arXiv subjects

Li Yin

Publications and source records attributed to Li Yin.

At least 19 recordsLinked to original sources

arXiv2026

@skills: Attention is all you have

There are 56,804 public agent skills today, and teams write many more privately. The dominant delivery model is installation: once installed, a skill's description remains in the system prompt, competing for fewer than 100 reliable trigger slots. This leaves the long tail with no practical path to use and forces teams' own playbooks to compete for the same scarce space. We observe that installation bundles three separable functions: content, persistence, and automatic triggering. Only the last requires prompt residency. We therefore propose @skills, an open protocol that separates them. A path addresses any skill, subtree, or collection, and reading a skill is sufficient to use it, so nothing is installed or made resident. The operation vendors a copy at the same path into a project's Git-tracked tree for adaptation and ownership. The operation adds one .gitignore-style line, the only element that costs prompt residency. A directory is a menu, making bundles ordinary directories rather than all-or-nothing units. The protocol requires no manifest, lockfile, or registration, and SKILL.md remains unchanged. @skills is additive, ships as an installable package, and turns any agent that can read files and run commands into a client through a single instruction file. Its open specification is at https://github.com/SylphAI-Inc/atskills and it is implemented in the AdaL CLI at https://adalagent.ai . Because paths address skills well but cannot find them, the protocol is paired with a free hub at https://atskills.one for corpus-wide search and ranking, repository-free hosting, private and team collections, and one-screen authoring. The hub is optional: gh: and local paths resolve without it, and indexed GitHub skills retain their gh: identities. Install less, use more.

cs.AI↗

arXiv2026

Compile, Then Page: Executable SOP Programs and a Capability-Gated Runtime for Procedural LLM Agents

Enterprise agents must follow long-horizon, conditional, safety-critical standard operating procedures (SOPs). We compile machine-readable SOP constraints into executable pseudo-code and run them with a program-guided (PG) stack machine that pages the active frame while an LLM performs semantic execution. A three-arm SOPBench study across six models separates representation from runtime: compiled text never significantly hurts and gains up to 16.0 points where official prose underperforms. Runtime guidance is capability-gated. Two strong models independently show positive seven-domain PG contrasts (58:19 and 75:31 discordant pairs), whereas weak models are harmed. A full-program cursor ablation (active frame first, complete program retained) recovers much of the strong-model refusal gain; selective visibility adds a smaller improvement. Paired probe and audit measurements track this divide to spontaneous state discipline rather than reconstruction ability. On Bank the three primary arms rise from 70.4 to 86.4 to 92.8, with 100% refusal correctness. Practical guidance: compile first; enable active-frame paging only after a model-level discipline check.

cs.AI↗

arXiv2026

Mesh-Aware Epipolar Matching for Multi-View Multi-Person 3D Pose Estimation in Basketball

Multi-view multi-person 3D pose estimation in team sports scenarios remains challenging due to player occlusions, appearance similarity caused by team uniforms, and the scarcity of annotated multi-view data, all of which limit the effectiveness and generalization capability of learning-based methods. In contrast, the performance of training-free approaches is inherently constrained by the accuracy of 2D keypoint detection and the robustness of cross-view association. To address these challenges, we propose Mesh-Aware Epipolar Matching (MAEM), a training-free framework for multi-view multi-person 3D pose estimation. Our method employs a monocular 3D human mesh recovery model as the frontend and introduces a two-stage epipolar matching strategy based on the recovered mesh outputs. Specifically, the proposed framework combines disjoint-set-union-based clustering with per-joint triangulation to achieve robust cross-view association and accurate 3D pose reconstruction. Experiments on two public multi-view basketball datasets demonstrate that MAEM consistently outperforms existing training-free association baselines while achieving competitive RGB-only performance in both indoor and outdoor basketball scenarios. MAEM achieves MPJPE/PA-MPJPE scores of 59.8/40.7 mm on SportCenter EPFL and 74.0/51.8 mm on Human-M3 Basketball, highlighting the effectiveness of dense mesh geometry for cross-view association without requiring target-domain training or fine-tuning.

