Search arXiv⌕ Search

arXiv subjects

Yin Chen

Publications and source records attributed to Yin Chen.

At least 19 recordsLinked to original sources
arXiv2026

EnvPilot: Systematic Design and Evaluation of an Experience-Augmented Agent for Software Environment Setup

Environment Setup is a critical yet complex task in software engineering that relies heavily on expert knowledge. Existing automated environment setup methods lack the ability to accumulate experience from past execution trajectories and to evolve over time. As a result, their performance is limited because they often perform redundant exploration, ignore useful past solutions, and fail to generalize across diverse software ecosystems. We present the systematic design and empirical validation of EnvPilot, an experience-augmented agent that operationalizes trajectory-derived experience reuse for software environment setup. EnvPilot maintains an expandable Trajectory-Derived Memory (TDM), initialized with 667 high-quality experiences. It systematically transforms implicit knowledge from historical execution trajectories into structured experience and retrieves the most relevant guidance during task execution through the Context-aware Retrieval mechanism. This enables EnvPilot to combine multiple validated setup strategies, providing more precise and detailed guidance than methods that rely solely on static project files or web retrieval. To evaluate EnvPilot, we construct AES-Bench, a multilingual benchmark of 112 real-world GitHub instances across 9 programming languages. Experiments show that EnvPilot achieves a new state-of-the-art (SOTA) with a 75.00% Pass@1 success rate while reducing reasoning costs. Our empirical study shows that both the structured experience representation and the Context-aware Retrieval mechanism are essential.

cs.SE↗
arXiv2026

Strain-driven orbital-selective reconstruction and bicollinear-to-stripe evolution in FeTe

FeTe, as a representative parent material among iron-based superconductors, provides an ideal platform for exploring the interplay among orbital-selective correlations, magnetism, and unconventional superconductivity. However, a unified picture of the correlated electronic structure and magnetism of FeTe under strain remains to be fully clarified. Here, combining density functional theory plus dynamical mean-field theory and Heisenberg model analysis, we uncover an orbital-selective reconstruction of the correlated electronic structure and reveal a strain-driven trajectory from bicollinear to stripe antiferromagnetism (AFM) via an intermediate competing staggered $n$-mer AFM regime in FeTe. Moderate strain gives rise to a regime where more coherent quasiparticles coexist with suppressed local moments. Further strain drives FeTe into an incoherent correlated regime with robust local moments and Fe-$3d_{z^2}$-dominated low-energy states. These results establish a strain-driven trajectory across distinct magnetic and correlated electronic states in FeTe.

cond-mat.supr-con↗
arXiv2026

Janus: Disaggregating Attention and Experts for Scalable MoE Inference

Serving large Mixture-of-Experts (MoE) models is challenging because of their large memory footprints, heterogeneous resource demands, and highly dynamic inference workloads. Most existing MoE inference systems deploy the entire model as a monolithic unit, forcing attention and MoE layers to share the same resource configuration despite their different scaling behaviors and resource bottlenecks. Such coarse-grained provisioning leads to resource inefficiency and suboptimal performance. We present JANUS, a scalable and resource-efficient MoE inference system built around three key principles. First, JANUS disaggregates attention and MoE layers onto separate GPU worker pools, enabling independent resource provisioning for the two layer types, and uses an adaptive two-phase communication mechanism for low-latency data exchange. Second, because MoE-layer execution is often memory-bound and highly sensitive to activated-expert imbalance, JANUS introduces a lightweight, microsecond-scale activation scheduler that balances per-layer activated experts across MoE instances to reduce inference latency. Third, JANUS employs a fine-grained, SLO-aware resource scaling scheme that jointly selects attention resources, MoE resources, and expert placement to minimize GPU cost under token-level SLOs. Evaluation shows that JANUS improves per-GPU throughput by up to 4.7x over state-of-the-art MoE inference baselines while satisfying token-level latency SLOs.

cs.DC↗
arXiv2026

Bidirectional Learning of Facial Action Units and Expressions via Structured Semantic Mapping across Heterogeneous Datasets

