Category
591 papers
Experimental results indicate that such complex designs exceed the capabilities of existing approaches, whereas the proposed LLM-based hardware development framework is the only one capable of producing functional end-to-end design.
This study examined the growth of Robusta coffee seedlings treated with ecoenzymes derived from fruit peel and addressed the limitations of red-yellow podzolic soil, which is deficient in macronutrients and organic matter. The objective of this study was to determine whether the application of different concentrations of ecoenzymes could enhance the growth of Robusta coffee seedlings. This study employed a non-factorial completely randomized design (CRD) consisting of 5 treatments with 4 replicates. The treatments were as follows: K0 = application of 0 mL/L ecoenzyme concentration (Control), K1 = 25 mL/L ecoenzyme concentration, K2 = 50 mL/L ecoenzyme concentration, K3 = 75 mL/L ecoenzyme concentration, and K4 = 100 mL/L ecoenzyme concentration. The research data were analyzed using Analysis of Variance (ANOVA) and Duncan’s Multiple Range Test (DMRT) at the 5% significance level using the DSAASTAT software. The results showed that the application of different ecoenzyme concentrations had a significant effect on improving the growth of Robusta coffee in terms of seedling height, stem diameter, leaf length, leaf width, root length, and dry weight. The optimal ecoenzyme treatment dose was K2 (a concentration of 50 mL/L), which increased seedling height by 6.91 cm, stem diameter by 1.88 mm, leaf length by 7.62 cm, leaf width by 3.35 cm, root length by 9.17 cm, and dry weight by 0.48 g.
A mascot is a visual identity element that plays an important role in shaping the image and appeal of an event. Porsimaptar (Sports and Arts Week for Students, Pupils, and Cadets) is a prestigious annual competition organized by the Semarang Police Academy (Akpol). The official Porsimaptar logo features the silhouette of Gajah Mada, a national hero who is a symbol of unity and fighting spirit, but until now has not been realized in the form of a representative three-dimensional (3D) mascot. This design aims to create a 3D mascot with the character of Gajah Mada that elevates the visual identity of the Porsimaptar logo while presenting a fun and friendly impression to all participants. The method used is Design Thinking with five stages: Empathize, Define, Ideate, Prototype, and Test. The design process was carried out using Blender software as the main medium for 3D modeling and rendering. This 3D mascot is expected to be a consistent and adaptive visual asset for various Porsimaptar promotional media.
The results highlight that gamification features can have a significant impact if introduced in a classical testing environment, possibly increasing effectiveness, efficiency and perceived engagement of the testers.
Reach audiences
Advertise in front of researchers, engineers, and readers.
Public v5.0.0 artifact for the T1 truth-blind lifecycle-conditioned metamorphic-testing study. Contents frozen T1 protocol and scenario definitions source code and native Tier-A OpenSees provider 24 constructed cases and 72 scientific process records detector manifests, post-freeze evaluation tables, information-flow canaries, and QA reports full SHA-256 manifest and public sanitization report Verified results 72/72 scientific summaries matched the frozen reference 27/27 detected cases localized at Top-1 45/45 non-detected cases correctly returned no localization 36/36 held-out records matched 10/10 information-flow canaries passed 0 scientific retries Scope This is an implementation-verification and evidence archive, not a claim of full engineering qualification. The public mirror normalizes only local absolute path prefixes; controlled scientific values and outcomes are unchanged. Zenodo DOI: https://doi.org/10.5281/zenodo.22234369 Licensing: code/scripts BSD-3-Clause; data/docs CC BY 4.0; linked OpenSees materials retain bundled upstream terms.
Edge AI systems are emerging from the convergence of IoT, edge computing, AI, agentic AI, and embodied, physical generative edge AI delivering adaptive, autonomous behaviour under physical, cyber, and operational constraints while remaining trustworthy. This article frames edge AI as a complex system-of-systems in which hardware, software, models, and data continuously co-evolve across heterogeneous “multi-X” environments: multiple systems, modalities, and agents distributed from the edge to the cloud. The article argues that as edge AI technologies are maturing, there is a need for a standardised, application-agnostic reference architecture to provide a shared lexicon and taxonomy, reduce integration errors, and expose opportunities for reusable assets and productive interoperability and standardisation. The paper grounds this need in systems engineering and introduces a quad-optimisation paradigm for balancing competing objectives during design and operation. The article presents a design framework and a multi-dimensional architecture organised into three complementary views: quality properties for trustworthiness and dependability, a layered technology stack within each tier, and a processing continuum that partitions intelligence across edge-to-cloud tiers. Finally, the article discusses value creation, interoperability, and how a 2 common baseline supports the development of complex edge AI systems-of-systems and their verification, validation, testing and benchmarking.
This note collects the constructions, proofs, and prior-art positioning behind the exact-rational certificates in the Podium library (podium.verify): barrier certificates for abort safety, Karush-Kuhn-Tucker certificates for the online convex solves, control-Lyapunov certificates, sum-of-squares certificates, and an optimality-gap bracket for nonconvex quadratically constrained quadratic programs. None of the underlying certificate mathematics is new; the note documents the exact instantiation, its verification, and its relation to the exact-SOS/moment, rational-SDP-recovery, and QCQP-duality literature, so that a math-inclined reader can audit the certificate layer directly. In detail, guidance and control software relies on numerical optimization and floating-point synthesis whose executions are difficult to verify directly. We study the problem of verifying solver outputs: given a certificate produced by an untrusted floating-point solver, we re-verify it in exact rational arithmetic, so that acceptance constitutes a proof that carries no rounding error. We apply this to four properties: abort safety, via barrier certificates; optimality of the online convex solves, via Karush-Kuhn-Tucker conditions for quadratic and second-order cone programs; closed-loop stability, via Lyapunov ellipsoids from a discrete-time linear-quadratic regulator; and polynomial set containment, via sum-of-squares certificates. For nonconvex quadratically constrained quadratic programs (QCQPs), we specialize the exact rational recovery of a semidefinite certificate (Peyrl-Parrilo; Kaltofen et al.) to the S-procedure dual, yielding a tolerance-free, machine-checkable optimality-gap bracket: a rational lower bound from the S-procedure dual and a rational upper bound from a feasible point, coinciding exactly at a global optimum. The note's original analysis characterizes how the certified lower bound degrades under rational rounding of the dual multiplier: the value loss is second-order in the rounding denominator in the nonsingular case and first-order in the trust-region hard case, with soundness retained in both, and it vanishes when the optimal multiplier is rational. For several constraints the S-procedure need not be tight, and the bracket then certifies a duality gap with both endpoints exact. Each result is accompanied by a verifying test in the reference implementation.
