arXiv subjects

Yufeng Han

Publications and source records attributed to Yufeng Han.

7 recordsLinked to original sources

arXiv2026

Simulation-Supervised Foundation Models for Retention Time Prediction in High-Performance Liquid Chromatography beyond Experimental Data Coverage

Accurate prediction of high-performance liquid chromatography (HPLC) retention times (RTs) across diverse molecules and chromatographic methods remains challenging because experimental training data cover only a limited region of chemical and method spaces. Here, we develop FUSE-RT (Foundation model Unifying Simulation and Experimental supervision for Retention Time), a multitask foundation model that integrates RT data from 179 chromatographic methods and adapts to unseen molecules and methods using limited target-domain data. To extend transferability beyond experimental coverage, we introduce simulation-to-real (Sim2Real) transfer learning, in which molecular representations learned from large-scale computational data are transferred to experimental RT prediction. Specifically, we use PolyOmics, comprising 39 properties for approximately 21,400 molecules generated by molecular dynamics and density-functional theory calculations, as auxiliary supervision. We evaluate generalization under molecular, method, and joint molecular--method distribution shifts. Simulation-derived supervision substantially improves transfer beyond the experimental molecular domain, particularly under pronounced coverage gaps and few-shot adaptation. Moreover, RT-prediction error decreases systematically with increasing simulation-data size, following a significant power-law relationship. These results establish Sim2Real transfer as a scalable strategy for extending RT prediction beyond the finite coverage of experimental chromatographic data.

physics.chem-ph↗

arXiv2026

Jiuge-Tuiqiao: An Interpretable Human-AI System for Classical Chinese Poetry Refinement

Classical Chinese poetry composition has long valued Tuiqiao, the iterative refinement of words, imagery, and prosody. However, many current AI poetry systems follow a one-shot generation paradigm, which reduces users to prompt providers and weakens their creative agency. We present Jiuge-Tuiqiao, an interactive human-AI collaborative system for classical Chinese poetry composition. The system is designed around a triadic model: user-driven control, ancient-guided evidence, and AI-assisted generation. Users can lock characters or lines, receive real-time prosody feedback, and obtain interpretable refinement suggestions grounded in high-frequency collocations, PPL-ranked classical lines, and structured knowledge extracted from classical encyclopedias. This design turns AI from an autonomous generator into a background assistant that supports the user's own process of poetic refinement. Preliminary experiments and user feedback suggest that Jiuge-Tuiqiao improves controllability, interpretability, and user engagement in classical poetry composition.

cs.AI↗

arXiv2025

MEIC-DT: Memory-Efficient Incremental Clustering for Long-Text Coreference Resolution with Dual-Threshold Constraints

In the era of large language models (LLMs), supervised neural methods remain the state-of-the-art (SOTA) for Coreference Resolution. Yet, their full potential is underexplored, particularly in incremental clustering, which faces the critical challenge of balancing efficiency with performance for long texts. To address the limitation, we propose \textbf{MEIC-DT}, a novel dual-threshold, memory-efficient incremental clustering approach based on a lightweight Transformer. MEIC-DT features a dual-threshold constraint mechanism designed to precisely control the Transformer's input scale within a predefined memory budget. This mechanism incorporates a Statistics-Aware Eviction Strategy (\textbf{SAES}), which utilizes distinct statistical profiles from the training and inference phases for intelligent cache management. Furthermore, we introduce an Internal Regularization Policy (\textbf{IRP}) that strategically condenses clusters by selecting the most representative mentions, thereby preserving semantic integrity. Extensive experiments on common benchmarks demonstrate that MEIC-DT achieves highly competitive coreference performance under stringent memory constraints.

cs.IR↗

arXiv2025

ImCoref-CeS: An Improved Lightweight Pipeline for Coreference Resolution with LLM-based Checker-Splitter Refinement

