arXiv subjects

Yicheng Mao

Publications and source records attributed to Yicheng Mao.

9 recordsLinked to original sources

arXiv2026

OpenAI4S: Code as Action, Science as Sessions

AI co-scientists could accelerate computational research, but over a long-running study the workflow also has to stay inspectable, resumable and reproducible, which requires persistent computational state and provenance. Here we present OpenAI4S, an open-source scientific research agent built around the principle of \emph{Code as Action, Science as Sessions}. OpenAI4S combines a persistent computing runtime with research-session management: orchestration is handled through structured tool calls, while scientific actions are represented as complete code cells executed in persistent Python and R kernels. An append-only Action Ledger, per-cell execution records, versioned artifacts, environment records, and workspace checkpoints preserve how results were produced and support session recovery, branching, and extension. Configurable sandboxing, permission controls, and code and trajectory screening provide complementary safeguards. We evaluate OpenAI4S on 36 research scenarios spanning retrosynthesis, molecular dynamics, protein binder design, protein mutation, catalyst screening, and mineral spectroscopy, measuring scientific task accuracy, workflow completeness, and reproducibility of the resulting repositories. OpenAI4S achieves an overall score of 7.83, compared with 5.7--6.4 for a general-purpose coding harness evaluated with three frontier models, with the largest gains on long-horizon and computation-intensive workflows. These results suggest that integrating persistent execution with session-level provenance can improve the reliability of AI-assisted scientific workflows. Environment specification and full rerunnability remain weak for every evaluated system, ours included, so reproducibility is still an open problem for scientific agents. The system is available under the MIT license at \href{https://github.com/PKU-YuanGroup/OpenAI4S}{github.com/PKU-YuanGroup/OpenAI4S}.

cs.AI↗

arXiv2026

Integrating adaptive human behavior into epidemic models with large language models

Infectious disease transmission is shaped by patterns of human interaction, which adapt as epidemic conditions change. Capturing these context-dependent behaviors remains a fundamental challenge for epidemic models. Here, we recast this challenge by using large language models (LLMs) to represent adaptive human behavior within mechanistic epidemic models. We operationalize this idea through Generative Adaptive Behavioral Layer for Epidemics (GABLE), which adapts LLMs to infer behavioral responses to epidemic and policy conditions and translates them into age-structured contact matrices coupled to a mechanistic epidemic model. Applied to COVID-19 in France, GABLE reproduced responses in population mixing and age-specific contact structures that remained epidemiologically informative. In short-term forecasting, LLM-generated contact matrices outperformed mobility-driven matrices derived from real-world mobility data, with the largest gains at longer horizons. GABLE also extends beyond forecasting to prospective policy evaluation by projecting behavioral and epidemic responses to candidate interventions before implementation. When supplied with subsequently implemented policies, GABLE reproduced epidemic trajectories and generated distinct responses to alternative policy timing and composition. By leveraging LLMs as a flexible behavioral layer, GABLE provides a framework for coupling context-sensitive behavioral generation with epidemic dynamics.

physics.soc-ph↗

arXiv2026

Data Mixing as Mixture Experiment: Response Surface Methodology and Optimal Design for Large Language Model Pretraining

Data mixing is a central design problem in large language model pretraining: given a fixed token budget, practitioners must decide how much data to allocate to each domain. Recent proxy-based methods address this problem by training small models on candidate mixtures, fitting a response model, and using the response to select mixtures for larger-scale training. We show that this workflow has the structure of a classical mixture experiment. Under this view, data domains are mixture components, token shares are component proportions, proxy-training runs are experimental design points, and validation loss defines a response surface over the probability simplex. We develop this formulation using sparse second-order Scheffé response-surface models and construct model-robust $\mathcal{I}$-optimal designs for proxy data-mixing experiments. Using RegMix as an empirical case study, we demonstrate how the framework can both interpret observed mixture responses and design more efficient proxy experiments. The Scheffé analysis shows that domain value is strongly relational: several domains that are weak under additive effects become favourable through pairwise interactions, especially through combinations with web-derived text. The sparse Scheffé model preserves mixture rankings across model scales and remains competitive with a flexible machine-learning predictor while providing an explicit decomposition of additive and interaction effects. In a simulation study calibrated to observed proxy-training responses, model-robust $\mathcal{I}$-optimal designs recover the relevant mixture ordering after removing about 25\% of the original proxy runs. These results suggest that LLM data mixing should be treated not only as a prediction problem, but also as an experimental-design problem in which the proxy mixtures themselves can be chosen to improve statistical efficiency.

cs.AI↗

arXiv2026

LatentChem: From Textual CoT to Latent Thinking in Chemical Reasoning

Current chemical large language models (LLMs) predominantly rely on explicit Chain-of-Thought (CoT) to solve complex reasoning problems. However, forcing nonverbal tacit chemical logic into discrete natural language imposes a fundamental ``modality mismatch,'' creating an artificial bottleneck for reasoning. We introduce LatentChem, a reasoning interface that decouples chemical logic from linguistic generation, enabling the model to process information via continuous thought vectors and dynamic perception. Our investigation reveals a pivotal emergent behavior: spontaneous internalization, defined here as self-selected under outcome-only optimization. When optimized for task success, the model abandons verbose textual derivations in favor of implicit latent computation, suggesting that it identifies the continuous manifold as a more native substrate for chemical logic. This paradigm shift also proves to be a superior computational strategy: LatentChem achieves a 59.88\% non-tie win rate against the strong CoT baseline on the rigorous ChemCoTBench, while delivering a broad 10.84$\times$ average reduction in reasoning step overhead (5.96$\times$ wall-clock speedup) across all evaluated benchmarks. Our results provide empirical evidence that chemical reasoning is more naturally and effectively realized as continuous latent dynamics rather than discretized linguistic trajectories.

