# Neue Publikationen in Ziel-Venues – 2026-10-02 Neue Werke: 8 · davon mit DE-Affiliation: 8 Venues: 60 · abgefragt ab publication_date 2026-09-02 · letzter Lauf: 2026-10-02T13:08:34 ## IEEE Transactions on Very Large Scale Integration (VLSI) Systems (TVLSI) — 1 neu · _core_ ### Guest Editorial Special Issue on the IEEE Computer Society Annual Symposium on VLSI—ISVLSI 2025 ⚑ DE - **DOI:** [10.1109/tvlsi.2026.3722384](https://doi.org/10.1109/tvlsi.2026.3722384) - **Metadaten:** Erschienen: 2026-09-28 · Vol. 34, Issue 10, S. 3075-3077 · OpenAlex seit 2026-09-29 - **DE-Institutionen:** Technische Universität Dresden; Brandenburg University of Technology Cottbus-Senftenberg - **Autoren:** - Γεώργιος Κεραμίδας — Aristotle University of Thessaloniki (GR) - Diana Goehringer — Technische Universität Dresden (DE) - Paraskevas Kitsos — University of Peloponnese (GR) - Nikolaos Voros — University of Peloponnese (GR) - Michael Huebner — Brandenburg University of Technology Cottbus-Senftenberg (DE) - **Keywords:** VLSI; integrated circuit design ## ACM Transactions on Embedded Computing Systems (TECS) — 1 neu · _arch_ ### Compatibility Ratio as a Guideline for Hardware/DNN Co-Design of Embedded Accelerators ⚑ DE - **DOI:** [10.1145/3841637](https://doi.org/10.1145/3841637) - **Metadaten:** Erschienen: 2026-09-03 · OpenAlex seit 2026-09-04 - **DE-Institutionen:** University of Lübeck - **Autoren:** - Lukas Groth — University of Lübeck (DE) - Andrija Nešković — University of Lübeck (DE) - Rainer Buchty — University of Lübeck (DE) - Mladen Bereković — University of Lübeck (DE) - Saleh Mulhem — University of Lübeck (DE) - **Topics:** Advanced Neural Network Applications; Parallel Computing and Optimization Techniques; Embedded Systems Design Techniques - **Keywords:** hardware-software co-design; deep neural network accelerators; workload mapping; systolic arrays; edge inference; design space exploration - **Abstract:** Increasingly large deep neural networks (DNNs) pose a significant challenge regarding required compute capabilities and energy consumption, especially at the edge. This challenge generally necessitates dedicated inference hardware accelerators. Hardware/software co-design can improve overall system performance by jointly optimizing the hardware micro-architecture and DNN workload mapping. It is however not easy to quantify how well such an accelerator generalizes to workloads not considered during hardware/software co-design. In response to this, we introduce the Compatibility Ratio (CR) as a simple guideline for evaluating performance trade-offs between optimal hardware micro-architecture configurations across different workloads. CR allows us to quantify the performance trade-offs of deploying a workload on an accelerator optimized for a different workload. We demonstrate CR through two case studies on a systolic array-based accelerator. First, we apply CR to explore the design space of the accelerator across 13 DNN workloads. In this case study, CR analysis showed that the choice of representative workload during co-design can implicitly increase the normalized area-latency cost of unconsidered workloads by more than 30% in the evaluated design space. Furthermore, we use CR to analyze how well our systolic array-based accelerator template can generalize beyond a single DNN workload to cover a family of DNN workloads. Our findings show that, for the considered accelerator, a DNN model-family optimized configuration might occupy an effective middle ground between highly targeted single- and general-purpose configurations. Second, we use CR as a guideline for a practical memory-retargeting decision in a specialized variant of our accelerator template. In this case study, CR quantifies whether a memory-retargeted accelerator derivative is justified under the selected memory-area and latency objective. For this two-configuration memory-retargeting case, analytical CR differs from implementation-level CR by 0.01, corresponding to one percentage point on the normalized