arXiv subjects

Keyi Wang

Publications and source records attributed to Keyi Wang.

At least 19 recordsLinked to original sources

arXiv2026

Day-Ahead Electricity Price Forecasting Using a Multivariate Group Lasso Method

Electricity price signals in modern power systems exhibit complex dependence structures that render forecasting inherently challenging. Our analysis of real-world pricing signals from the California Independent System Operator (CAISO) reveals complex temporal group effects, whereby the influence of explanatory variables on electricity prices persists across consecutive blocks of time due to underlying economic and operational drivers. In response, we propose a multivariate statistical method based on a Group Lasso formulation to forecast the vector of day-ahead electricity prices, by leveraging multi-feature temporal group effects. Our approach is evaluated on two full years of electricity prices from CAISO, demonstrating considerable improvements in point and probabilistic forecast metrics compared to a wide array of statistical and deep learning methods. Theoretical and empirical analyses confirm the effectiveness of the proposed approach in modeling realistic group effects, maintaining both interpretability and low computational complexity. When retrospectively evaluated on test data from a recent international electricity price forecasting challenge, the proposed method ranked in second place, despite having access to significantly less information than competing approaches. Finally, the proposed method is independently validated against two operational electricity price forecasting systems in CAISO, demonstrating competitive predictive performance and practical relevance.

stat.AP↗

arXiv2026

Multi-Gait Learning for Humanoid Robots Using Reinforcement Learning with Selective Adversarial Motion Prior

Learning diverse locomotion skills for humanoid robots in a unified reinforcement learning framework remains challenging due to the conflicting requirements of stability and dynamic expressiveness across different gaits. We present a multi-gait learning approach that enables a humanoid robot to master five distinct gaits -- walking, goose-stepping, running, stair climbing, and jumping -- using a consistent policy structure, action space, and reward formulation. The key contribution is a selective Adversarial Motion Prior (AMP) strategy: AMP is applied to periodic, stability-critical gaits (walking, goose-stepping, stair climbing) where it accelerates convergence and suppresses erratic behavior, while being deliberately omitted for highly dynamic gaits (running, jumping) where its regularization would over-constrain the motion. Policies are trained via PPO with domain randomization in simulation and deployed on a physical 12-DOF humanoid robot through zero-shot sim-to-real transfer. Quantitative comparisons demonstrate that selective AMP outperforms a uniform AMP policy across all five gaits, achieving faster convergence, lower tracking error, and higher success rates on stability-focused gaits without sacrificing the agility required for dynamic ones.

cs.RO↗

arXiv2026

GenericAgent: A Token-Efficient Self-Evolving LLM Agent via Contextual Information Density Maximization (V1.0)

Long-horizon large language model (LLM) agents are fundamentally limited by context. As interactions become longer, tool descriptions, retrieved memories, and raw environmental feedback accumulate and push out the information needed for decision-making. At the same time, useful experience gained from tasks is often lost across episodes. We argue that long-horizon performance is determined not by context length, but by how much decision-relevant information is maintained within a finite context budget. We present GenericAgent (GA), a general-purpose, self-evolving LLM agent system built around a single principle: context information density maximization. GA implements this through four closely connected components: a minimal atomic tool set that keeps the interface simple, a hierarchical on-demand memory that only shows a small high-level view by default, a self-evolution mechanism that turns verified past trajectories into reusable SOPs and executable code, and a context truncation and compression layer that maintains information density during long executions. Across task completion, tool use efficiency, memory effectiveness, self-evolution, and web browsing, GA consistently outperforms leading agent systems while using significantly fewer tokens and interactions, and it continues to evolve over time. Project: https://github.com/lsdefine/GenericAgent

cs.CL↗

arXiv2026

SEA-Eval: A Benchmark for Evaluating Self-Evolving Agents Beyond Episodic Assessment

