Fanjiang Xu

High-energy, ultrafast pulse generation from mid-infrared (MIR) fiber lasers is often limited by pulse splitting caused by excessive nonlinear phase accumulation, especially in anomalous-dispersion cavities. Dissipative soliton resonance (DSR) offers a route to energy scaling by accommodating additional gain through temporal pulse broadening under peak-power clamping. Here, we numerically demonstrate DSR operation in a net-normal-dispersion Er³⁺: ZBLAN mode-locked fiber laser at 2.8 µm, using an As₂S₃ fiber for dispersion and nonlinear management. Systematic parameter sweeps show that the gain saturation energy primarily governs pulse-energy scaling, whereas the output-coupling ratio controls peak-power extraction and the operation regime. The spectral filter bandwidth and saturable absorber parameters define the stability window and mediate transitions among dissipative solitons, DSR pulses, noise-like pulses, multi-pulse states, and unstable operation. After coordinated cavity optimization, a stable rectangular DSR pulse with a maximum energy of 408.52 nJ is obtained under a 55 nm filter bandwidth, relaxing the narrow-filtering requirement reported in previous MIR DSR designs. A high-peak-power DSR state with 3112.1 W peak power and 316.58 nJ energy is also achieved at a 95% output-coupling ratio. By tuning the As₂S₃ fiber length, DSR dynamics are also accessed near the zero-dispersion boundary and in the anomalous-dispersion regime, where spike-on-pedestal temporal profiles and dual-peak spectra emerge. This work advances the understanding of MIR DSR dynamics and offers design guidance for compact, high-energy ultrafast sources at 2.8 µm.

Moment estimation for stochastic differential equations (SDEs) is fundamental to the formal reasoning and verification of stochastic dynamical systems, yet remains challenging and is rarely available in closed form. In this paper, we study time-homogeneous SDEs with polynomial drift and diffusion, and investigate when their moments can be computed exactly. We formalize the notion of moment-solvable SDEs and propose a generic symbolic procedure that, for a given monomial, attempts to construct a finite-dimensional linear ordinary differential equation (ODE) system governing its moment, thereby enabling exact computation. We introduce a syntactic class of pro-solvable SDEs, characterized by a block-triangular structure, and prove that all polynomial moments of any pro-solvable SDE admit such finite ODE representations. This class strictly generalizes linear SDEs and includes many nonlinear models. Experimental results demonstrate the effectiveness of our approach.

The World Wide Web needs reliable predictive capabilities to respond to changes in user behavior and usage patterns. Time series forecasting (TSF) is a key means to achieve this goal. In recent years, the large language models (LLMs) for TSF (LLM4TSF) have achieved good performance. However, there is a significant difference between pretraining corpora and time series data, making it hard to guarantee forecasting quality when directly applying LLMs to TSF; fine-tuning LLMs can mitigate this issue, but often incurs substantial computational overhead. Thus, LLM4TSF faces a dual challenge of prediction performance and compute overhead. To address this, we aim to explore a method for improving the forecasting performance of LLM4TSF while freezing all LLM parameters to reduce computational overhead. Inspired by in-context learning (ICL), we propose LVICL. LVICL uses our vector-injected ICL to inject example information into a frozen LLM, eliciting its in-context learning ability and thereby enhancing its performance on the example-related task (i.e., TSF). Specifically, we first use the LLM together with a learnable context vector adapter to extract a context vector from multiple examples adaptively. This vector contains compressed, example-related information. Subsequently, during the forward pass, we inject this vector into every layer of the LLM to improve forecasting performance. Compared with conventional ICL that adds examples into the prompt, our vector-injected ICL does not increase prompt length; moreover, adaptively deriving a context vector from examples suppresses components harmful to forecasting, thereby improving model performance. Extensive experiments demonstrate the effectiveness of our approach.

Image restoration represents a promising approach for addressing the inherent defects of image content distortion. Standard image restoration approaches suffer from high storage cost and the requirement towards the known degradation pattern, including type and degree, which can barely be satisfied in dynamic practical scenarios. In contrast, all-in-one image restoration (AiOIR) eliminates multiple degradations within a unified model to circumvent the aforementioned issues. However, according to our causal analysis, we disclose that two significant defects still exacerbate the effectiveness and generalization of AiOIR models: 1) the spurious correlation between non-degradation semantic features and degradation patterns; 2) the biased estimation of degradation patterns. To obtain the true causation between degraded images and restored images, we propose Causal-deconfounding Wavelet-disentangled Prompt Network (CWP-Net) to perform effective AiOIR. CWP-Net introduces two modules for decoupling, i.e., wavelet attention module of encoder and wavelet attention module of decoder. These modules explicitly disentangle the degradation and semantic features to tackle the issue of spurious correlation. To address the issue stemming from the biased estimation of degradation patterns, CWP-Net leverages a wavelet prompt block to generate the alternative variable for causal deconfounding. Extensive experiments on two all-in-one settings prove the effectiveness and superior performance of our proposed CWP-Net over the state-of-the-art AiOIR methods.