CR scale. ## ACS Nano (ACSNANO) — 1 neu ### Selective Ultrafast Excitation and Dephasing of Gap Plasmons into Hot Carriers in Structural Color Generation ⚑ DE - **DOI:** [10.1021/acsnano.6c12999](https://doi.org/10.1021/acsnano.6c12999) - **Metadaten:** Erschienen: 2026-10-02 · OpenAlex seit 2026-10-02 - **DE-Institutionen:** Ludwig-Maximilians-Universität München - **Autoren:** - Manobina Karmakar — University of Central Florida (US); Ludwig-Maximilians-Universität München (DE) - Ayon Jyoti Karmakar — Indian Institute of Technology Kharagpur (IN) - Aritra Biswas — University of Central Florida (US) - Mahdi Soudi — University of Central Florida (US) - Tianyi Guo — University of Central Florida (US) - Pablo Cencillo‐Abad — University of Central Florida (US) - Prasanta Kumar Datta — Indian Institute of Technology Kharagpur (IN) - Debashis Chanda — University of Central Florida (US) - **Topics:** Plasmonic and Surface Plasmon Research; Metamaterials and Metasurfaces Applications; Gold and Silver Nanoparticles Synthesis and Applications - **Abstract:** Abstract Plasmon-induced hot-carriers generated through Landau damping underpin applications in plasmonic optoelectronics, energy conversion, and photocatalysis. However, it remains unclear whether these carriers retain the spatial localization of their parent plasmon modes or rapidly lose this spatial memory through ultrafast scattering and transport. Here, we experimentally probe the spatial and temporal dynamics of plasmon-induced hot-carriers in a strongly confined plasmonic structural color consisting of disordered aluminum nanoparticle ensembles on near-field cavities. Strong interparticle coupling produces hybridized and localized gap plasmon modes with intense nanoscale electromagnetic confinement, which we directly visualize using scattering-type scanning near-field optical microscopy. Ultrafast pump–probe measurements selectively excite a homogeneously broadened sub-ensemble of plasmonic modes, enabling indirect tracking of pump-generated, plasmon-induced hot-carrier dynamics. We observe that hot-carriers undergo only partial spatial delocalization and retain signatures of the initial plasmonic localization for hundreds of picoseconds. The perturbed sub-ensemble reaches maximum line widths of 0.54 ± 0.06 eV (0.36 ± 0.05 eV) under 630 nm (700 nm) excitation. On a phenomenological localization scale from 0 (localized) to 1 (fully delocalized), this corresponds to 0.35 ± 0.10 and 0.09 ± 0.08 for 630 and 700 nm excitation, respectively. The wavelength-dependent fraction shows that the degree of preserved localization is optically tunable, set by the carrier energy. This persistence of spatial memory, likely arising from disorder-induced scattering within the nanoparticle ensemble, reveals an unexpected spatial dimension of hot-carrier dynamics and highlights the role of plasmonic confinement and structural disorder in governing nanoscale energy flow, with implications for hot-carrier-based optoelectronics. ## Advanced Materials (ADVMAT) — 1 neu ### Ionic Cocrystal Programming of Spatial Ionic Organization in Carbon Nitrides Enables Highly Efficient H 2 O 2 Photosynthesis ⚑ DE - **DOI:** [10.1002/adma.75258](https://doi.org/10.1002/adma.75258) - **Metadaten:** Erschienen: 2026-10-01 · OpenAlex seit 2026-10-02 - **DE-Institutionen:** Max Planck Institute of Colloids and Interfaces; Center for Advanced Systems Understanding - **Autoren:** - Anna Lo Presti — Max Planck Institute of Colloids and Interfaces (DE) - Zahra Hajiahmadi — Center for Advanced Systems Understanding (DE) - Sam A. J. Hillman — Imperial College London (GB) - Mu Xiao — The University of Queensland (AU) - James Robert Durrant — University of Oxford (GB); Imperial College London (GB) - Thomas D. Kühne — Center for Advanced Systems Understanding (DE) - Christian Mark Pelicano — Max Planck Institute of Colloids and Interfaces (DE) - **Topics:** Advanced Photocatalysis Techniques; Advanced Battery Materials and Technologies; Electrocatalysts for