Current LLM-based agents demonstrate strong performance in episodic task execution but remain constrained by static toolsets and episodic amnesia, failing to accumulate experience across task boundaries. This paper formalizes the Self-Evolving Agent (SEA) from the perspective of digital embodiment and continuous cross-task evolution, introduces the Evolutionary Flywheel as its minimal sufficient architecture, and presents SEA-Eval -- the first benchmark designed specifically for evaluating SEAs. Grounded in Flywheel theory, SEA-Eval establishes SR and T as primary metrics and, through sequential task stream design, is designed to quantify evolutionary gain, evolutionary stability, and implicit alignment convergence. Empirical evaluation reveals that, under comparable success rates, token consumption differs by up to 31.2 times between frameworks on individual tasks, with divergent evolutionary trajectories emerging under sequential analysis -- demonstrating that success rate alone creates a capability illusion and that the sequential convergence of $T$ is the key criterion for distinguishing genuine evolution from pseudo-evolution.

cs.AI↗

arXiv2026

White Dwarfs with Infrared Excess from LAMOST Data Release 11

Infrared (IR) excess observed around white dwarfs (WDs) is typically attributed to companions or debris disks. These systems are interesting because they offer a unique opportunity to study the late stages of stellar evolution and the interactions between WDs and surrounding material. The 11th data release (DR11) of the Large Sky Area Multi-Object Fiber Spectroscopic Telescope (LAMOST) - one of the largest spectroscopic surveys to date - has recently provided spectra for 3092 WDs, many of which have yet to be systematically investigated for IR excess. In this study, we cross-correlated the LAMOST DR11 WD catalog with optical and IR surveys, including the Sloan Digital Sky Survey (SDSS), Two Micron All-Sky Survey (2MASS), UKIRT Infrared Deep Sky Survey (UKIDSS), and Wide-field Infrared Survey Explorer (WISE). We performed spectral energy distribution fitting using the VOSA tool for 1818 WDs and identified 167 IR excess WD candidates. After excluding 23 sources with potential contamination within 6" and five additional sources identified through WISE ccf flag analysis, we identified 139 objects with candidate IR excess. These include 30 candidate WD + M dwarf binaries (18 new systems), 19 candidate WD + brown dwarf (BD) binaries (eight new systems), 66 candidate WD + dust disks (38 new systems), and 24 candidate either WD + BD or WD + dust disks (19 new systems). Given the limited spatial resolution of WISE, all candidate systems require follow-up IR observations for confirmation, such as high spatial resolution imaging or IR spectroscopy. This will help expand the parameter space of dust disks, allowing us to explore a broader range of possibilities.

astro-ph.SR↗

arXiv2026

Evaluation and Benchmarking Suite for Financial Large Language Models and Agents

Over the past three years, the financial services industry has witnessed Large Language Models (LLMs) and agents transitioning from the exploration stage to readiness and governance stages. Financial large language models (FinLLMs), such as open FinGPT and proprietary BloombergGPT , have great potential in financial applications, including retrieving real-time data, tutoring, analyzing sentiment of social media, analyzing SEC filings, and agentic trading. However, general-purpose LLMs and agents lack financial expertise and often struggle to handle complex financial reasoning. This paper presents an evaluation and benchmarking suite that covers the lifecycle of FinLLMs and FinAgents. This suite led by SecureFinAI Lab includes an evaluation pipeline and a governance framework collaborating with Linux Foundation and PyTorch Foundation, a FinLLM Leaderboard with HuggingFace, an AgentOps framework with Red Hat, and a documentation website with Rensselear Center of Open Source. Our collaborative development evolves through three stages: FinLLM Exploration (2023), FinLLM Readiness (2024), and FinAI Governance (2025). The proposed suite serves as an open platform that enables researchers and practitioners to perform both quantitative and qualitative analysis of different FinLLMs and FinAgents, fostering a more robust and reliable FinAI ecosystem.

cs.CE↗

arXiv2026

zkFinGPT: Zero-Knowledge Proofs for Financial Generative Pre-trained Transformers

