Weikang Zhang, Alison Watts
Nov 19, 2025·q-fin.ST·PDF Crypto enthusiasts claim that buying and holding crypto assets yields high returns, often citing Bitcoin's past performance to promote other tokens and fuel fear of missing out. However, understanding the real risk-return trade-off and what factors affect future crypto returns is crucial as crypto becomes increasingly accessible to retail investors through major brokerages. We examine the HODL strategy through two independent analyses. First, we implement 480 million Monte Carlo simulations across 378 non-stablecoin crypto assets, net of trading fees and the opportunity cost of 1-month Treasury bills, and find strong evidence of survivorship bias and extreme downside concentration. At the 2-3 year horizon, the median excess return is -28.4 percent, the 1 percent conditional value at risk indicates that tail scenarios wipe out principal after all costs, and only the top quartile achieves very large gains, with a mean excess return of 1,326.7 percent. These results challenge the HODL narrative: across a broad set of assets, simple buy-and-hold loads extreme downside risk onto most investors, and the miracles mostly belong to the luckiest quarter. Second, using a Bayesian multi-horizon local projection framework, we find that endogenous predictors based on realized risk-return metrics have economically negligible and unstable effects, while macro-finance factors, especially the 24-week exponential moving average of the Fear and Greed Index, display persistent long-horizon impacts and high cross-basket stability. Where significant, a one-standard-deviation sentiment shock reduces forward top-quartile mean excess returns by 15-22 percentage points and median returns by 6-10 percentage points over 1-3 year horizons, suggesting that macro-sentiment conditions, rather than realized return histories, are the dominant indicators for future outcomes.
Matěj Hejda, Ekaterina Malysheva, Dafydd Owen-Newns, Qusay Raghib Ali Al-Taai, Weikang Zhang, Ignacio Ortega-Piwonka, Julien Javaloyes, Edward Wasige, Victor Dolores-Calzadilla, José M. L. Figueiredo, Bruno Romeira, Antonio Hurtado
Excitable optoelectronic devices represent one of the key building blocks for implementation of artificial spiking neurons in neuromorphic (brain-inspired) photonic systems. This work introduces and experimentally investigates an opto-electro-optical (O/E/O) artificial neuron built with a resonant tunnelling diode (RTD) coupled to a photodetector as a receiver and a vertical cavity surface emitting laser as a the transmitter. We demonstrate a well defined excitability threshold, above which this neuron produces 100 ns optical spiking responses with characteristic neural-like refractory period. We utilise its fan-in capability to perform in-device coincidence detection (logical AND) and exclusive logical OR (XOR) tasks. These results provide first experimental validation of deterministic triggering and tasks in an RTD-based spiking optoelectronic neuron with both input and output optical (I/O) terminals. Furthermore, we also investigate in theory the prospects of the proposed system for its nanophotonic implementation with a monolithic design combining a nanoscale RTD element and a nanolaser; therefore demonstrating the potential of integrated RTD-based excitable nodes for low footprint, high-speed optoelectronic spiking neurons in future neuromorphic photonic hardware.
Qusay Raghib Ali Al-Taai, Matěj Hejda, Weikang Zhang, Bruno Romeira, José M. L. Figueiredo, Edward Wasige, Antonio Hurtado
This work reports a nanostructure resonant tunnelling diode-photodetector (RTD-PD) device and demonstrates its operation as a controllable, optically-triggered excitable spike generator. The top contact layer of the device is designed with a nanopillar structure 500 nm in diameter) to restrain the injection current, yielding therefore lower energy operation for spike generation. We demonstrate experimentally the deterministic optical triggering of controllable and repeatable neuron-like spike patterns in the nanostructure RTD-PDs. Moreover, we show the device's ability to deliver spiking responses when biased in both regions adjacent to the negative differential conductance (NDC) region, the so-called 'peak' and 'valley' points of the current-voltage ($I$-$V$) characteristic. This work also demonstrates experimentally key neuron-like dynamical features in the nanostructure RTD-PD, such as a well-defined threshold (in input optical intensity) for spike firing, as well as the presence of spike firing refractory time. The optoelectronic and chip-scale character of the proposed system together with the deterministic, repeatable and well controllable nature of the optically-elicited spiking responses render this nanostructure RTD-PD element as a highly promising solution for high-speed, energy-efficient optoelectronic artificial spiking neurons for novel light-enabled neuromorphic computing hardware.