Facial action unit (AU) detection and facial expression (FE) recognition can be jointly viewed as affective facial behavior tasks, representing fine-grained muscular activations and coarse-grained holistic affective states, respectively. Despite their inherent semantic correlation, existing studies predominantly focus on knowledge transfer from AUs to FEs, while bidirectional learning remains insufficiently explored. In practice, this challenge is further compounded by heterogeneous data conditions, where AU and FE datasets differ in annotation paradigms (frame-level vs.\ clip-level), label granularity, and data availability and diversity, hindering effective joint learning. To address these issues, we propose a Structured Semantic Mapping (SSM) framework for bidirectional AU--FE learning under different data domains and heterogeneous supervision. SSM consists of three key components: (1) a shared visual backbone that learns unified facial representations from dynamic AU and FE videos; (2) semantic mediation via a Textual Semantic Prototype (TSP) module, which constructs structured semantic prototypes from fixed textual descriptions with learnable context prompts for supervision and cross-task alignment in a shared semantic space; and (3) a Dynamic Prior Mapping (DPM) module that incorporates FACS-derived prior knowledge and learns data-adaptive bidirectional association matrices in the textual semantic space for explicit knowledge transfer. Extensive experiments on popular AU detection and FE recognition benchmarks show that SSM consistently outperforms its single-task and multi-task baselines and achieves competitive performance against task-specific methods. The FE-to-AU results further show that holistic expression semantics provides useful supervision for fine-grained AU learning across heterogeneous datasets.

cs.CV↗
arXiv2026

Respect Your Zero-Shot Uncertainty: Conservative Calibration for Test-Time-Adapted Vision-Language Models

Test-time adaptation (TTA) can improve the recognition accuracy of vision-language models under distribution shift, but often degrades calibration, making predictive confidence unreliable for downstream decision-making. Many existing label-free calibration approaches are either coupled to prompt optimization or rely on logit-range statistics that provide only a coarse characterization of the predictive distribution. We show that TTA can increase confidence and reduce entropy even when the top-1 prediction and its correctness remain unchanged, a failure mode we term prediction-preserving sharpening. Across diverse TTA methods and benchmarks, larger entropy reductions relative to paired zero-shot predictions are associated with greater increases in Expected Calibration Error (ECE). On entropy-reduced samples, confidence gains also tend to exceed accuracy gains. Based on these findings, we propose Zero-Shot-Anchored Entropy Calibration (ZAEC), a label-free post-hoc method that uses zero-shot entropy as a sample-specific uncertainty reference. ZAEC selectively restores the zero-shot entropy of sharpened predictions through minimal temperature scaling while leaving all other predictions unchanged. It requires no labeled calibration data or learned parameters and preserves class rankings and classification accuracy. Across five TTA methods and 15 datasets, ZAEC achieves the lowest post-hoc macro-average ECE on ViT-B/16, with consistent gains on RN50.

cs.CV↗
arXiv2026

$\texttt{iNORG}$: An open-source quantum impurity solver package based on the natural orbitals renormalization group

In the context of dynamical mean-field theory (DMFT) calculations for strongly correlated electron systems, quantum impurity solvers play a central computational role in treating correlated lattice models and realistic materials. Consequently, developing efficient and robust quantum impurity solvers remains a key challenge. In this paper, we present an open-source quantum impurity solver package based on the natural orbitals renormalization group (NORG) method, dubbed $\texttt{iNORG}$. This software delivers high accuracy with reduced computational cost by optimizing the bath representation using natural orbitals and incorporating advanced features such as efficient Hilbert space selection and efficient algorithms for computing Green's functions. We first introduce the basic principle of the NORG method and then discuss the implementation details. The software framework, major features, and installation procedure for $\texttt{iNORG}$ are explained as well. Finally, several simple examples are presented to demonstrate the usage of $\texttt{iNORG}$.

cond-mat.str-el↗
arXiv2026

Arachne: Orchestrating Cascades for Efficient Text-to-Video Model Training

The rising demand for AI-generated videos is fueled by advances in large-scale Text-to-Video (T2V) models, trained on extensive datasets of video clips spanning diverse resolutions and durations. To address this data heterogeneity, current training methods often use a bucketing strategy that groups samples into discrete buckets for efficiency. However, this approach struggles to scale with compute and data volumes under static parallelism schemes, such as data and sequence parallelism, leading to significant workload imbalances and hardware under-utilization. In this paper, we present Arachne, a novel training framework for efficient T2V model training at scale. Arachne decomposes the training process into fine-grained computational units, called \textit{cascades}, orchestrating their distributed execution and synchronization across the cluster through coordinated spatial and temporal optimization. Our comprehensive evaluation demonstrates that Arachne reduces iteration time by up to 65\% over leading frameworks, exhibiting a positive scaling trend where its performance advantages amplify as training scale grows.