Energy Conversion - **Abstract:** ABSTRACT Controlling the spatial organization of ions in polymeric photocatalysts remains a fundamental challenge for optimizing charge transport and catalytic selectivity in solar‐driven hydrogen peroxide (H 2 O 2 ) synthesis. In conventional poly(heptazine imide) (PHI), alkali cations are incorporated through molten‐salt‐mediated equilibration, producing statistically distributed ionic environments with little spatial correlation. Here, we demonstrate that organic‐inorganic ionic cocrystals (ICCs) serve as molecularly preorganized precursors that promote spatially correlated ionic environments during framework formation, enabling controlled incorporation of Li + into imide‐bridged N sites. This ICC‐programmed architecture decouples composition from photocatalytic function, demonstrating that photocatalytic activity is governed by the ion incorporation pathway and the resulting spatial ionic organization, rather than cation identity or content alone. The resulting ionic carbon nitride, CALiK, achieves an apparent quantum yield of 58% at 410 nm, the highest among the benchmarked state‐of‐the‐art photocatalysts, while requiring substantially lower inorganic salt loading than conventional ionothermal synthesis. Operando‐relevant spectroscopic and electrochemical analyses reveal directional charge migration driven by internal electrostatic polarization and enhanced accumulation of long‐lived photogenerated electrons. Representative density functional theory models suggest that spatially correlated ionic motifs and cyano‐derived electron‐accepting domains generate polarized charge landscapes that lower the energetic barrier for the two‐electron oxygen reduction pathway. ## Applied Physics Letters (APL) — 1 neu ### Laser-driven kA sub-100 ps current pulse generation and transport ⚑ DE - **DOI:** [10.1063/5.0341256](https://doi.org/10.1063/5.0341256) - **Metadaten:** Erschienen: 2026-09-28 · Vol. 129, Issue 13 · OpenAlex seit 2026-10-02 - **DE-Institutionen:** Helmholtz-Zentrum Dresden-Rossendorf; Technische Universität Dresden - **Autoren:** - Khalil Aliane — Centre National de la Recherche Scientifique (FR); Université de Bordeaux (FR); Commissariat à l'Énergie Atomique et aux Énergies Alternatives (FR); Centre Lasers Intenses et Applications (FR) - B. Etchessahar — CEA CESTA (FR) - A. Hirsch-Passicos — Helmholtz-Zentrum Dresden-Rossendorf (DE) - C. L. C. Lacoste — Centre National de la Recherche Scientifique (FR); Université de Bordeaux (FR); Commissariat à l'Énergie Atomique et aux Énergies Alternatives (FR); Institut National de la Recherche Scientifique (CA); Centre Lasers Intenses et Applications (FR); Hydro-Québec (CA) - J. Schilz — Helmholtz-Zentrum Dresden-Rossendorf (DE); Technische Universität Dresden (DE) - R. Bouillaud — Centre National de la Recherche Scientifique (FR); Université de Bordeaux (FR); Commissariat à l'Énergie Atomique et aux Énergies Alternatives (FR); Centre Lasers Intenses et Applications (FR) - L. Merzeau — Centre National de la Recherche Scientifique (FR); Université de Bordeaux (FR); Commissariat à l'Énergie Atomique et aux Énergies Alternatives (FR); Centre Lasers Intenses et Applications (FR) - F. Blais — Centre National de la Recherche Scientifique (FR); Université de Bordeaux (FR); Commissariat à l'Énergie Atomique et aux Énergies Alternatives (FR); Centre Lasers Intenses et Applications (FR) - V. Royer — CEA CESTA (FR) - Ulrich Schramm — Helmholtz-Zentrum Dresden-Rossendorf (DE); Technische Universität Dresden (DE) - Emmanuel d’Humières — Centre National de la Recherche Scientifique (FR); Université de Bordeaux (FR); Commissariat à l'Énergie Atomique et aux Énergies Alternatives (FR); Centre Lasers Intenses et Applications (FR) - Didier Raffestin — Centre National de la Recherche Scientifique (FR); Université de Bordeaux (FR); Commissariat à l'Énergie Atomique et aux Énergies Alternatives (FR); Centre Lasers Intenses et Applications (FR) - Florian Kroll — Helmholtz-Zentrum Dresden-Rossendorf (DE) - Josefine Metzkes-Ng — Helmholtz-Zentrum Dresden-Rossendorf (DE) - M. Bardon — Centre National de la Recherche Scientifique (FR); Université de Bordeaux (FR); Commissariat à l'Énergie Atomique et aux Énergies Alternatives (FR); Centre Lasers Intenses et Applications (FR) - **Topics:** Laser-Plasma Interactions and Diagnostics; Laser-Matter Interactions and Applications; Pulsed Power Technology Applications - **Abstract:** We report the generation and transport of laser-driven discharge currents in the sub-100 ps, kiloampere regime. A target and transport architecture was developed to mitigate dispersion, distortion, reflections, and the risk of electrical breakdown, enabling efficient extraction and guided propagation of the laser-induced discharge current. Experiments performed with a petawatt-class laser demonstrate measured peak currents up to 1.6 kA into 50 Ω (∼80 kV) with a measured pulse duration of 70 ps, limited by the 6 GHz diagnostic bandwidth. Bandwidth-corrected extrapolation suggests intrinsic pulses of a few tens of picoseconds with multi-kiloampere amplitudes, requiring confirmation by higher-bandwidth measurements. These results establish laser-driven targets as a viable source of ultrafast, high-amplitude electrical pulses and open new opportunities for applications in pulsed power and strong-field science. ## Journal of Applied Physics (JAP) — 1 neu ### Field-induced modulation of absorption in waveguides with InGaN quantum well active region ⚑ DE - **DOI:** [10.1063/5.0344838](https://doi.org/10.1063/5.0344838) - **Metadaten:** Erschienen: 2026-10-01 · Vol. 140, Issue 13 · OpenAlex seit 2026-10-02 - **DE-Institutionen:** Chemnitz University of Technology - **Autoren:** - Shemshat Kerimova — Chemnitz University of Technology (DE) - Ulrich Theodor Schwarz — Chemnitz University of Technology (DE) - Muhammed Aktaş — Institute of High Pressure Physics (PL) - P. Perlin — Institute of High Pressure Physics (PL) - Anna Kafar — Institute of High Pressure Physics (PL) - **Topics:** GaN-based semiconductor devices and materials; Advanced Fiber Laser Technologies; Plasmonic and Surface Plasmon Research - **Abstract:** Bias-dependent modal absorption and gain in InGaN/GaN multi-section Fabry–Pérot ridge laser diodes emitting between 415 and 470 nm are investigated. Optical gain spectra are extracted using the Hakki–Paoli method, allowing the modal absorption of the absorber section to be determined under forward and reverse bias conditions. The absorption decreases with increasing reverse bias and reaches a minimum near the flatband condition, where the applied field compensates the internal polarization field and reduces the quantum-confined Stark effect (QCSE). The extracted flatband voltage shifts toward larger reverse bias for longer emission wavelengths. In addition, the transparency-energy shift increases with emission wavelength, indicating stronger QCSE in higher-indium-content quantum wells. These results provide insight into field-induced modulation of optical transitions in InGaN quantum wells and differential absorption as a figure of merit for electro-absorption modulators. ## Optics Express (OE) — 1 neu ### Pulsed 978 nm swept-source laser for high-speed multiphoton microscopy ⚑ DE - **DOI:** [10.1364/oe.613074](https://doi.org/10.1364/oe.613074) - **Metadaten:** Erschienen: 2026-09-03 · Vol. 34, Issue 20, S. 38259 · OpenAlex seit 2026-09-04 - **DE-Institutionen:** University of Lübeck - **Autoren:** - Stefan Reinhards — keine Affiliation - Tonio Kutscher — University of Lübeck (DE) - Kimberley Goodwin — University of Lübeck (DE) - Sebastian Karpf — University of Lübeck (DE) - **Topics:** Advanced Fluorescence Microscopy Techniques; Digital Holography and Microscopy; Optical Coherence Tomography Applications - **Keywords:** swept-source laser; pulsed laser; multiphoton microscopy; high-speed imaging; femtosecond laser - **Abstract:** Spectro-temporal laser imaging by diffractive excitation (SLIDE) enables two-photon microscopy (TPM) with 2D