arXiv Science⌕ Search

arXiv subjects

Wenjun Tang

Publications and source records attributed to Wenjun Tang.

8 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↗

ViCo-SAM3: Vision-Conditioned Alignment for Open-Vocabulary Camouflaged Object Segmentation

Open-vocabulary camouflaged object segmentation (OVCOS) aims to segment unseen camouflaged objects under text guidance. We observe that SAM3 still suffers from a pronounced semantic gap between global textual semantics and fine-grained pixel-level visual cues in OVCOS. Meanwhile, fully fine-tuning the text encoder introduces heavy parameter overhead and risks overfitting to training categories, which compromises open-vocabulary representation flexibility. To address these issues, we propose ViCo-SAM3, a Vision-Conditioned alignment framework designed for OVCOS. Specifically, we introduce vision-conditioned (ViCo) module, which dynamically modulates text embeddings with global visual context, enabling textual representations to adapt to the current image content and thereby effectively bridging the semantic gap between vision and text. Building on this, we further design a vision-conditioned cross-modal binding (ViCoBind) module to enhance cross-modal interaction and semantic alignment between visual and textual representations. Without bells and whistles, ViCo-SAM3 achieves state-of-the-art performance on the OVCamo benchmark and demonstrates strong generalization.

cs.CV↗

A 65nm 8b-Activation 8b-Weight SRAM-Based Charge-Domain Computing-in-Memory Macro Using A Fully-Parallel Analog Adder Network and A Single-ADC Interface

Performing data-intensive tasks in the von Neumann architecture is challenging to achieve both high performance and power efficiency due to the memory wall bottleneck. Computing-in-memory (CiM) is a promising mitigation approach by enabling parallel in-situ multiply-accumulate (MAC) operations within the memory with support from the peripheral interface and datapath. SRAM-based charge-domain CiM (CD-CiM) has shown its potential of enhanced power efficiency and computing accuracy. However, existing SRAM-based CD-CiM faces scaling challenges to meet the throughput requirement of high-performance multi-bit-quantization applications. This paper presents an SRAM-based high-throughput ReLU-optimized CD-CiM macro. It is capable of completing MAC and ReLU of two signed 8b vectors in one CiM cycle with only one A/D conversion. Along with non-linearity compensation for the analog computing and A/D conversion interfaces, this work achieves 51.2GOPS throughput and 10.3TOPS/W energy efficiency, while showing 88.6% accuracy in the CIFAR-10 dataset.

cs.AR↗

Smart Meters Integration in Distribution System State Estimation with Collaborative Filtering and Deep Gaussian Process

The problem of state estimations for electric distribution system is considered. A collaborative filtering approach is proposed in this paper to integrate the slow time-scale smart meter measurements in the distribution system state estimation, in which the deep Gaussian process is incorporated to infer the fast time-scale pseudo measurements and avoid anomalies. Numerical tests have demonstrated the higher estimation accuracy of the proposed method.

eess.SY↗

Appendix for Nonparametric Multivariate Probability Density Forecast in Smart Grids With Deep Learning

This paper proposes a nonparametric multivariate density forecast model based on deep learning. It not only offers the whole marginal distribution of each random variable in forecasting targets, but also reveals the future correlation between them. Differing from existing multivariate density forecast models, the proposed method requires no a priori hypotheses on the forecasted joint probability distribution of forecasting targets. In addition, based on the universal approximation capability of neural networks, the real joint cumulative distribution functions of forecasting targets are well-approximated by a special positive-weighted deep neural network in the proposed method. Numerical tests from different scenarios were implemented under a comprehensive verification framework for evaluation, including the very short-term forecast of the wind speed, wind power, and the day-ahead forecast of the aggregated electricity load. Testing results corroborate the superiority of the proposed method over current multivariate density forecast models considering the accordance with reality, prediction interval width, and correlations between different random variables.

eess.SY↗

Data-Driven Online Interactive Bidding Strategy for Demand Response

Demand response (DR), as one of the important energy resources in the future's grid, provides the services of peak shaving, enhancing the efficiency of renewable energy utilization with a short response period, and low cost. Various categories of DR are established, e.g. automated DR, incentive DR, emergency DR, and demand bidding. However, with the practical issue of the unawareness of residential and commercial consumers' utility models, the researches about demand bidding aggregator involved in the electricity market are just at the beginning stage. For this issue, the bidding price and bidding quantity are two required decision variables while considering the uncertainties due to the market and participants. In this paper, we determine the bidding and purchasing strategy simultaneously employing the smart meter data and functions. A two-agent deep deterministic policy gradient method is developed to optimize the decisions through learning historical bidding experiences. The online learning further utilizes the daily newest bidding experience attained to ensure trend tracing and self-adaptation. Two environment simulators are adopted for testifying the robustness of the model. The results prove that when facing diverse situations the proposed model can earn the optimal profit via off/online learning the bidding rules and robustly making the proper bid.

cs.AI↗

Machine Learning Approach to Uncovering Residential Energy Consumption Patterns Based on Socioeconomic and Smart Meter Data

The smart meter data analysis contributes to better planning and operations for the power system. This study aims to identify the drivers of residential energy consumption patterns from the socioeconomic perspective based on the consumption and demographic data using machine learning. We model consumption patterns by representative loads and reveal the relationship between load patterns and socioeconomic characteristics. Specifically, we analyze the real-world smart meter data and extract load patterns by clustering in a robust way. We further identify the influencing socioeconomic attributes on load patterns to improve our method's interpretability. The relationship between consumers' load patterns and selected socioeconomic features is characterized via machine learning models. The findings are as follows. (1) Twelve load clusters, consisting of six for weekdays and six for weekends, exhibit a diverse pattern of lifestyle and a difference between weekdays and weekends. (2) Among various socioeconomic features, age and education level are suggested to influence the load patterns. (3) Our proposed analytical model using feature selection and machine learning is proved to be more effective than XGBoost and conventional neural network model in mapping the relationship between load patterns and socioeconomic features.

cs.LG↗

A Multivariate Density Forecast Approach for Online Power System Security Assessment

A multivariate density forecast model based on deep learning is designed in this paper to forecast the joint cumulative distribution functions (JCDFs) of multiple security margins in power systems. Differing from existing multivariate density forecast models, the proposed method requires no a priori hypotheses on the distribution of forecasting targets. In addition, based on the universal approximation capability of neural networks, the value domain of the proposed approach has been proven to include all continuous JCDFs. The forecasted JCDF is further employed to calculate the deterministic security assessment index evaluating the security level of future power system operations. Numerical tests verify the superiority of the proposed method over current multivariate density forecast models. The deterministic security assessment index is demonstrated to be more informative for operators than security margins as well.

eess.SY↗