cs.DC↗
arXiv2026

An invariant-theoretic approach to three weight enumerators of self-dual quantum codes

This article is a continuation of our recent work (Yin Chen and Runxuan Zhang, Shape enumerators of self-dual NRT codes over finite fields. SIAM J. Discrete Math. 38 (2024), no. 4, 2841-2854) in the setting of quantum error-correcting codes. We use algebraic invariant theory to study three weight enumerators of formally self-dual quantum codes over arbitrary finite fields. We derive a quantum analogue of Gleason's theorem, demonstrating that the weight enumerator of a formally self-dual quantum code can be expressed algebraically by two polynomials. We also show that the double weight enumerator of a formally self-dual quantum code can be expressed algebraically by five polynomials. We explicitly compute the complete weight enumerators of some special self-dual quantum codes. Our approach illustrates the potential of employing algebraic invariant theory to compute weight enumerators of self-dual quantum codes.

cs.IT↗
arXiv2026

Self-Evolving Agents as Dynamic Graph Transformation: A Survey and New Perspective

Large language model (LLM)-based agents are increasingly becoming self-evolving systems that persist across interactions, maintain memories, use tools, acquire skills, refine workflows, and coordinate with other agents. These capabilities make agent states structural and dynamic: entities, relations, attributes, dependencies, and execution structures change with new evidence, feedback, and environmental conditions. Existing graph-agent surveys typically treat graphs as support structures for agent functions rather than as evolving substrates, while self-evolving-agent surveys focus on agent-level mechanisms and rarely discuss graph topology evolution. Thus, the coupling between evolving agent state and dynamic graph topology remains underexplored. This survey connects these two research lines by framing \textit{agent evolution as dynamic graph transformation}. We model agent state as a dynamic graph, where memories, tools, skills, workflows, and inter-agent relations are represented as typed nodes, edges, and subgraphs updated through schema-constrained rewrites. Based on this formulation, we organize existing dynamic-graph-based methods for self-evolving agents into four taxonomies: node/feature evolution, edge/topology evolution, subgraph activation, and cross-component co-evolution. Building on this taxonomy, we propose dynamic graph learning as reusable infrastructure for self-evolving agents and map nine dynamic-graph-learning subfields to agent-evolution capabilities, discussing their adaptations and possible failure modes. Finally, we discuss five types of graph-aware evaluation and governance protocols from a dynamic-graph perspective, which complement end-task evaluation. The goal is to provide a compact structural lens for designing and governing self-evolving agents.

cs.AI↗
arXiv2026

Swift-SVD: Theoretical Optimality Meets Practical Efficiency in Low-Rank LLM Compression

The deployment of Large Language Models is constrained by the memory and bandwidth demands of static weights and dynamic Key-Value cache. SVD-based compression provides a hardware-friendly solution to reduce these costs. However, existing methods suffer from two key limitations: some are suboptimal in reconstruction error, while others are theoretically optimal but practically inefficient. In this paper, we propose Swift-SVD, an activation-aware, closed-form compression framework that simultaneously guarantees theoretical optimum, practical efficiency and numerical stability. Swift-SVD incrementally aggregates covariance of output activations given a batch of inputs and performs a single eigenvalue decomposition after aggregation, enabling training-free, fast, and optimal layer-wise low-rank approximation. We employ effective rank to analyze local layer-wise compressibility and design a dynamic rank allocation strategy that jointly accounts for local reconstruction loss and end-to-end layer importance. Extensive experiments across six LLMs and eight datasets demonstrate that Swift-SVD outperforms state-of-the-art baselines, achieving optimal compression accuracy while delivering 3-70X speedups in end-to-end compression time. Our code is available at https://github.com/hiahei/Swift-SVD.

cs.CL↗
arXiv2026

Mosaic: Towards Efficient Training of Multimodal Models with Spatial Resource Multiplexing

