arXiv subjects

Sen Hu

Publications and source records attributed to Sen Hu.

At least 37 records · Page 2Linked to original sources

arXiv2025

RepoMaster: Autonomous Exploration and Understanding of GitHub Repositories for Complex Task Solving

The ultimate goal of code agents is to solve complex tasks autonomously. Although large language models (LLMs) have made substantial progress in code generation, real-world tasks typically demand full-fledged code repositories rather than simple scripts. Building such repositories from scratch remains a major challenge. Fortunately, GitHub hosts a vast, evolving collection of open-source repositories, which developers frequently reuse as modular components for complex tasks. Yet, existing frameworks like OpenHands and SWE-Agent still struggle to effectively leverage these valuable resources. Relying solely on README files provides insufficient guidance, and deeper exploration reveals two core obstacles: overwhelming information and tangled dependencies of repositories, both constrained by the limited context windows of current LLMs. To tackle these issues, we propose RepoMaster, an autonomous agent framework designed to explore and reuse GitHub repositories for solving complex tasks. For efficient understanding, RepoMaster constructs function-call graphs, module-dependency graphs, and hierarchical code trees to identify essential components, providing only identified core elements to the LLMs rather than the entire repository. During autonomous execution, it progressively explores related components using our exploration tools and prunes information to optimize context usage. Evaluated on the adjusted MLE-bench, RepoMaster achieves a 110% relative boost in valid submissions over the strongest baseline OpenHands. On our newly released GitTaskBench, RepoMaster lifts the task-pass rate from 40.7% to 62.9% while reducing token usage by 95%. Our code and demonstration materials are publicly available at https://github.com/QuantaAlpha/RepoMaster.

cs.SE↗

arXiv2025

GAP: Graph-Assisted Prompts for Dialogue-based Medication Recommendation

Medication recommendations have become an important task in the healthcare domain, especially in measuring the accuracy and safety of medical dialogue systems (MDS). Different from the recommendation task based on electronic health records (EHRs), dialogue-based medication recommendations require research on the interaction details between patients and doctors, which is crucial but may not exist in EHRs. Recent advancements in large language models (LLM) have extended the medical dialogue domain. These LLMs can interpret patients' intent and provide medical suggestions including medication recommendations, but some challenges are still worth attention. During a multi-turn dialogue, LLMs may ignore the fine-grained medical information or connections across the dialogue turns, which is vital for providing accurate suggestions. Besides, LLMs may generate non-factual responses when there is a lack of domain-specific knowledge, which is more risky in the medical domain. To address these challenges, we propose a \textbf{G}raph-\textbf{A}ssisted \textbf{P}rompts (\textbf{GAP}) framework for dialogue-based medication recommendation. It extracts medical concepts and corresponding states from dialogue to construct an explicitly patient-centric graph, which can describe the neglected but important information. Further, combined with external medical knowledge graphs, GAP can generate abundant queries and prompts, thus retrieving information from multiple sources to reduce the non-factual responses. We evaluate GAP on a dialogue-based medication recommendation dataset and further explore its potential in a more difficult scenario, dynamically diagnostic interviewing. Extensive experiments demonstrate its competitive performance when compared with strong baselines.

cs.CL↗

arXiv2025

Pastiche Novel Generation Creating: Fan Fiction You Love in Your Favorite Author's Style

Great novels create immersive worlds with rich character arcs, well-structured plots, and nuanced writing styles. However, current novel generation methods often rely on brief, simplistic story outlines and generate details using plain, generic language. To bridge this gap, we introduce the task of Pastiche Novel Generation, which requires the generated novels to imitate the distinctive features of the original work, including understanding character profiles, predicting plausible plot developments, and writing concrete details using vivid, expressive language. To achieve this, we propose WriterAgent, a novel generation system designed to master the core aspects of literary pastiche. WriterAgent is trained through a curriculum learning paradigm, progressing from low-level stylistic mastery to high-level narrative coherence. Its key tasks include language style learning, character modeling, plot planning, and stylish writing, ensuring comprehensive narrative control. To support this, WriterAgent leverages the WriterLoRA framework, an extension of LoRA with hierarchical and cumulative task-specific modules, each specializing in a different narrative aspect. We evaluate WriterAgent on multilingual classics like Harry Potter and Dream of the Red Chamber, demonstrating its superiority over baselines in capturing the target author's settings, character dynamics, and writing style to produce coherent, faithful narratives.