Financial Generative Pre-trained Transformers (FinGPT) with multimodal capabilities are now being increasingly adopted in various financial applications. However, due to the intellectual property of model weights and the copyright of training corpus and benchmarking questions, verifying the legitimacy of GPT's model weights and the credibility of model outputs is a pressing challenge. In this paper, we introduce a novel zkFinGPT scheme that applies zero-knowledge proofs (ZKPs) to high-value financial use cases, enabling verification while protecting data privacy. We describe how zkFinGPT will be applied to three financial use cases. Our experiments on two existing packages reveal that zkFinGPT introduces substantial computational overhead that hinders its real-world adoption. E.g., for LLama3-8B model, it generates a commitment file of $7.97$MB using $531$ seconds, and takes $620$ seconds to prove and $2.36$ seconds to verify.

cs.CE↗

arXiv2025

Reasoning Models Ace the CFA Exams

Previous research has reported that large language models (LLMs) demonstrate poor performance on the Chartered Financial Analyst (CFA) exams. However, recent reasoning models have achieved strong results on graduate-level academic and professional examinations across various disciplines. In this paper, we evaluate state-of-the-art reasoning models on a set of mock CFA exams consisting of 980 questions across three Level I exams, two Level II exams, and three Level III exams. Using the same pass/fail criteria from prior studies, we find that most models clear all three levels. The models that pass, ordered by overall performance, are Gemini 3.0 Pro, Gemini 2.5 Pro, GPT-5, Grok 4, Claude Opus 4.1, and DeepSeek-V3.1. Specifically, Gemini 3.0 Pro achieves a record score of 97.6% on Level I. Performance is also strong on Level II, led by GPT-5 at 94.3%. On Level III, Gemini 2.5 Pro attains the highest score with 86.4% on multiple-choice questions while Gemini 3.0 Pro achieves 92.0% on constructed-response questions.

cs.AI↗

arXiv2025

FinAuditing: A Financial Taxonomy-Structured Multi-Document Benchmark for Evaluating LLMs

Going beyond simple text processing, financial auditing requires detecting semantic, structural, and numerical inconsistencies across large-scale disclosures. As financial reports are filed in XBRL, a structured XML format governed by accounting standards, auditing becomes a structured information extraction and reasoning problem involving concept alignment, taxonomy-defined relations, and cross-document consistency. Although large language models (LLMs) show promise on isolated financial tasks, their capability in professional-grade auditing remains unclear. We introduce FinAuditing, a taxonomy-aligned, structure-aware benchmark built from real XBRL filings. It contains 1,102 annotated instances averaging over 33k tokens and defines three tasks: Financial Semantic Matching (FinSM), Financial Relationship Extraction (FinRE), and Financial Mathematical Reasoning (FinMR). Evaluations of 13 state-of-the-art LLMs reveal substantial gaps in concept retrieval, taxonomy-aware relation modeling, and consistent cross-document reasoning. These findings highlight the need for realistic, structure-aware benchmarks. We release the evaluation code at https://github.com/The-FinAI/FinAuditing and the dataset at https://huggingface.co/collections/TheFinAI/finauditing. The task currently serves as the official benchmark of an ongoing public evaluation contest at https://open-finance-lab.github.io/SecureFinAI_Contest_2026/.

cs.CL↗

arXiv2025

The African Languages Lab: A Collaborative Approach to Advancing Low-Resource African NLP

Despite representing nearly one-third of the world's languages, African languages remain critically underserved by modern NLP technologies, with 88\% classified as severely underrepresented or completely ignored in computational linguistics. We present the African Languages Lab (All Lab), a comprehensive research initiative that addresses this technological gap through systematic data collection, model development, and capacity building. Our contributions include: (1) a quality-controlled data collection pipeline, yielding the largest validated African multi-modal speech and text dataset spanning 40 languages with 19 billion tokens of monolingual text and 12,628 hours of aligned speech data; (2) extensive experimental validation demonstrating that our dataset, combined with fine-tuning, achieves substantial improvements over baseline models, averaging +23.69 ChrF++, +0.33 COMET, and +15.34 BLEU points across 31 evaluated languages; and (3) a structured research program that has successfully mentored fifteen early-career researchers, establishing sustainable local capacity. Our comparative evaluation against Google Translate reveals competitive performance in several languages while identifying areas that require continued development.