With the wide adoption of Multimodal Models (MMs) in real-world scenarios, it is significant to efficiently train emerging MMs that exhibit increasingly complex module architectures. For MM deployment, existing works allocate a GPU to only one MM module in a temporal-multiplexing manner; this compromises training efficiency because a single module often fails to achieve high GPU utilization. To improve GPU utilization and enable efficient MM training, we propose deploying MMs in a temporal-spatial multiplexing manner, allowing multiple MM modules to colocate on a GPU with well-controlled resource quotas. In this paper, we propose Apollo, an efficient MM training system that applies temporal-spatial multiplexing. We first develop a flexible and lightweight execution engine that supports MM training with arbitrary resource quotas, and then build a comprehensive and accurate performance model to estimate module execution time under different allocation plans. With the performance model, we further adopt effective heuristics to derive high-quality MM deployment plans efficiently. Testbed experiments confirm that Apollo effectively improves the training efficiency of popular MMs, with a training speedup of up to 1.31x.

cs.DC↗
arXiv2026

CALVO: Improve Serving Efficiency for LLM Inferences with Intense Network Demands

Distributed prefix caching has become a core technique for efficient LLM serving. However, for long-context requests with high cache hit ratios, retrieving reusable KVCache blocks from remote servers has emerged as a new performance bottleneck. Such network-intensive LLM inference is expected to become increasingly common as agentic AI workloads continue to grow. However, existing LLM inference engines remain largely compute-centric: they treat KVCache loading as a subordinate phase to GPU execution and often fail to account for its delay explicitly during scheduling. We present CALVO, an LLM serving engine that treats KVCache loading as a first-class concern. CALVO decouples KVCache loading and GPU computation into independently managed, asynchronously progressing stages, enabling better utilization of network, PCIe, and computation resources. In addition, CALVO incorporates KVCache loading delay as an explicit component of per-request service cost, leading to more accurate scheduling decisions. Experiments on a real testbed with diverse long-context workloads show that CALVO substantially improves the efficiency of network-intensive LLM inference, achieving up to 61.67% higher SLO attainment than the baseline.

cs.DC↗
arXiv2026

Geometry and classifications of some $ω$-Lie algebras

Using group actions and orbit-stabilizer methods, we study the geometry of isomorphism classes of finite-dimensional $ω$-Lie algebras over a field $\mathbb{K}$ of characteristic $\neq 2$ and establish a one-to-one correspondence between the set of isomorphism classes and the orbit space of a stabilizer of $ω$. We also apply techniques from computational ideal theory to explore the geometric structure of the affine variety of all 3-dimensional $ω$-Lie algebras over $\mathbb{K}$, showing that this variety is a 6-dimensional irreducible affine variety and a complete intersection. As an application, we derive a complete classification of all 3-dimensional $ω$-Lie algebras over an algebraically closed field of characteristic $\neq 2$, up to $ω$-Lie algebra isomorphism.

math.RA↗
arXiv2026

Modular matrix invariants under some transpose actions

Consider the special linear group of degree $2$ over an arbitrary finite field, acting on the full space of $2 \times 2$-matrices by transpose. We explicitly construct a generating set for the corresponding modular matrix invariant ring, demonstrating that this ring is a hypersurface. Using a recent result on $a$-invariants of Cohen-Macaulay algebras, we determine the Hilbert series of this invariant ring, and our method avoids seeking the generating relation. Additionally, we prove that the modular matrix invariant ring of the group of upper triangular $2 \times 2$-matrices is also a hypersurface.

