The passively Q-switched mode-locked (QML) operation of a Tm:CaGdAlO₄ (Tm:CALGO) bulk laser pumped by a wavelength-tunable Ti:sapphire oscillator using molybdenum disulfide (MoS₂) as a saturable absorber was demonstrated. By using an output coupler with 3% transmittance, Q-switched mode-locked operation can be achieved with 3.56 W absorbed pump power. At a pump power of 6.77 W, laser pulses with the maximum average power of 1.44 W were obtained, corresponding to a slope efficiency of 21.3%. The laser delivered pulses centered at 1927 nm with a repetition frequency of 131.6 MHz. The experimental results confirm the promising application of the MoS₂ in high-power Q-switched mode-locked solid-state lasers at 2 µm.

Data-driven artificial intelligence systems effectively enhance accuracy and robustness by utilizing multi-view learning to aggregate consistent and complementary information from multi-source data. As one of the most important branches of multi-view learning, multi-view anchor clustering greatly reduces the time complexity via learning similarity graphs between anchors and data instead of data-to-data similarities, which has gained widespread attention in data-driven artificial intelligence domains. However, two issues still exist in current methods: (1) They commonly utilize orthogonal regularization to enhance anchor diversity, which may lead to a distorted anchor distribution, e.g., some clusters might have few or even no corresponding anchors. (2) They only utilize view-sharing anchors to aggregate complementary and consistent information between views, which may fail to ensure anchor robustness due to the heterogeneity gap between views. To this end, self-supervised, multi-view, semantics-aware anchor clustering (SMA2C) is proposed, containing multi-view representation alignment, adaptive anchor selection, and global spectral optimization. Specifically, SMA2C devises dual-level contrastive learning on representations and clustering partitioning between views within a deep encoding–decoding architecture to achieve multi-granularity alignment of views with the heterogeneity gap. Meanwhile, SMA2C introduces adaptive anchor selection via filtering outliers and refining clusters to enhance correlations between clusters and anchors, which ensures the diversity and discriminability of anchors. Finally, extensive evaluations across four real-world datasets confirm that SMA2C establishes a new benchmark for multi-view anchor clustering. In particular, SMA2C achieves a 6.75% improvement in accuracy over the second-best result on the HW dataset.

The degenerate optical parametric oscillator (OPO) is demonstrated to generate high-power, broadband mid-infrared MgO-doped periodically poled lithium niobate (MgO:PPLN) femtosecond laser at 151 MHz, synchronously pumped by a commercial Kerr-lens mode-locked Yb:KGW oscillator at 1028 nm. The average power of the degenerate OPO centered at 2056 nm is as high as 740 mW, which is the highest output power from a reported 2 μm degenerate femtosecond OPO, pumped by a bulk solid-state laser. The full width at half maximum (FWHM) spectral bandwidth of the degenerate OPO is 87.4 nm, corresponding to a theoretical, Fourier-limited pulse duration of 51 fs. These remarkable results indicate that degenerate OPO is a great potential candidate technology for generating high-power and few-cycle femtosecond pulses around 2 μm. Such mid-infrared sources are well-suited for high harmonic generation, a pumping source for mid- to far-infrared OPO.

From the perspective of neuromorphic dynamics, this paper constructs a novel Two-Dimensional Neural-Sine Coupled Hyperchaotic Map (2D-NSCHM) by embedding Mish and Swish activation functions into a sine-modulated coupling framework to introduce controllable nonlinear gradients. Unlike traditional polynomial maps prone to dynamical degradation due to simplified energy landscapes, the embedded neural activation functions introduce complex non-monotonic potential barriers. These barriers induce symmetry breaking in the phase space, effectively eliminating periodic windows and maximizing the system's global ergodicity and positive Lyapunov exponents. To bridge the gap between continuous dynamics and discrete cryptography while preserving the intrinsic entropy of the source, we introduce a hybrid post-processing quantization architecture comprising MurmurHash3 and infinite impulse response (IIR) feedback to simulate historical dependence in physical systems. This mechanism effectively suppresses finite-precision artifacts, ensuring the generated sequences strictly comply with the NIST SP 800-22 statistical test suite.Leveraging this robust entropy source, we design a resilient image encryption architecture that integrates a statistical multiplexing backup mechanism. This mechanism exploits the Alpha channel as a latent physical degree of freedom to embed low-frequency statistical components, effectively mapping localized physical perturbations (e.g., cropping) into a recoverable global distribution. Experimental results demonstrate that after global scrambling and tensor diffusion, the system achieves recognizable information reconstruction even under 75% data cropping or 70% salt-and-pepper noise. The proposed 2D-NSCHM, synthesized with its quantization strategy, demonstrates significant potential as a hardware-efficient entropy source, providing a robust framework for secure information processing and neuromorphic computing in lossy and stochastic physical environments.