cs.CL↗

arXiv2025

Dex1B: Learning with 1B Demonstrations for Dexterous Manipulation

Generating large-scale demonstrations for dexterous hand manipulation remains challenging, and several approaches have been proposed in recent years to address this. Among them, generative models have emerged as a promising paradigm, enabling the efficient creation of diverse and physically plausible demonstrations. In this paper, we introduce Dex1B, a large-scale, diverse, and high-quality demonstration dataset produced with generative models. The dataset contains one billion demonstrations for two fundamental tasks: grasping and articulation. To construct it, we propose a generative model that integrates geometric constraints to improve feasibility and applies additional conditions to enhance diversity. We validate the model on both established and newly introduced simulation benchmarks, where it significantly outperforms prior state-of-the-art methods. Furthermore, we demonstrate its effectiveness and robustness through real-world robot experiments. Our project page is at https://jianglongye.com/dex1b

cs.RO↗

arXiv2025

MultiFinBen: Benchmarking Large Language Models for Multilingual and Multimodal Financial Application

Real-world financial analysis involves information across multiple languages and modalities, from reports and news to scanned filings and meeting recordings. Yet most existing evaluations of LLMs in finance remain text-only, monolingual, and largely saturated by current models. To bridge these gaps, we present MultiFinBen, the first expert-annotated multilingual (five languages) and multimodal (text, vision, audio) benchmark for evaluating LLMs in realistic financial contexts. MultiFinBen introduces two new task families: multilingual financial reasoning, which tests cross-lingual evidence integration from filings and news, and financial OCR, which extracts structured text from scanned documents containing tables and charts. Rather than aggregating all available datasets, we apply a structured, difficulty-aware selection based on advanced model performance, ensuring balanced challenge and removing redundant tasks. Evaluating 21 leading LLMs shows that even frontier multimodal models like GPT-4o achieve only 46.01% overall, stronger on vision and audio but dropping sharply in multilingual settings. These findings expose persistent limitations in multilingual, multimodal, and expert-level financial reasoning. All datasets, evaluation scripts, and leaderboards are publicly released.

cs.CL↗

arXiv2025

FinTagging: Benchmarking LLMs for Extracting and Structuring Financial Information

Accurate interpretation of numerical data in financial reports is critical for markets and regulators. Although XBRL (eXtensible Business Reporting Language) provides a standard for tagging financial figures, mapping thousands of facts to over 10k US GAAP concepts remains costly and error prone. Existing benchmarks oversimplify this task as flat, single step classification over small subsets of concepts, ignoring the hierarchical semantics of the taxonomy and the structured nature of financial documents. Consequently, these benchmarks fail to evaluate Large Language Models (LLMs) under realistic reporting conditions. To bridge this gap, we introduce FinTagging, the first comprehensive benchmark for structure aware and full scope XBRL tagging. We decompose the complex tagging process into two subtasks: (1) FinNI (Financial Numeric Identification), which extracts entities and types from heterogeneous contexts including text and tables; and (2) FinCL (Financial Concept Linking), which maps extracted entities to the full US GAAP taxonomy. This two stage formulation enables a fair assessment of LLMs' capabilities in numerical reasoning and taxonomy alignment. Evaluating diverse LLMs in zero shot settings reveals that while models generalize well in extraction, they struggle significantly with fine grained concept linking, highlighting critical limitations in domain specific structure aware reasoning.

cs.CL↗

arXiv2025

FinRL Contests: Benchmarking Data-driven Financial Reinforcement Learning Agents

Financial reinforcement learning (FinRL) is now a practical paradigm for financial engineering. However, applying RL strategies to real-world trading tasks remains a challenge for individuals, as it is error-prone and engineering-heavy. The non-stationarity of financial data, low signal-to-noise ratios, and various market frictions require deep accumulations. Although numerous FinRL methods have been developed for tasks such as stock/crypto trading and portfolio management, the lack of standardized task definitions, real-time high-quality datasets, close-to-real market environments, and robust baselines has hindered consistent reproduction in both open-source community and FinTech industry. To bridge this gap, we organized a series of FinRL Contests from 2023 to 2025, covering a diverse range of financial tasks such as stock trading, order execution, crypto trading, and the use of large language model (LLM)-engineered signals. These contests attracted 200+ participants from 100+ institutions over 20+ countries. To encourage participations, we provided starter kits featuring GPU-optimized parallel market environments, ensemble learning, and comprehensive instructions. In this paper, we summarize these benchmarking efforts, detailing task formulations, data curation pipelines, environment implementations, evaluation protocols, participant performance, and organizational insights. It guides our follow-up FinRL contests, and also provides a reference for FinAI contests alike.

