Ningxin Su

Yuxuan Zhao, Sijia Chen, Ningxin Su

LLMs have shown strong performance across diverse financial tasks, yet portfolio management (PM), a critical financial decision-making task, remains poorly benchmarked. Existing benchmarks exhibit two main gaps: they ignore cross-asset correlation structures, thereby failing to distinguish genuinely diversified portfolios from concentrated ones, and fail to evaluate the complete PM decision pipeline in real-world scenarios. We introduce PortBench, a benchmark spanning six heterogeneous asset classes over ten years. PortBench consists of two complementary layers: a static QA dataset of 6,269 correlation-based questions across seven task templates, and a dynamic five-stage allocation pipeline that mirrors the full PM decision cycle. To evaluate these layers, we introduce two dedicated metrics: a dual-layer correlation score that measures whether proposed portfolios exploit inter-class hedging and avoid intra-class concentration, and CEPS, a metric that quantifies how reasoning errors compound across pipeline stages. We further assess strategy robustness and investor alignment under three historical stress regimes and risk profiles. Evaluating ten frontier LLMs, we find that despite strong performance on static financial QA, 90\% of model-profile combinations fail to outperform a basic equal-weight allocation, and models that satisfy every procedural constraint still suffer catastrophic drawdowns under stress. Our source code is available at \href{https://github.com/AgenticFinLab/portbench}{this https URL}.

Yuxuan Zhao, Sijia Chen, Ningxin Su

Multi-agent systems (MAS) have emerged as a prominent paradigm for leveraging large language models (LLMs) to tackle complex tasks. However, the mechanisms governing the effectiveness of MAS built upon publicly available LLMs, specifically the underlying rationales for their success or failure, remain largely unexplored. In this paper, we revisit MAS through the perspective of uncertainty, considering both intra- and inter-agent dynamics by investigating entropy transitions during problem-solving across various topologies and six benchmark tasks. By analyzing 245 features spanning token-, trajectory-, and round-level entropy, we counterintuitively find that a single agent outperforms MAS in approximately 43.3% of cases, and that uncertainty dynamics are largely determined during the first round of interaction. Furthermore, we provide three key observations: 1) Certainty Preference: reducing uncertainty at any stage for any agent is critical for guaranteeing correct solutions; 2) Base Uncertainty: base models with lower entropy during problem-solving directly benefit MAS performance; and 3) Task Awareness: entropy dynamics of MAS play varying roles across different tasks. Building on these insights, we introduce a simple yet effective algorithm, the Entropy Judger, to select solutions from MAS's pass@k results, leading to consistent accuracy improvements across all MAS configurations and tasks. Our source code is available at https://github.com/AgenticFinLab/multiagent-entropy.

Yi Liu, TingFeng Hui, Wei Zhang et al.

Scalable AI agents training relies on interactive environments that faithfully simulate the consequences of agent actions. Manually crafted environments are expensive to build, brittle to extend, and fundamentally limited in diversity. A promising direction is to replace manually crafted environments with LLM-simulated counterparts. However, this paradigm hinges on an unexamined core assumption: LLMs can accurately simulate environmental feedback. In practice, LLM-simulated environments suffer from hallucinations, logical inconsistencies, and silent state drift failures that corrupt agent reward signals and compound the construction costs that the paradigm was designed to eliminate. To address this gap, we propose EnvSimBench with four contributions: 1) We provide the first formal definition and operationalization of Environment Simulation Ability (EnvSim Ability) as a quantifiable research objective. 2) We construct EnvSimBench, a rigorous benchmark covering 400 samples across 167 diverse environments, equipped with verifiable labels and fine-grained difficulty stratification along three axes. 3) Systematic evaluations reveal that all state-of-the-art language models suffer from a universal state change cliff: they achieve near-perfect accuracy on tasks when the environment state remains invariant, yet fail catastrophically when multiple states need simultaneous updates. This finding exposes EnvSim Ability as a critical yet largely unaddressed capability gap. 4) We design a constraint-driven simulation pipeline that substantially reduces hallucination, boosts environment synthesis yield by 6.8%, and cuts costs by over 90%. Overall, EnvSimBench serves as both a diagnostic framework and a practical optimization path for reliable LLM-based environment simulation, establishing a foundation for scalable agent training. Code and data are available at https://github.com/cookieApril/EnvSimBench

Sijia Chen, Ningxin Su, Baochun Li

In the context of personalized federated learning, existing approaches train a global model to extract transferable representations, based on which any client could train personalized models with a limited number of data samples. Self-supervised learning is considered a promising direction as the global model it produces is generic and facilitates personalization for all clients fairly. However, when data is heterogeneous across clients, the global model trained using SSL is unable to learn high-quality personalized models. In this paper, we show that when the global model is trained with SSL without modifications, its produced representations have fuzzy class boundaries. As a result, personalized learning within each client produces models with low accuracy. In order to improve SSL towards better accuracy without sacrificing its advantage in fairness, we propose Calibre, a new personalized federated learning framework designed to calibrate SSL representations by maintaining a suitable balance between more generic and more client-specific representations. Calibre is designed based on theoretically-sound properties, and introduces (1) a client-specific prototype loss as an auxiliary training objective; and (2) an aggregation algorithm guided by such prototypes across clients. Our experimental results in an extensive array of non-i.i.d.~settings show that Calibre achieves state-of-the-art performance in terms of both mean accuracy and fairness across clients. Code repo: https://github.com/TL-System/plato/tree/main/examples/ssl/calibre.

Qifu Wen, Xi Zeng, Zihan Zhou et al.

Early stopping monitors global validation loss and halts all parameter updates simultaneously, which is computationally costly for large transformers due to the extended time required for validation inference. We propose \textit{GradES}, a novel gradient-based early stopping approach that operates within transformer components (attention projections and Feed-Forward layer matrices). We found that different components converge at varying rates during fine-tuning for both language and vision-language models. \textit{GradES} tracks the magnitude of gradient changes in backpropagation for these matrices during training. When a projection matrix's magnitude of gradient changes fall below a convergence threshold $τ$, we exclude that projection matrix from further updates individually, eliminating costly validation passes while allowing slow converging matrices to continue learning. \textit{GradES} speeds up training time by 1.57--7.22$\times$ while simultaneously enhancing generalization through early prevention of overfitting, resulting in 1.2\% higher average accuracy in language tasks and 3.88\% on multimodal benchmarks.