imaging rates in the kHz range and is capable of performing real-time volumetric imaging at video rates. However, the application of SLIDE microscopy is constrained by its available excitation wavelength. To date, imaging has only been demonstrated with central wavelengths of 1064 nm and 775 nm. The underlying cause is the spectral limitation of appropriate fiber-based active gain media, such as ytterbium and erbium. These media play a crucial role in SLIDE in augmenting laser pulse power for two-photon excitation. In the case of commonly used green fluorescent dyes such as Alexa Fluor 488 or green fluorescent protein (GFP), it is necessary to amplify light with a center wavelength in the range of 976 nm. Ytterbium appears to be a preferred medium for this task, but its quasi-three-level lasing behavior at this wavelength poses significant challenges in its application. This study presents a high-power swept-source laser light source around 978 nm. The approach utilizes an all-fiber Fourier domain mode locking master oscillator power amplifier (FDML-MOPA) with a central wavelength of 978 nm and a sweep bandwidth of 4 nm. The light source employs a Mach-Zehnder electro-optical modulator (EOM) to generate 30 ps pulses at 82 MHz (512 pulses per sweep). The pulses are amplified by 41 dB across three amplifier stages. This configuration enables the generation of pulse peak powers of up to 305 W. Initial experiments with classic raster scanning and SLIDE scanning demonstrate the potential of this laser for high-speed GFP multiphoton microscopy using a pulse-on-demand method. ## Sensors (SENSORS) — 1 neu ### TexSortML: A Dataset and Baseline Evaluation for Garment Classification in Automated Textile Sorting ⚑ DE - **DOI:** [10.3390/s26196236](https://doi.org/10.3390/s26196236) - **Metadaten:** Erschienen: 2026-09-30 · Vol. 26, Issue 19, S. 6236 · OpenAlex seit 2026-10-02 - **DE-Institutionen:** Technische Hochschule Augsburg; University of Augsburg; Institut für Textiltechnik Augsburg (Germany) - **Autoren:** - Julian Schanz — Technische Hochschule Augsburg (DE) - Martin Kohnle — Technische Hochschule Augsburg (DE) - Fabian Kopf — Technische Hochschule Augsburg (DE) - Sebastian Geldhäuser — University of Augsburg (DE); Institut für Textiltechnik Augsburg (Germany) (DE) - Mesut Cetin — Technische Hochschule Augsburg (DE); University of Augsburg (DE); Institut für Textiltechnik Augsburg (Germany) (DE) - Alexandra Teynor — Technische Hochschule Augsburg (DE) - **Topics:** Textile materials and evaluations; 3D Shape Modeling and Analysis; Generative Adversarial Networks and Image Synthesis - **Abstract:** The growing volume of textile waste, driven by fast fashion and shortened garment lifespans, underscores the urgent need for scalable, automated sorting solutions. Progress is hindered by the absence of realistic, public datasets reflecting the variability of industrial textile sorting. We address this gap with TexSortML, a dataset of 751 post-consumer garments across 11 categories, captured under three controlled difficulty stages that simulate increasing handling disorder on a conveyor belt. On this dataset, we compare Convolutional Neural Networks (CNNs), Vision Transformers (ViTs), and CLIP-based zero-shot classifiers under five-fold cross-validation. A ViT fine-tuned directly on our data is the most accurate and most stable model at every stage, losing only 9.3 percentage points between the most and least controlled condition, against 16.6–16.9 points for the CNN baselines. The variant additionally pre-trained on external second-hand clothing data matches it only under controlled presentation and degrades sharply once garments are rotated or crumpled. CLIP-based zero-shot classifiers remain far behind and degrade the fastest. Rather than proposing a new architecture, this work contributes the dataset and a reproducible benchmark, and shows that extra training data can hurt when it is cleaner than the deployment condition. The dataset is publicly available on Zenodo.