cs.CL↗

arXiv2024

On the Hilbert Space of the Chern-Simons Matrix Model, Deformed Double Current Algebra Action, and the Conformal Limit

A Chern-Simons matrix model was proposed by Dorey, Tong, and Turner to describe non-Abelian fractional quantum Hall effect. In this paper we study the Hilbert space of the Chern-Simons matrix model from a geometric quantization point of view. We show that the Hilbert space of the Chern-Simons matrix model can be identified with the space of sections of a line bundle on the quiver variety associated to a framed Jordan quiver. We compute the character of the Hilbert space using localization technique. Using a natural isomorphism between vortex moduli space and a Beilinson-Drinfeld Schubert variety, we prove that the ground states wave functions are flat sections of a bundle of conformal blocks associated to a WZW model. In particular they solve a Knizhnik-Zamolodchikov equation. We show that there exists a natural action of the deformed double current algebra (DDCA) on the Hilbert space, moreover the action is irreducible. We define and study the conformal limit of the Chern-Simons matrix model. We show that the conformal limit of the Hilbert space is an irreducible integrable module of $\widehat{\mathfrak{gl}}(n)$ with level identified with the matrix model level. Moreover, we prove that $\widehat{\mathfrak{gl}}(n)$ generators can be obtained from scaling limits of matrix model operators, which settles a conjecture of Dorey-Tong-Turner. The key to the proof is the construction of a Yangian $Y(\mathfrak{gl}_n)$ action on the conformal limit of the Hilbert space, which we expect to be equivalent to the $Y(\mathfrak{gl}_n)$ action on the integrable $\widehat{\mathfrak{gl}}(n)$ modules constructed by Uglov. We also characterize eigenvectors and eigenvalues of the matrix model Hilbert space with respect to a maximal commutative subalgebra of Yangian.

math-ph↗

arXiv2024

Quantum Algebra of Chern-Simons Matrix Model and Large $N$ Limit

In this paper we study the algebra of quantum observables of the Chern-Simons matrix model which was originally proposed by Susskind and Polychronakos to describe electrons in fractional quantum Hall effects. We establish the commutation relations for its generators and study the large $N$ limit of its representation. We show that the large $N$ limit algebra is isomorphic to the uniform in $N$ algebra studied by Costello, which is isomorphic to the deformed double current algebra studied by Guay. Under appropriate scaling limit, we show that the large $N$ limit algebra degenerates to a Lie algebra which admits a surjective map to the affine Lie algebra of $\mathfrak{u}(p)$. This leads to a complete proof of the large $N$ emergence of the $\mathfrak{u}(p)$ current algebra as proposed by Dorey, Tong and Turner. This also suggests a rigorous derivation of edge excitation of a fractional quantum Hall droplet.

math.QA↗

arXiv2024

Are LLM-based Evaluators Confusing NLG Quality Criteria?

Some prior work has shown that LLMs perform well in NLG evaluation for different tasks. However, we discover that LLMs seem to confuse different evaluation criteria, which reduces their reliability. For further verification, we first consider avoiding issues of inconsistent conceptualization and vague expression in existing NLG quality criteria themselves. So we summarize a clear hierarchical classification system for 11 common aspects with corresponding different criteria from previous studies involved. Inspired by behavioral testing, we elaborately design 18 types of aspect-targeted perturbation attacks for fine-grained analysis of the evaluation behaviors of different LLMs. We also conduct human annotations beyond the guidance of the classification system to validate the impact of the perturbations. Our experimental results reveal confusion issues inherent in LLMs, as well as other noteworthy phenomena, and necessitate further research and improvements for LLM-based evaluation.

