Searcharxiv⌕ Search

arXiv subjects

Shuanghong Li

Publications and source records attributed to Shuanghong Li.

4 recordsLinked to original sources

Reliable Evaluation of AI Agents for Investment Research: A Framework for Performance and Validity Assessment

We consider the problem of evaluating AI agents undertaking investment-research tasks, where agents interact with external information sources and computational tools to produce financial analyses and research reports. An important challenge is that the quality of an agent's final output does not necessarily reflect the quality of its execution process, while the reliability of the resulting assessment also depends on the validity of the evaluation procedure. Furthermore, investment-research tasks differ considerably in their requirements for numerical accuracy, information retrieval, financial analysis and supporting evidence, making it difficult to assess agent performance using a uniform set of criteria. We propose a domain-oriented evaluation framework that distinguishes outcome quality from execution quality and accommodates these differences through task-specific criteria, deterministic validators and independent large language model judges. We also introduce an evaluation-validity mechanism that examines the integrity of evaluation inputs, reference information, supporting evidence and grading outputs, with expert review used to resolve identified deficiencies and significant grading disagreements. Numerical results show that the proposed framework can distinguish differences in agent performance across evaluation dimensions, while identifying potential reliability issues in the evaluation process. The associated implementation and evaluation materials will be made available on GitHub.

cs.AI↗

The Ubiquity of White-Light Flares on the Sun

White-light flares (WLFs) have long been considered rare on the Sun, as most magnetic reconnections occur in the corona, making it difficult to transport sufficient energy downward to heat the very low layers for WL continuum emission. However, it is also argued that WL continuum emission is a common feature of all flares, and the rare detection of them is largely due to the limited sensitivity and resolution of telescopes. Recently, we developed a new method to identify WLFs, which exhibits extraordinary detection capability. Combined with SDO/HMI's over 15 years of continuous photospheric observations, it provides us an unprecedented opportunity to address the long-standing debate about WLF ubiquity. A total of 1951 flares (547 B-class, 1316 C-class, and 88 M-class) from 769 low-activity active regions (ARs; flare number $<$ 10) observed during January 2011 - May 2025 are investigated. Among these, 1012 are identified as WLFs, giving an overall occurrence rate of 51.9\%. The occurrence rates for B-, C-, and M-class flares are 36.9\%, 56.9\%, and 85.2\%, respectively, showing a clear increase with flare class. The occurrence rate is also highly correlated with the AR's sunspot number. For the first time, we find that WL continuum radiation is widely present even among B-class flares. Different WLFs contain varying numbers of WL kernels, which exhibit distinct properties, evolutionary behaviors and different relationships with GOES soft X-ray flux and its time derivative. These results strongly suggest that WLFs are a common phenomenon on the Sun.

astro-ph.SR↗

Detection and analysis of white-light emission in solar flares through light curve diagnostics

White-light flares (WLFs) are crucial for understanding the energy transport and heating processes in the lower solar atmosphere. Systematic studies are highly necessary. However, most WLFs are very weak and difficult to detect. To address this, we propose a new method of detecting WLFs. Through the observations of SDO/HMI, the light curve of each pixel in the flaring region can be obtained. By subtracting the slowly varying background, we obtained a series of rapidly varying radiative pulses. Pixels for which radiative pulses during flares significantly exceed those occurring before and after the flare were identified as WL emission regions. We applied our method to the detection of the X2.2 flare on September 6, 2017 and validated the method. We found that the WL emission in this flare exhibits two phases, and that different regions show distinct WL emission properties. We also detected the WL emission in all the flares (1 X-class, 2 M-class, and 20 C-class) occurred in active region NOAA 12887. It was found that 15 of the 23 flares are WLFs (1 X-class, 2 M-class, and 12 C-class). The occurrence rate of WLFs in this active region is $\sim65\%$. Surprisingly, the occurrence rate of WLFs in C-class flares even reaches up to $60\%$. It should be noted that most of these C-class WLFs are below C5.0. In addition, a C1.0 WLF was identified; this is the lowest GOES-class event with confirmed WL emission to date. These results demonstrate that WL emission is ubiquitous in most flares, even down to C-class events.

astro-ph.SR↗

Statistics of Matrix Elements of Operators in a Disorder-Free SYK model

Recently, studies have explored the statistics of matrix elements of local operators in the Lieb-Liniger model. It was found that the probability distribution function for off-diagonal matrix elements $\langle \boldsymbolμ|\mathcal{O}|\boldsymbolλ \rangle$ within the same macro-state is well described by the Fréchet distributions. This represents a significant development for the Eigenstate Thermalization Hypothesis (ETH). In this paper, we investigate a similar phenomenon in another solvable model: the disorder-free Sachdev-Ye-Kitaev (SYK) model. The Hamiltonian of this model consists of 4-body interactions of Majorana fermions. Unlike the conventional SYK model, the coupling strengths in this model are fixed to a constant, earning it the name ``disorder-free.'' We evaluate the matrix elements of operators constructed from products of $n$ Majorana fermions: $\mathcal{O} = χ_{a_1}χ_{a_2}\ldots χ_{a_n}$. For a general choice of indices and $n \geq 4$, we find that the statistics of the off-diagonal matrix elements are well-fitted by a generalized inverse Gaussian distribution rather than Fréchet distributions.

cond-mat.stat-mech↗