A hands-on, test-first tutorial for building real software with AI while keeping human control of design, scope, and evidence. Worked examples use Aider, Python, and pytest.
SUPER DNS ONE is a fully differentiable, three‑dimensional finite‑volume solver for the compressible Navier–Stokes equations. It is designed for high‑fidelity civilian research: Medical flows – cardiovascular haemodynamics, respiratory aerosol transport, micro‑fluidic drug delivery. Atmospheric and environmental physics – turbulent boundary layers, pollutant dispersion, cloud microphysics. Civil aviation – aerodynamic analysis, noise reduction, wake turbulence. Hypersonic civilian transport – real‑gas effects, shock capturing, high‑speed boundary‑layer transition. And More. This software is not intended, tested, or authorised for military applications, weapons development, or any form of armed conflict.
Density-functional-theory dataset for Fe substitution in BaTiO3 and the competition between its tetragonal and hexagonal polymorphs, together with SrTiO3 and CaTiO3 control systems. The archive holds 700 structures. For every calculation it gives the unrelaxed input geometry and the converged relaxed geometry, together with the total energies and the configurational degeneracies needed to reproduce the reported averages, and the VASP input templates. 486 Fe-substituted BaTiO3 configurations, tetragonal and hexagonal, with and without one oxygen vacancy, at 1.85, 3.70 and 5.56 at.% Fe (all symmetry-inequivalent arrangements, each with its degeneracy) 95 Fe-substituted SrTiO3 and CaTiO3 controls 7 undoped host primitive cells 72 repeats at Ueff = 2 and 6 eV 4 configurations with the FeO6 shell held at a uniform bond length (Jahn-Teller reference) 8 electronic-structure runs: band unfolding, high-resolution density of states, frozen-ion SCF and the Fe-free reference cell 27 supercell-shape and relaxation-scheme tests 1 symmetry-disabled test The first three blocks are the 588 converged production relaxations the study is built on. Computational details are given in the paper, arXiv:2603.22571. The VASP input files used are included under inputs/. Pseudopotentials. VASP POTCAR files are licensed by the VASP Software GmbH and are not redistributed here. inputs/POTCAR_provenance.txt names the four PAW datasets (PAW_PBE Ba_sv 06Sep2000, Fe 06Sep2000, Ti_pv 07Sep2000, O_s 07Sep2000) and the species orderings that define the concatenation order for each structure. Start from README.md; manifest.csv indexes every deposited structure. Funding. Deutsche Forschungsgemeinschaft, Project-ID 463184206, SFB 1548. Computing time on the Lichtenberg high-performance computer at NHR4CES, TU Darmstadt (Project ID p0026451).
Replication package for "LLM-Based Exploratory Testing Charter Generation: An Ablation Study of Prompt Structure" (Arbaz Surti; under review at Automated Software Engineering, Springer). This version corresponds to the round-1 major revision and supersedes v1. Contents: the evaluation dataset (25 requirement specifications); all 375 generated charters across the three prompt conditions (ETCG / Intermediate / Baseline); rubric scores; scripts/analysis.py, the single source of truth that regenerates every table, figure and reported value in the manuscript plus a traceability index; cross-model judge and multi-model generation data; the human inter-rater inputs (both raters' scoring keys and the scoring sheet); and the pre-repair scoring snapshot used by the Threats-to-Validity provenance analysis. Data under CC BY 4.0; code under the MIT licence (see LICENSE-CODE).
ABSTRACT: Version 9 changes the epistemic purpose of this research record. Rather than extending or defending the previously proposed explanatory framework, it reframes the archived observations as a falsification challenge. Speculative mechanisms proposed in earlier versions - including Orch-OR 2.0, Hybrid Thermal Entanglement, and relationalentanglement interpretations - are explicitly treated here as hypotheses to be tested against conventional software, platform, statistical, cognitive, and physiological explanations. Earlier versions are intentionally preserved as part of the chronological research record so that the development of the observations, inferences, errors, corrections, and hypotheses remains auditable. Version 9 does not retract the underlying archived observations merely because their proposed mechanism may be wrong. Its purpose is to separate the evidential record from the explanatory model and to invite independent researchers and platform engineers to reproduce the observations, identify ordinary mechanisms where sufficient, and isolate any residual that survives appropriate controls.
In his latest book, Professor Eugene Fitzgerald examines the forces that turn breakthroughs into value — and why innovation resists simple formulas.
MIT students designed, built, and tested a jet engine with AI copilots, assessing AI’s usefulness in developing high-performance aerospace systems.
Researchers developed an auditing technique to test generative AI models for malicious capabilities, without prompting them for illegal outputs.
A USAF cadet and a Lincoln Laboratory researcher found AI chatbots can help nontechnical service members produce viable software applications for their unique problems.
Our latest voice model has improved precision and lower latency to make voice interactions more fluid, natural and precise.
Education Innovation