cs.CL↗

arXiv2024

From Beginner to Expert: Modeling Medical Knowledge into General LLMs

Recently, large language model (LLM) based artificial intelligence (AI) systems have demonstrated remarkable capabilities in natural language understanding and generation. However, these models face a significant challenge when it comes to sensitive applications, such as reasoning over medical knowledge and answering medical questions in a physician-like manner. Prior studies attempted to overcome this challenge by increasing the model size (>100B) to learn more general medical knowledge, while there is still room for improvement in LLMs with smaller-scale model sizes (<100B). In this work, we start from a pre-trained general LLM model (AntGLM-10B) and fine-tune it from a medical beginner towards a medical expert (called AntGLM-Med-10B), which leverages a 3-stage optimization procedure, i.e., general medical knowledge injection, medical domain instruction tuning, and specific medical task adaptation. Our contributions are threefold: (1) We specifically investigate how to adapt a pre-trained general LLM in medical domain, especially for a specific medical task. (2) We collect and construct large-scale medical datasets for each stage of the optimization process. These datasets encompass various data types and tasks, such as question-answering, medical reasoning, multi-choice questions, and medical conversations. (3) Specifically for multi-choice questions in the medical domain, we propose a novel Verification-of-Choice approach for prompting engineering, which significantly enhances the reasoning ability of LLMs. Remarkably, by combining the above approaches, our AntGLM-Med-10B model can outperform the most of LLMs on PubMedQA, including both general and medical LLMs, even when these LLMs have larger model size.

cs.CL↗

arXiv2023

S2M: Converting Single-Turn to Multi-Turn Datasets for Conversational Question Answering

Supplying data augmentation to conversational question answering (CQA) can effectively improve model performance. However, there is less improvement from single-turn datasets in CQA due to the distribution gap between single-turn and multi-turn datasets. On the other hand, while numerous single-turn datasets are available, we have not utilized them effectively. To solve this problem, we propose a novel method to convert single-turn datasets to multi-turn datasets. The proposed method consists of three parts, namely, a QA pair Generator, a QA pair Reassembler, and a question Rewriter. Given a sample consisting of context and single-turn QA pairs, the Generator obtains candidate QA pairs and a knowledge graph based on the context. The Reassembler utilizes the knowledge graph to get sequential QA pairs, and the Rewriter rewrites questions from a conversational perspective to obtain a multi-turn dataset S2M. Our experiments show that our method can synthesize effective training resources for CQA. Notably, S2M ranks 1st place on the QuAC leaderboard at the time of submission (Aug 24th, 2022).

cs.CL↗

arXiv2023

AdapterDistillation: Non-Destructive Task Composition with Knowledge Distillation

Leveraging knowledge from multiple tasks through introducing a small number of task specific parameters into each transformer layer, also known as adapters, receives much attention recently. However, adding an extra fusion layer to implement knowledge composition not only increases the inference time but also is non-scalable for some applications. To avoid these issues, we propose a two-stage knowledge distillation algorithm called AdapterDistillation. In the first stage, we extract task specific knowledge by using local data to train a student adapter. In the second stage, we distill the knowledge from the existing teacher adapters into the student adapter to help its inference. Extensive experiments on frequently asked question retrieval in task-oriented dialog systems validate the efficiency of AdapterDistillation. We show that AdapterDistillation outperforms existing algorithms in terms of accuracy, resource consumption and inference time.

cs.LG↗

arXiv2023

Feynman geometry

In this paper we introduce a notion of Feynman geometry on which quantum field theories could be properly defined. A strong Feynman geometry is a geometry when the vector space of $A_\infty$ structures is finite dimensional. A weak Feynman geometry is a geometry when the vector space of $A_\infty$ structures is infinite dimensional while the relevant operators are of trace-class. We construct families of Feynman geometries with "continuum" as their limit.

hep-th↗

arXiv2022

Ancient and recent collisions revealed by phosphate minerals in the Chelyabinsk meteorite