Weikang Zhang, Zimo Zhu, Zhichuan Yang, Chen Huang, Wenqiang Lei, See-Kiong Ng
Simulating Standardized Patients with cognitive impairment offers a scalable and ethical solution for clinical training. However, existing methods rely on discrete prompt engineering and fail to capture the heterogeneity of deficits across varying domains and severity levels. To address this limitation, we propose StsPatient for the fine-grained simulation of cognitively impaired patients. We innovatively capture domain-specific features by extracting steering vectors from contrastive pairs of instructions and responses. Furthermore, we introduce a Stochastic Token Modulation (STM) mechanism to regulate the intervention probability. STM enables precise control over impairment severity while mitigating the instability of conventional vector methods. Comprehensive experiments demonstrate that StsPatient significantly outperforms baselines in both clinical authenticity and severity controllability.
Weikang Zhang, Matěj Hejda, Qusay Raghib Ali Al-Taai, Dafydd Owen-Newns, Bruno Romeira, José M. L. Figueiredo, Joshua Robertson, Edward Wasige, Antonio Hurtado
We report a multi-modal spiking neuron that allows optical and electronic input and control, and wavelength-multiplexing operation, for use in novel high-speed neuromorphic sensing and computing functionalities. The photonic-electronic neuron is built with a micro-scale, nanostructure resonant tunnelling diode (RTD) with photodetection (PD) capability. Leveraging the advantageous intrinsic properties of this RTD-PD system, namely highly nonlinear characteristics, photo-sensitivity, light-induced I-V curve shift, and the ability to deliver excitable responses under electrical and optical inputs, we successfully achieve flexible neuromorphic spike activation and inhibition regimes through photonic-electrical control. We also demonstrate the ability of this RTD-PD spiking sensing-processing neuron to operate under the simultaneous arrival of multiple wavelength-multiplexed optical signals, due to its large photodetection spectral window (covering the 1310 and 1550 nm telecom wavelength bands). Our results highlight the potential of RTD photonic-electronic neurons to reproduce multiple key excitatory and inhibitory spiking regimes, at high speed (ns-rate spiking responses, with faster sub-ns regimes theoretically predicted) and low energy (requiring only ~10 mV and ~150 microW, electrical and optical input amplitudes, respectively), similar in nature to those commonly found in the biological neurons of the visual system and the brain. This work offers a highly promising approach for the realisation of high-speed, energy-efficient photonic-electronic spiking neurons and spiking neural networks, enabling multi-modal and multi-wavelength operation for sensing and information processing tasks. This work therefore paves the way for innovative high-speed, photonic-electronic, and spike-based neuromorphic sensing and computing systems and artificial intelligence hardware.
Xiaomi LLM-Core Team, :, Bangjun Xiao, Bingquan Xia, Bo Yang, Bofei Gao, Bowen Shen, Chen Zhang, Chenhong He, Chiheng Lou, Fuli Luo, Gang Wang, Gang Xie, Hailin Zhang, Hanglong Lv, Hanyu Li, Heyu Chen, Hongshen Xu, Houbin Zhang, Huaqiu Liu, Jiangshan Duo, Jianyu Wei, Jiebao Xiao, Jinhao Dong, Jun Shi, Junhao Hu, Kainan Bao, Kang Zhou, Lei Li, Liang Zhao, Linghao Zhang, Peidian Li, Qianli Chen, Shaohui Liu, Shihua Yu, Shijie Cao, Shimao Chen, Shouqiu Yu, Shuo Liu, Tianling Zhou, Weijiang Su, Weikun Wang, Wenhan Ma, Xiangwei Deng, Bohan Mao, Bowen Ye, Can Cai, Chenghua Wang, Chengxuan Zhu, Chong Ma, Chun Chen, Chunan Li, Dawei Zhu, Deshan Xiao, Dong Zhang, Duo Zhang, Fangyue Liu, Feiyu Yang, Fengyuan Shi, Guoan Wang, Hao Tian, Hao Wu, Heng Qu, Hongfei Yi, Hongxu An, Hongyi Guan, Xing Zhang, Yifan Song, Yihan Yan, Yihao Zhao, Yingchun Lai, Yizhao Gao, Yu Cheng, Yuanyuan Tian, Yudong Wang, Zhen Tang, Zhengju Tang, Zhengtao Wen, Zhichao Song, Zhixian Zheng, Zihan Jiang, Jian Wen, Jiarui Sun, Jiawei Li, Jinlong Xue, Jun Xia, Kai Fang, Menghang Zhu, Nuo Chen, Qian Tu, Qihao Zhang, Qiying Wang, Rang Li, Rui Ma, Shaolei Zhang, Shengfan Wang, Shicheng Li, Shuhao Gu, Shuhuai Ren, Sirui Deng, Tao Guo, Tianyang Lu, Weiji Zhuang, Weikang Zhang, Weimin Xiong, Wenshan Huang, Wenyu Yang, Xin Zhang, Xing Yong, Xu Wang, Xueyang Xie, Yilin Jiang, Yixin Yang, Yongzhe He, Yu Tu, Yuanliang Dong, Yuchen Liu, Yue Ma, Yue Yu, Yuxing Xiang, Zhaojun Huang, Zhenru Lin, Zhipeng Xu, Zhiyang Chen, Zhonghua Deng, Zihan Zhang, Zihao Yue
