Kaichao Liang

PSBC LLM Team, Huawei LLM Team, Ruihan Long et al.

Large language models (LLMs) drive significant financial innovations, yet their high-concurrency deployment is severely bottlenecked by KV cache memory overhead, which inflates infrastructure costs and throttles scalability. To address this, we propose YouZhi-LLM, a highly efficient financial LLM empowered by a comprehensive structural transition and training pipeline natively built on the Huawei Ascend ecosystem. At its algorithmic core, YouZhi-LLM features a layer-adaptive GQA-to-MLA transition framework that dynamically assigns per-layer FreqFold sizes, maximizing KV-cache compression while minimizing perplexity degradation. To recover representation capacity and inject domain expertise, the Ascend-based training pipeline seamlessly integrates generalized knowledge distillation with financial-specific supervised fine-tuning. Evaluations demonstrate the superiority of this systematic approach, with the adaptive transition reducing perplexity degradation by up to 35% over uniform baselines. Crucially, when evaluated on Ascend NPUs via vLLM-Ascend, the massive KV-cache reduction translates directly into deployment efficiency. Compared to their respective base models, YouZhi-7B yields a 12.3% improvement in average financial benchmark score alongside a 2.69$\times$ increase in maximum concurrency; similarly, YouZhi-14B achieves a 7.0% accuracy gain and a 2.43$\times$ concurrency boost, establishing a new paradigm for cost-effective, high-throughput financial inference.

Mingyang Liu, Qingcan Kang, Yuke Wang et al.

Incorporating news into time series forecasting is appealing because news can reveal abrupt exogenous events that historical values alone cannot recover. However, existing LLM-based news-forecasting pipelines face two practical limitations: relevant news articles often exceed the model's context window, and iterative retrieval of supplementary news is typically unguided, leading to redundant updates and slow convergence. We address these issues with a novel framework that combines importance-aware news compression and process-level retrieval supervision. First, we train an importance reward model that estimates the forecasting utility of each article and uses this signal to allocate compression budgets during sequential pairwise fusion, preserving informative content within a fixed context limit. Second, we introduce a process reward model (PRM) that ranks multiple supplementary-news candidates conditioned on the current error profile and the history of previously selected articles, replacing one-shot blind retrieval with quality-controlled selection. Both components are trained offline using historical data with ground truth; inference uses the frozen filtering logic and compression modules without any reflection loop. Experiments on finance, energy, traffic, and bitcoin forecasting benchmarks show that our method improves prediction accuracy over strong baselines, significantly reduces the number of refinement iterations compared to the iterative baseline, and remains effective when relevant articles span thousands of tokens.

Da Chen, Guangyu Hu, Kaihong Xu et al.

Extracting high-fidelity 2D contours from Scanning Electron Microscope (SEM) images is critical for calibrating Optical Proximity Correction (OPC) models. While foundation models like Segment Anything 2 (SAM2) are promising, adapting them to specialized domains with scarce annotated data is a major challenge. This paper presents a case study on adapting SAM2 for SEM contour extraction in a few-shot setting. We propose SegSEM, a framework built on two principles: a data-efficient fine-tuning strategy that adapts by selectively training only the model's encoders, and a robust hybrid architecture integrating a traditional algorithm as a confidence-aware fallback. Using a small dataset of 60 production images, our experiments demonstrate this methodology's viability. The primary contribution is a methodology for leveraging foundation models in data-constrained industrial applications.

Zixiao Wang, Jieya Zhou, Su Zheng et al.

Recent decades have witnessed remarkable advancements in artificial intelligence (AI), including large language models (LLMs), image and video generative models, and embodied AI systems. These advancements have led to an explosive increase in the demand for computational power, challenging the limits of Moore's Law. Optical lithography, a critical technology in semiconductor manufacturing, faces significant challenges due to its high costs. To address this, various lithography simulators have been developed. However, many of these simulators are limited by their inadequate photoresist modeling capabilities. This paper presents TorchResist, an open-source, differentiable photoresist simulator.TorchResist employs an analytical approach to model the photoresist process, functioning as a white-box system with at most twenty interpretable parameters. Leveraging modern differentiable programming techniques and parallel computing on GPUs, TorchResist enables seamless co-optimization with other tools across multiple related tasks. Our experimental results demonstrate that TorchResist achieves superior accuracy and efficiency compared to existing solutions. The source code is publicly available.

Muge Du, Kaichao Liang, Yinong Liu et al.

Metal artefacts in CT images may disrupt image quality and interfere with diagnosis. Recently many deep-learning-based CT metal artefact reduction (MAR) methods have been proposed. Current deep MAR methods may be troubled with domain gap problem, where methods trained on simulated data cannot perform well on practical data. In this work, we experimentally investigate two image-domain supervised methods, two dual-domain supervised methods and two image-domain unsupervised methods on a dental dataset and a torso dataset, to explore whether domain gap problem exists or is overcome. We find that I-DL-MAR and DudoNet are effective for practical data of the torso dataset, indicating the domain gap problem is solved. However, none of the investigated methods perform satisfactorily on practical data of the dental dataset. Based on the experimental results, we further analyze the causes of domain gap problem for each method and dataset, which may be beneficial for improving existing methods or designing new ones. The findings suggest that the domain gap problem in deep MAR methods remains to be addressed.