physics.chem-ph↗

arXiv2026

DCEDesignSA: A MATLAB-based Graphical User Interface for Discrete Choice Experiment Design Using Simulated Annealing

DCEDesignSA is a freely available MATLAB package for generating Bayesian D-optimal discrete choice experiment designs. It employs Simulated Annealing to efficiently search the design space and maximise the Bayesian D-optimality criterion under user-specified prior distributions. The toolbox features an interactive graphical user interface, enabling researchers without programming expertise to define experimental settings, generate optimal designs, and export survey-ready designs directly to Qualtrics. DCEDesignSA supports interaction terms in utility, no-choice alternatives, and presentation order effects.

stat.CO↗

arXiv2026

Neural Posterior Estimation for Spatial Individual-Level Epidemic Models

Spatial individual-level models (ILMs) provide a flexible framework for modelling infectious disease transmission across populations with known locations. Bayesian inference for these models relies on Markov chain Monte Carlo (MCMC), which requires repeated likelihood evaluation and, when parts of the epidemic trajectory are unobserved, data-augmented sampling over high-dimensional latent variables. This computational cost limits the applicability of MCMC to large populations and to settings requiring inference across multiple outbreaks. We propose using neural posterior estimation (NPE) for amortised Bayesian inference in spatial ILMs. NPE trains a conditional normalising flow on simulated data to approximate the posterior directly, bypassing likelihood evaluation at inference time. We compare two embedding architectures: a convolutional neural network (CNN) operating on the population-level incidence curve and a graph neural network (GNN) operating on individual-level infection and location data. In a simulation study under full observation, stochastic removals, and partial observation, both variants produce well-calibrated posteriors, with the GNN embedding yielding lower error and narrower credible intervals for the spatial transmission parameters. We apply the framework to a spatial SEIR model on 1,177 farm locations from the 2001 UK foot-and-mouth disease outbreak. GNN-NPE maintains calibrated coverage and is substantially faster than MCMC on a per-epidemic basis.

stat.CO↗

arXiv2026

Simulated Annealing for Model-Robust Partial Profile Choice Designs in Healthcare Preference Studies

Discrete Choice Experiments (DCEs) investigate participants' preferences by observing their choice behavior in hypothetical scenarios and are widely used in the domain of healthcare. To reduce participants' cognitive burden, especially when dealing with a large number of attributes, researchers often employ partial profile designs. In these designs, certain attributes within each choice set are kept constant. Current literature on partial profile designs mainly focuses on main-effects models rather than interaction-effect models, with certain partial profile designs even incapable of estimating interaction effects. To address this issue, this paper introduces an Simulated Annealing (SA) algorithm to construct partial profile designs based on an interaction-effects model. During the experimental design phase, the existence and magnitude of interaction effects are often unknown. Therefore, this paper proposes a model-robust experimental design strategy. Through extensive simulation experiments and a real-life case study, we demonstrate that our SA model-robust partial profile design performs relatively well regardless of the underlying model.

stat.ME↗

arXiv2025

Digital Gatekeepers: Exploring Large Language Model's Role in Immigration Decisions

With globalization and increasing immigrant populations, immigration departments face significant work-loads and the challenge of ensuring fairness in decision-making processes. Integrating artificial intelligence offers a promising solution to these challenges. This study investigates the potential of large language models (LLMs),such as GPT-3.5 and GPT-4, in supporting immigration decision-making. Utilizing a mixed-methods approach,this paper conducted discrete choice experiments and in-depth interviews to study LLM decision-making strategies and whether they are fair. Our findings demonstrate that LLMs can align their decision-making with human strategies, emphasizing utility maximization and procedural fairness. Meanwhile, this paper also reveals that while ChatGPT has safeguards to prevent unintentional discrimination, it still exhibits stereotypes and biases concerning nationality and shows preferences toward privileged group. This dual analysis highlights both the potential and limitations of LLMs in automating and enhancing immigration decisions.

cs.CL↗

arXiv2025

Constructing Bayesian Optimal Designs for Discrete Choice Experiments by Simulated Annealing

Discrete choice experiments (DCEs) investigate the attributes that influence individuals' choices when selecting among various options. To enhance the quality of the estimated choice models, researchers opt for Bayesian optimal designs that utilize existing information about the attributes' preferences. Given the nonlinear nature of choice models, the construction of an appropriate design requires efficient algorithms. Among these, the coordinate-exchange (CE) algorithm is commonly employed for constructing designs based on the MNL model. However, as a hill-climbing method, the CE algorithm tends to quickly converge to local optima, potentially limiting the quality of the resulting designs. We propose the use of a simulated annealing (SA) algorithm to construct Bayesian optimal designs. This algorithm accepts both superior and inferior solutions, avoiding premature convergence and allowing a more thorough exploration of potential solutions. Consequently, it ultimately obtains higher-quality choice designs compared to the CE algorithm. Our work represents the first application of an SA algorithm in constructing Bayesian optimal designs for DCEs. Through extensive computational experiments, we demonstrate that the SA designs generally outperform the CE designs in terms of statistical efficiency, especially when the prior preference information is highly uncertain.

stat.ME↗