAE=4.13 kg, MAPE=16.60%), indicating that combining visual and morphometric information under strict group-based evaluation provides reliable non-contact body weight estimation on unseen livestock.
Extracting heart girth and body length from 2D images with age and sex, XGBoost predicts body weight in unseen Surti goats at R²=0.7955
This study used Mask R-CNN to segment goats from two-dimensional images and extract heart girth and body length, combined these with age and sex metadata, trained XGBoost, Random Forest and CatBoost and integrated them with weights of 0.5/0.3/0.2; on a validation set split by Animal ID (37 goats) XGBoost reached R²=0.8514 (RMSE=4.26 kg, MAE=3.47 kg, MAPE=16.27%), and on a completely unseen test set (37 goats) XGBoost performed best with R²=0.7955 (MSE=29.15 kg², RMSE=5.40 kg, MAE=4.01 kg, MAPE=16.38%) while the weighted ensemble recorded R²=0.7713 (MSE=32.60 kg², RMSE=5.71 kg, MAE=4.13 kg, MAPE=16.60%), indicating that combining visual and morphometric information under strict group-based evaluation provides reliable non-contact body weight estimation on unseen livestock.
This study used Mask R-CNN to segment goats from two-dimensional images and extract heart girth and body length, combined these with age and sex metadata, trained XGBoost, Random Forest and CatBoost and integrated them with weights of 0.5/0.3/0.2; on a validation set split by Animal ID (37 goats) XGBoost reached R²=0.8514 (RMSE=4.26 kg, MAE=3.47 kg, MAPE=16.27%), and on a completely unseen test set (37 goats) XGBoost performed best with R²=0.7955 (MSE=29.15 kg², RMSE=5.40 kg, MAE=4.01 kg, MAPE=16.38%) while the weighted ensemble recorded R²=0.7713 (MSE=32.60 kg², RMSE=5.71 kg, MAE=4.13 kg, MAPE=16.60%), indicating that combining visual and morphometric information under strict group-based evaluation provides reliable non-contact body weight estimation on unseen livestock.
Jurnal Riset Teknik Komputer Sep 28, 2026
The article proposes a computer-vision pipeline that combines color-based segmentation with a convolutional neural network to detect healthy, blue-stain, and fungal regions on Sengon wood surfaces: RGB is transformed into HSV and YCrCb to exploit hue, saturation, and chrominance differences, thresholding and morphological operations yield candidate regions, these are cropped into fixed-size patches and classified by a lightweight CNN, and precision, recall, F1-score, accuracy, IoU, and Dice coefficient form the evaluation design; the article states it is a research design, so empirical result values are not yet presented and will be reported after the target dataset is tested.
The article proposes a computer-vision pipeline that combines color-based segmentation with a convolutional neural network to detect healthy, blue-stain, and fungal regions on Sengon wood surfaces: RGB is transformed into HSV and YCrCb to exploit hue, saturation, and chrominance differences, thresholding and morphological operations yield candidate regions, these are cropped into fixed-size patches and classified by a lightweight CNN, and precision, recall, F1-score, accuracy, IoU, and Dice coefficient form the evaluation design; the article states it is a research design, so empirical result values are not yet presented and will be reported after the target dataset is tested.
For Sengon wood in Jepara, researchers propose a color-segmentation plus lightweight-CNN pipeline for blue stain and fungal detection, with no empirical results yet
The article proposes a computer-vision pipeline that combines color-based segmentation with a convolutional neural network to detect healthy, blue-stain, and fungal regions on Sengon wood surfaces: RGB is transformed into HSV and YCrCb to exploit hue, saturation, and chrominance differences, thresholding and morphological operations yield candidate regions, these are cropped into fixed-size patches and classified by a lightweight CNN, and precision, recall, F1-score, accuracy, IoU, and Dice coefficient form the evaluation design; the article states it is a research design, so empirical result values are not yet presented and will be reported after the target dataset is tested.
The article proposes a computer-vision pipeline that combines color-based segmentation with a convolutional neural network to detect healthy, blue-stain, and fungal regions on Sengon wood surfaces: RGB is transformed into HSV and YCrCb to exploit hue, saturation, and chrominance differences, thresholding and morphological operations yield candidate regions, these are cropped into fixed-size patches and classified by a lightweight CNN, and precision, recall, F1-score, accuracy, IoU, and Dice coefficient form the evaluation design; the article states it is a research design, so empirical result values are not yet presented and will be reported after the target dataset is tested.
Research Square Sep 28, 2026
The work proposes a UAV control scheme that combines a Recursive Gaussian Process (RGP) with a feedback linearization controller, using a purpose-designed Kalman-filter-based tuning algorithm to compensate online for the nonlinear dynamics left uncanceled by feedback linearization, implements the entire RGP pipeline directly onboard a real quadrotor (the ANT-X Lab drone), and validates it for single-axis horizontal position control in indoor experiments: across repetitive and non-repetitive sinusoidal references, the geometric mean root mean square error improves by 38% to 75% relative to the same controller without RGP, with geometric mean tracking errors between 4.5 cm (constant-amplitude sinusoid at omega=1 rad/s) and 9.85 cm (sweep sinusoid).
The work proposes a UAV control scheme that combines a Recursive Gaussian Process (RGP) with a feedback linearization controller, using a purpose-designed Kalman-filter-based tuning algorithm to compensate online for the nonlinear dynamics left uncanceled by feedback linearization, implements the entire RGP pipeline directly onboard a real quadrotor (the ANT-X Lab drone), and validates it for single-axis horizontal position control in indoor experiments: across repetitive and non-repetitive sinusoidal references, the geometric mean root mean square error improves by 38% to 75% relative to the same controller without RGP, with geometric mean tracking errors between 4.5 cm (constant-amplitude sinusoid at omega=1 rad/s) and 9.85 cm (sweep sinusoid).
Recursive Gaussian Process compensation cuts quadrotor single-axis horizontal tracking error by 38% to 75%
The work proposes a UAV control scheme that combines a Recursive Gaussian Process (RGP) with a feedback linearization controller, using a purpose-designed Kalman-filter-based tuning algorithm to compensate online for the nonlinear dynamics left uncanceled by feedback linearization, implements the entire RGP pipeline directly onboard a real quadrotor (the ANT-X Lab drone), and validates it for single-axis horizontal position control in indoor experiments: across repetitive and non-repetitive sinusoidal references, the geometric mean root mean square error improves by 38% to 75% relative to the same controller without RGP, with geometric mean tracking errors between 4.5 cm (constant-amplitude sinusoid at omega=1 rad/s) and 9.85 cm (sweep sinusoid).
The work proposes a UAV control scheme that combines a Recursive Gaussian Process (RGP) with a feedback linearization controller, using a purpose-designed Kalman-filter-based tuning algorithm to compensate online for the nonlinear dynamics left uncanceled by feedback linearization, implements the entire RGP pipeline directly onboard a real quadrotor (the ANT-X Lab drone), and validates it for single-axis horizontal position control in indoor experiments: across repetitive and non-repetitive sinusoidal references, the geometric mean root mean square error improves by 38% to 75% relative to the same controller without RGP, with geometric mean tracking errors between 4.5 cm (constant-amplitude sinusoid at omega=1 rad/s) and 9.85 cm (sweep sinusoid).

Page 5 · showing 10