The collision history of asteroids is an important archive of inner Solar System evolution. Evidence for these collisions is brought to Earth by meteorites, which can preserve impact-reset radioisotope mineral ages. However, as meteorites often preserve numerous mineral ages, their interpretation is controversial. Here, we combine analysis of phosphate U-Pb ages and mineral microtextures to construct a collision history for the highly shocked Chelyabinsk meteorite. We show that phosphate U-Pb ages in the meteorite are independent of thermal history at macro-to-microscales, correlating instead with phosphate microtexture. Isotopic data from pristine phosphate domains is largely concordant, whereas fracture-damaged domains universally display discordance. Combining both populations best constrains upper (4,473 +/- 11 Ma) and lower intercept (-9 +/- 55 Ma, i.e., within error of the present day) U-Pb ages for Chelyabinsk phosphates. We conclude that all phosphate U-Pb ages were completely reset during an ancient high energy collision. Fracture-damaged phosphate domains experienced further Pb-loss during mild collisional heating in the geologically recent past, and must be targeted to properly constrain a lower intercept age. Targeting textural sub-populations of phosphate grains can significantly improve the calculation and interpretation of U-Pb ages, permitting more robust reconstruction of both ancient and recent asteroidal collision histories.

astro-ph.EP↗

arXiv2021

Mars: new insights and unresolved questions

Mars exploration motivates the search for extraterrestrial life, the development of space technologies, and the design of human missions and habitations. Here we seek new insights and pose unresolved questions relating to the natural history of Mars, habitability, robotic and human exploration, planetary protection, and the impacts on human society. Key observations and findings include:(1)high escape rates of early Mars' atmosphere, including loss of water, impact present-day habitability;(2)putative fossils on Mars will likely be ambiguous biomarkers for life;(3)microbial contamination resulting from human habitation is unavoidable;(4)based on Mars' current planetary protection category, robotic payload(s) should characterize the local martian environment for any life-forms prior to human habitation. Some of the outstanding questions are:(1)which interpretation of the hemispheric dichotomy of the planet is correct;(2)to what degree did deep-penetrating faults transport subsurface liquids to Mars' surface;(3)in what abundance are carbonates formed by atmospheric processes;(4)what properties of martian meteorites could be used to constrain their source locations;(5)the origin(s) of organic macromolecules;(6)was/is Mars inhabited;(7)how can missions designed to uncover microbial activity in the subsurface eliminate potential false positives caused by microbial contaminants from Earth;(8)how can we ensure that humans and microbes form a stable and benign biosphere;(9)should humans relate to putative extraterrestrial life from a biocentric viewpoint (preservation of all biology), or anthropocentric viewpoint of expanding habitation of space? Studies of Mars' evolution can shed light on the habitability of extrasolar planets. In addition, Mars exploration can drive future policy developments and confirm (or put into question) the feasibility and/or extent of human habitability of space.

astro-ph.EP↗

arXiv2021

Copula Averaging for Tail Dependence in Insurance Claims Data

Analysing dependent risks is an important task for insurance companies. A dependency is reflected in the fact that information about one random variable provides information about the likely distribution of values of another random variable. Insurance companies in particular must investigate such dependencies between different lines of business and the effects that an extreme loss event, such as an earthquake or hurricane, has across multiple lines of business simultaneously. Copulas provide a popular model-based approach to analysing the dependency between risks, and the coefficient of tail dependence is a measure of dependence for extreme losses. Besides commonly used empirical estimators for estimating the tail dependence coefficient, copula fitting can lead to estimation of such coefficients directly or can verify their existence. Generally, a range of copula models is available to fit a data set well, leading to multiple different tail dependence results; a method based on Bayesian model averaging is designed to obtain a unified estimate of tail dependence. In this article, this model-based coefficient estimation method is illustrated through a variety of copula fitting approaches and results are presented for several simulated data sets and also a real general insurance loss data set.

stat.AP↗

arXiv2020

Overview of the CCKS 2019 Knowledge Graph Evaluation Track: Entity, Relation, Event and QA

Knowledge graph models world knowledge as concepts, entities, and the relationships between them, which has been widely used in many real-world tasks. CCKS 2019 held an evaluation track with 6 tasks and attracted more than 1,600 teams. In this paper, we give an overview of the knowledge graph evaluation tract at CCKS 2019. By reviewing the task definition, successful methods, useful resources, good strategies and research challenges associated with each task in CCKS 2019, this paper can provide a helpful reference for developing knowledge graph applications and conducting future knowledge graph researches.