math.AC↗
arXiv2026

Vector invariants for two-dimensional orthogonal groups over finite fields

Let $\mathbb{F}_{q}$ be a finite field of characteristic $2$ and $O_2^+(\mathbb{F}_{q})$ be the $2$-dimensional orthogonal group of plus type over $\mathbb{F}_{q}$. Consider the standard representation $V$ of $O_2^+(\mathbb{F}_{q})$ and the ring of vector invariants $\mathbb{F}_{q}[mV]^{O_2^+(\mathbb{F}_{q})}$ for any $m\in \mathbb{N}^{+}$. We prove a first main theorem for $(O_2^+(\mathbb{F}_{q}),V)$, i.e., we find a minimal generating set for $\mathbb{F}_{q}[mV]^{O_2^+(\mathbb{F}_{q})}$. As a consequence, we derive the Noether number $β_{mV}(O_2^+(\mathbb{F}_{q}))=\max\{q-1,m\}$. We construct a free basis for $\mathbb{F}_{q}[2V]^{O_2^+(\mathbb{F}_{q})}$ over a suitably chosen homogeneous system of parameters. We also obtain a generating set of the Hilbert ideal for $\mathbb{F}_{q}[mV]^{O_2^+(\mathbb{F}_{q})}$ which shows that the Hilbert ideal can be generated by invariants of degree $\leqslant q-1=\frac{|O_2^+(\mathbb{F}_{q})|}{2}$, positively confirming a conjecure of Derksen and Kemper for this particular case.

math.AC↗
arXiv2026

Rota-Baxter operators on $ω$-Lie algebras

This article explores Rota-Baxter operators on finite-dimensional $ω$-Lie algebras over a field of characteristic not 2. We provide several methods for constructing left-symmetric algebras, $ω$-Lie algebras, and Hom-Lie algebras via compatible Rota-Baxter operators on a given $ω$-Lie algebra. We also study the geometric structures of compatible Rota-Baxter operators of weight $0$ and isometric Rota-Baxter operators of weight $1$ over the field of complex numbers. In particular, we prove that the affine variety of all isometric Rota-Baxter operators of weight $1$ on any finite-dimensional non-Lie complex simple $ω$-Lie algebra is $1$-dimensional. Furthermore, we show that for every $4$-dimensional non-Lie complex $ω$-Lie algebra, there always exists a nilpotent compatible Rota-Baxter operator of weight $0$ such that the induced Hom-Lie algebra is nonabelian but solvable.

math.RA↗
arXiv2026

Hidden in-plane long-range order in an amorphized crystal

Solid materials are commonly classified as crystalline or amorphous based on the presence or absence of long-range order.Metal-organic frameworks (MOFs), like other solids,also display markedly different properties and functions in these two phases. Here, we identify a previously unrecognized structural state that retains long-range in-plane translational order while losing order along the stacking direction. Hypothesized since 1941 but not experimentally verified, this intermediate phase emerges in a crystalline MOFs via controlled thermal desolvation, which selectively disrupts the intrinsically weak interlayer interactions while preserving macroscopic structural coherence. Although the resulting material appears amorphous under conventional characterization, systematic synchrotron PXRD, total X-ray scattering, and low-dose high resolution TEM reveal clear in-plane periodicity. This material spontaneously delaminates in water into uniform, high-quality two-dimensional crystalline nanosheets, forming stable colloidal suspensions and exhibiting superlubricity comparable to graphene - but at less than 0.1% of the production cost. Our discovery finds a missing link within the long-standing crystalline-amorphous dichotomy, while providing an inherently scalable route to high-quality 2D crystals, and offering a conceptual and practical advance in phase engineering.

cond-mat.mtrl-sci↗
arXiv2026

KV-CoRE: Benchmarking Data-Dependent Low-Rank Compressibility of KV-Caches in LLMs

Large language models rely on kv-caches to avoid redundant computation during autoregressive decoding, but as context length grows, reading and writing the cache can quickly saturate GPU memory bandwidth. Recent work has explored KV-cache compression, yet most approaches neglect the data-dependent nature of kv-caches and their variation across layers. We introduce KV-CoRE KV-cache Compressibility by Rank Evaluation), an SVD-based method for quantifying the data-dependent low-rank compressibility of kv-caches. KV-CoRE computes the optimal low-rank approximation under the Frobenius norm and, being gradient-free and incremental, enables efficient dataset-level, layer-wise evaluation. Using this method, we analyze multiple models and datasets spanning five English domains and sixteen languages, uncovering systematic patterns that link compressibility to model architecture, training data, and language coverage. As part of this analysis, we employ the Normalized Effective Rank as a metric of compressibility and show that it correlates strongly with performance degradation under compression. Our study establishes a principled evaluation framework and the first large-scale benchmark of kv-cache compressibility in LLMs, offering insights for dynamic, data-aware compression and data-centric model development.

cs.CL↗