cs.CE↗

arXiv2025

FLAG-Trader: Fusion LLM-Agent with Gradient-based Reinforcement Learning for Financial Trading

Large language models (LLMs) fine-tuned on multimodal financial data have demonstrated impressive reasoning capabilities in various financial tasks. However, they often struggle with multi-step, goal-oriented scenarios in interactive financial markets, such as trading, where complex agentic approaches are required to improve decision-making. To address this, we propose \textsc{FLAG-Trader}, a unified architecture integrating linguistic processing (via LLMs) with gradient-driven reinforcement learning (RL) policy optimization, in which a partially fine-tuned LLM acts as the policy network, leveraging pre-trained knowledge while adapting to the financial domain through parameter-efficient fine-tuning. Through policy gradient optimization driven by trading rewards, our framework not only enhances LLM performance in trading but also improves results on other financial-domain tasks. We present extensive empirical evidence to validate these enhancements.

cs.AI↗

arXiv2025

Open FinLLM Leaderboard: Towards Financial AI Readiness

Financial large language models (FinLLMs) with multimodal capabilities are envisioned to revolutionize applications across business, finance, accounting, and auditing. However, real-world adoption requires robust benchmarks of FinLLMs' and FinAgents' performance. Maintaining an open leaderboard is crucial for encouraging innovative adoption and improving model effectiveness. In collaboration with Linux Foundation and Hugging Face, we create an open FinLLM leaderboard, which serves as an open platform for assessing and comparing AI models' performance on a wide spectrum of financial tasks. By demoncratizing access to advances of financial knowledge and intelligence, a chatbot or agent may enhance the analytical capabilities of the general public to a professional level within a few months of usage. This open leaderboard welcomes contributions from academia, open-source community, industry, and stakeholders. In particular, we encourage contributions of new datasets, tasks, and models for continual update. Through fostering a collaborative and open ecosystem, we seek to promote financial AI readiness.

cs.CE↗

arXiv2025

Revisiting Ensemble Methods for Stock Trading and Crypto Trading Tasks at ACM ICAIF FinRL Contest 2023-2024

Reinforcement learning has demonstrated great potential for performing financial tasks. However, it faces two major challenges: policy instability and sampling bottlenecks. In this paper, we revisit ensemble methods with massively parallel simulations on graphics processing units (GPUs), significantly enhancing the computational efficiency and robustness of trained models in volatile financial markets. Our approach leverages the parallel processing capability of GPUs to significantly improve the sampling speed for training ensemble models. The ensemble models combine the strengths of component agents to improve the robustness of financial decision-making strategies. We conduct experiments in both stock and cryptocurrency trading tasks to evaluate the effectiveness of our approach. Massively parallel simulation on a single GPU improves the sampling speed by up to $1,746\times$ using $2,048$ parallel environments compared to a single environment. The ensemble models have high cumulative returns and outperform some individual agents, reducing maximum drawdown by up to $4.17\%$ and improving the Sharpe ratio by up to $0.21$. This paper describes trading tasks at ACM ICAIF FinRL Contests in 2023 and 2024.

cs.CE↗

arXiv2024

A Report on Financial Regulations Challenge at COLING 2025

Financial large language models (FinLLMs) have been applied to various tasks in business, finance, accounting, and auditing. Complex financial regulations and standards are critical to financial services, which LLMs must comply with. However, FinLLMs' performance in understanding and interpreting financial regulations has rarely been studied. Therefore, we organize the Regulations Challenge, a shared task at COLING 2025. It encourages the academic community to explore the strengths and limitations of popular LLMs. We create 9 novel tasks and corresponding question sets. In this paper, we provide an overview of these tasks and summarize participants' approaches and results. We aim to raise awareness of FinLLMs' professional capability in financial regulations.

cs.CE↗