cs.AI↗

arXiv2019

Bivariate Gamma Mixture of Experts Models for Joint Insurance Claims Modeling

In general insurance, risks from different categories are often modeled independently and their sum is regarded as the total risk the insurer takes on in exchange for a premium. The dependence from multiple risks is generally neglected even when correlation could exist, for example a single car accident may result in claims from multiple risk categories. It is desirable to take the covariance of different categories into consideration in modeling in order to better predict future claims and hence allow greater accuracy in ratemaking. In this work multivariate severity models are investigated using mixture of experts models with bivariate gamma distributions, where the dependence structure is modeled directly using a GLM framework, and covariates can be placed in both gating and expert networks. Furthermore, parsimonious parameterisations are considered, which leads to a family of bivariate gamma mixture of experts models. It can be viewed as a model-based clustering approach that clusters policyholders into sub-groups with different dependencies, and the parameters of the mixture models are dependent on the covariates. Clustering is shown to be important in separating the data into sub-groupings where strong dependence is often present, even if the overall data set exhibits only weak dependence. In doing so, the correlation within different components features prominently in the model. It is shown that, by applying to both simulated data and a real-world Irish GI insurer data set, claim predictions can be improved.

stat.AP↗

arXiv2018

A graphical calculus for semi-groupal categories

Around the year 1988, Joyal and Street established a graphical calculus for monoidal categories, which provides a firm foundation for many explorations of graphical notations in mathematics and physics. For a deeper understanding of their work, we consider a similar graphical calculus for semi-groupal categories. We introduce two frameworks to formalize this graphical calculus, a topological one based on the notion of a processive plane graph and a combinatorial one based on the notion of a planarly ordered processive graph, which serves as a combinatorial counterpart of a deformation class of processive plane graphs. We demonstrate the equivalence of Joyal and Street's graphical calculus and the theory of upward planar drawings. We introduce the category of semi-tensor schemes, and give a construction of a free monoidal category on a semi-tensor scheme. We deduce the unit convention as a kind of quotient construction, and show an idea to generalize the unit convention. Finally, we clarify the relation of the unit convention and Joyal and Street's construction of a free monoidal category on a tensor scheme.

math.CO↗

arXiv2017

Motor Insurance Accidental Damage Claims Modeling with Factor Collapsing and Bayesian Model Averaging

Accidental damage is a typical component of motor insurance claim. Modeling of this nature generally involves analysis of past claim history and different characteristics of the insured objects and the policyholders. Generalized linear models (GLMs) have become the industry's standard approach for pricing and modeling risks of this nature. However, the GLM approach utilizes a single "best" model on which loss predictions are based, which ignores the uncertainty among the competing models and variable selection. An additional characteristic of motor insurance data sets is the presence of many categorical variables, within which the number of levels is high. In particular, not all levels of such variables may be statistically significant and rather some subsets of the levels may be merged to give a smaller overall number of levels for improved model parsimony and interpretability. A method is proposed for assessing the optimal manner in which to collapse a factor with many levels into one with a smaller number of levels, then Bayesian model averaging (BMA) is used to blend model predictions from all reasonable models to account for factor collapsing uncertainty. This method will be computationally intensive due to the number of factors being collapsed as well as the possibly large number of levels within factors. Hence a stochastic optimisation is proposed to quickly find the best collapsing cases across the model space.

stat.AP↗

arXiv2017

Entanglement Entropy of $AdS_5 \times S^5$ with massless flavors at non-zero temperature

We consider backreacted $AdS_5 \times S^5$ coupled with $N_f$ massless flavors introduced by D7-branes at non-zero temperature. The backreacted geometry is in the Veneziano limit. The temperature of this system is related to the event horizon at $r_h$. Dividing one of the spatial directions into a line segment with length $l$, we will calculate the entanglement entropy between the two subspaces. We study the behavior near the event horizon, and finally find that there exists phase transition phenomenon near the event horizon since the difference between the entanglement entropy of the connected minimal surface and the disconnected one changes sign.

hep-th↗