Coreference Resolution (CR) is a critical task in Natural Language Processing (NLP). Current research faces a key dilemma: whether to further explore the potential of supervised neural methods based on small language models, whose detect-then-cluster pipeline still delivers top performance, or embrace the powerful capabilities of Large Language Models (LLMs). However, effectively combining their strengths remains underexplored. To this end, we propose \textbf{ImCoref-CeS}, a novel framework that integrates an enhanced supervised model with LLM-based reasoning. First, we present an improved CR method (\textbf{ImCoref}) to push the performance boundaries of the supervised neural method by introducing a lightweight bridging module to enhance long-text encoding capability, devising a biaffine scorer to comprehensively capture positional information, and invoking a hybrid mention regularization to improve training efficiency. Importantly, we employ an LLM acting as a multi-role Checker-Splitter agent to validate candidate mentions (filtering out invalid ones) and coreference results (splitting erroneous clusters) predicted by ImCoref. Extensive experiments demonstrate the effectiveness of ImCoref-CeS, which achieves superior performance compared to existing state-of-the-art (SOTA) methods.

cs.CL↗

arXiv2025

MiniCPM4: Ultra-Efficient LLMs on End Devices

This paper introduces MiniCPM4, a highly efficient large language model (LLM) designed explicitly for end-side devices. We achieve this efficiency through systematic innovation in four key dimensions: model architecture, training data, training algorithms, and inference systems. Specifically, in terms of model architecture, we propose InfLLM v2, a trainable sparse attention mechanism that accelerates both prefilling and decoding phases for long-context processing. Regarding training data, we propose UltraClean, an efficient and accurate pre-training data filtering and generation strategy, and UltraChat v2, a comprehensive supervised fine-tuning dataset. These datasets enable satisfactory model performance to be achieved using just 8 trillion training tokens. Regarding training algorithms, we propose ModelTunnel v2 for efficient pre-training strategy search, and improve existing post-training methods by introducing chunk-wise rollout for load-balanced reinforcement learning and data-efficient tenary LLM, BitCPM. Regarding inference systems, we propose CPM.cu that integrates sparse attention, model quantization, and speculative sampling to achieve efficient prefilling and decoding. To meet diverse on-device requirements, MiniCPM4 is available in two versions, with 0.5B and 8B parameters, respectively. Furthermore, we construct a hybrid reasoning model, MiniCPM4.1, which can be used in both deep reasoning mode and non-reasoning mode. Evaluation results demonstrate that MiniCPM4 and MiniCPM4.1 outperform similar-sized open-source models across benchmarks, with the 8B variants showing significant speed improvements on long sequence understanding and generation.

cs.CL↗

arXiv2018

Electromagnetic control of the collective radiations with three-photon blockade

We study properties of collective radiations of coherently driven two three-level ladder-type atoms trapped in a single-mode cavity. Using the electromagnetically induced transparency technique, we show that the three-photon blockade effect can be observed and the properties of collective radiations are strongly dependent on the phase between two atoms. In the case of in-phase radiations, the frequency range to realize the three-photon blockade can be broadened as the control field increases. However, in the regime of the three-photon blockade, the property of collective radiations changes from hyperradiance to subradiance. In the case of out-of-phase radiations, hyperradiance accompanied with the three-photon blockade can be observed. The results presented in this paper show that our scheme is an attractive candidate to generate antibunched photon pairs and control the properties of collective radiations.

quant-ph↗

arXiv2018

Electromagnetic control of nonclassicality in cavity QED system

We present a study of the electromagnetic control of nonclassicality of the outgoing light field in a single atom cavity QED system. By exploring the energy eigenvalues and eigenstates, we show that the eigenstates are similar to the Jaynes-Cummings ladder of eigenstates and can be optically controlled by an external control field. Tuning the control field frequency to the one photon resonance, we show the superbunching behavior in the outgoing light field can be observed at the frequency of one photon resonance. We also show that there exists a magic control field intensity at which two photon blockade phenomenon can be significantly improved. In particular, it is possible to adjust the nonclassicality of the outgoing field from quantum to classical by varying the control field intensity. The work presented here provides an optical method to control statistical features of the outgoing field and can be useful for the nonclassical light generation, quantum gate operation and exotic quantum state generation.

quant-ph↗