We present MiMo-V2-Flash, a Mixture-of-Experts (MoE) model with 309B total parameters and 15B active parameters, designed for fast, strong reasoning and agentic capabilities. MiMo-V2-Flash adopts a hybrid attention architecture that interleaves Sliding Window Attention (SWA) with global attention, with a 128-token sliding window under a 5:1 hybrid ratio. The model is pre-trained on 27 trillion tokens with Multi-Token Prediction (MTP), employing a native 32k context length and subsequently extended to 256k. To efficiently scale post-training compute, MiMo-V2-Flash introduces a novel Multi-Teacher On-Policy Distillation (MOPD) paradigm. In this framework, domain-specialized teachers (e.g., trained via large-scale reinforcement learning) provide dense and token-level reward, enabling the student model to perfectly master teacher expertise. MiMo-V2-Flash rivals top-tier open-weight models such as DeepSeek-V3.2 and Kimi-K2, despite using only 1/2 and 1/3 of their total parameters, respectively. During inference, by repurposing MTP as a draft model for speculative decoding, MiMo-V2-Flash achieves up to 3.6 acceptance length and 2.6x decoding speedup with three MTP layers. We open-source both the model weights and the three-layer MTP weights to foster open research and community collaboration.
Yan Xie, Changkui Mao, Changsong Wu, Chao Lu, Chao Suo, Cheng Qian, Chun Yang, Danyang Zhu, Hengchang Xiong, Hongzhan Lu, Hongzhen Liu, Jiafu Liu, Jie Chen, Jie Dai, Junfeng Tang, Kai Liu, Kun Li, Lipeng Ge, Meng Sun, Min Luo, Peng Chen, Peng Wang, Shaodong Yang, Shibin Tang, Shibo Chen, Weikang Zhang, Xiao Ling, Xiaobo Du, Xin Wu, Yang Liu, Yi Jiang, Yihua Jin, Yin Huang, Yuli Zhang, Zhen Yuan, Zhiyuan Man, Zhongxiao Yao
As deep learning-based AI technologies gain momentum, the demand for general-purpose AI computing architectures continues to grow. While GPGPU-based architectures offer versatility for diverse AI workloads, they often fall short in efficiency and cost-effectiveness. Various Domain-Specific Architectures (DSAs) excel at particular AI tasks but struggle to extend across broader applications or adapt to the rapidly evolving AI landscape. M100 is Li Auto's response: a performant, cost-effective architecture for AI inference in Autonomous Driving (AD), Large Language Models (LLMs), and intelligent human interactions, domains crucial to today's most competitive automobile platforms. M100 employs a dataflow parallel architecture, where compiler-architecture co-design orchestrates not only computation but, more critically, data movement across time and space. Leveraging dataflow computing efficiency, our hardware-software co-design improves system performance while reducing hardware complexity and cost. M100 largely eliminates caching: tensor computations are driven by compiler- and runtime-managed data streams flowing between computing elements and on/off-chip memories, yielding greater efficiency and scalability than cache-based systems. Another key principle was selecting the right operational granularity for scheduling, issuing, and execution across compiler, firmware, and hardware. Recognizing commonalities in AI workloads, we chose the tensor as the fundamental data element. M100 demonstrates general AI computing capability across diverse inference applications, including UniAD (for AD) and LLaMA (for LLMs). Benchmarks show M100 outperforms GPGPU architectures in AD applications with higher utilization, representing a promising direction for future general AI computing.
Samiha Tariq, Weikang Zhang
This study explores the interdependent relationship between consumer credit and consumer confidence in the United States using monthly data from January 1978 to August 2024. Utilizing a Vector Error Correction Model (VECM), the analysis focuses on the interplay between household borrowing behaviour and consumer sentiment while controlling for macroeconomic factors such as interest rates, inflation, unemployment, and money supply. The results reveal a stable long-run equilibrium: heightened consumer confidence is associated with increased credit utilization, reflecting greater financial optimism among households. In the short run, shifts in consumer confidence exert relatively modest immediate influence on credit usage, whereas consumer credit adjusts slowly, displaying significant inertia. Impulse-response analysis confirms that shocks to consumer confidence generate sustained positive effects on borrowing, while unexpected increases in credit initially depress sentiment but only fleetingly. These findings underscore the critical role of the relationship between consumer confidence and credit-market dynamics and highlight its policy relevance for fostering balanced and stable household finances.
Yaqi Ma, Meizhen Huang, Xu Zhang, Weixiong Hu, Zishu Zhou, Kai Feng, Wenhui Li, Yong Chen, Chenxuan Lou, Weikang Zhang, Haoxi Ji, Yibo Wang, Zefei Wu, Xiaodong Cui, Wang Yao, Shichao Yan, Zi Yang Meng, Ning Wang
Two-dimensional electron systems in both magnetic fields and periodic potentials are described by Hofstadter butterfly, a fundamental problem of solid-state physics. While moiré systems provide a powerful method to realize this spectrum, previous experiments, however, have been limited to fractional flux quanta regime due to the difficulty of building ~ 50 nm periodic modulations. Here, we demonstrate a super-moiré strategy to overcome this challenge. By aligning monolayer graphene (G) with 1.0° twisted hexagonal boron nitride (t-hBN), a 63.2 nm bichromatic G/t-hBN super-moiré is constructed, made possible by exploiting the electrostatic nature of t-hBN potential. Under magnetic field B, magnetic Bloch states at integer flux quanta (1-9) are achieved and observed as integer Brown-Zak oscillations, expanding the flux quanta from factions to integers. Theoretical analysis reproduces these experimental findings. This work opens new avenues to study unexplored Hofstadter butterfly, explore emergent topological order at integer flux quanta and engineer long-wavelength periodic modulations.
Jincheng Guo, Xiaofeng Wang, Qichun Liu, Alexei V. Filippenko, Thomas G. Brink, Jingkun Zhao, WeiKang Zhang, Yi Yang, Jie Lin, Haowei Peng, Hailiang Chen, Davron O. Mirzaqulov, Shuhrat A. Ehgamberdiev, Bin Ma, Jun Mo, Cheng Liu, Gaobo Xi, Xiaojun Jiang, Danfeng Xiang, Jicheng Zhang
Jan 15, 2026·astro-ph.SR·PDF We present a physical characterization of TMTS J00063798+3104160 (J0006), a rapidly rotating,ultra-massive white dwarf (WD) identified in high-cadence light curves from the Tsinghua University-Ma Huateng Telescope for Survey (TMTS). A coherent 23-minute periodicity is detected in TMTS, TESS, and ZTF photometry. A time series of low-resolution spectra with the Keck-I 10 m telescope reveals broad, shallow hydrogen absorption features indicative of an extreme magnetic field and shows no evidence for radial-velocity variations. Atmospheric modeling yields a magnetic field strength of $\sim$ 250 MG, while Gaia astrometry and photometry imply a mass of 1.06 $\pm$ 0.01 M$_{\odot}$. A significant infrared excess is detected in the WISE W1 band and is well fitted by a 550 K blackbody, likely arising from residual material of a merger. We interpret the 23-minute photometric modulation as the rotation period of an isolated, massive WD formed likely through the merger of a double WD binary. With one of the shortest rotation periods known among candidate merger remnants and with constraints from a deep Einstein Probe X-ray nondetection, J0006 provides a rare and important observational window into the poorly explored intermediate stages of post-merger evolution.