cs.CV↗

arXiv2026

The Last Harness You'll Ever Build

AI agents are increasingly deployed on complex, domain-specific workflows -- navigating enterprise web applications that require dozens of clicks and form fills, orchestrating multi-step research pipelines that span search, extraction, and synthesis, automating code review across unfamiliar repositories, and handling customer escalations that demand nuanced domain knowledge. \textbf{Each new task domain requires painstaking, expert-driven harness engineering}: designing the prompts, tools, orchestration logic, and evaluation criteria that make a foundation model effective. We present a two-level framework that automates this process. At the first level, the \textbf{Harness Evolution Loop} optimizes a worker agent's harness $\mathcal{H}$ for a single task: a Worker Agent $W_{\mathcal{H}}$ executes the task, an Evaluator Agent $V$ adversarially diagnoses failures and scores performance, and an Evolution Agent $E$ modifies the harness based on the full history of prior attempts. At the second level, the \textbf{Meta-Evolution Loop} optimizes the evolution blueprint $\Lambda = (W_{\mathcal{H}}, \mathcal{H}^{(0)}, V, E)$ itself across diverse tasks, \textbf{learning a blueprint $\Lambda^{(\text{best})}$ that enables rapid harness convergence on any new task -- so that adapting an agent to a novel domain requires no human harness engineering at all.} We formalize the correspondence to meta-learning and present both algorithms. The framework \textbf{shifts manual harness engineering into automated harness engineering}, and takes one step further -- \textbf{automating the design of the automation itself}.

cs.AI↗

arXiv2025

MIRAD - A comprehensive real-world robust anomaly detection dataset for Mass Individualization

Social manufacturing leverages community collaboration and scattered resources to realize mass individualization in modern industry. However, this paradigm shift also introduces substantial challenges in quality control, particularly in defect detection. The main difficulties stem from three aspects. First, products often have highly customized configurations. Second, production typically involves fragmented, small-batch orders. Third, imaging environments vary considerably across distributed sites. To overcome the scarcity of real-world datasets and tailored algorithms, we introduce the Mass Individualization Robust Anomaly Detection (MIRAD) dataset. As the first benchmark explicitly designed for anomaly detection in social manufacturing, MIRAD captures three critical dimensions of this domain: (1) diverse individualized products with large intra-class variation, (2) data collected from six geographically dispersed manufacturing nodes, and (3) substantial imaging heterogeneity, including variations in lighting, background, and motion conditions. We then conduct extensive evaluations of state-of-the-art (SOTA) anomaly detection methods on MIRAD, covering one-class, multi-class, and zero-shot approaches. Results show a significant performance drop across all models compared with conventional benchmarks, highlighting the unresolved complexities of defect detection in real-world individualized production. By bridging industrial requirements and academic research, MIRAD provides a realistic foundation for developing robust quality control solutions essential for Industry 5.0. The dataset is publicly available at https://github.com/wu33learn/MIRAD.

cs.CV↗

arXiv2025

LAD-VF: LLM-Automatic Differentiation Enables Fine-Tuning-Free Robot Planning from Formal Methods Feedback

Large language models (LLMs) can translate natural language instructions into executable action plans for robotics, autonomous driving, and other domains. Yet, deploying LLM-driven planning in the physical world demands strict adherence to safety and regulatory constraints, which current models often violate due to hallucination or weak alignment. Traditional data-driven alignment methods, such as Direct Preference Optimization (DPO), require costly human labeling, while recent formal-feedback approaches still depend on resource-intensive fine-tuning. In this paper, we propose LAD-VF, a fine-tuning-free framework that leverages formal verification feedback for automated prompt engineering. By introducing a formal-verification-informed text loss integrated with LLM-AutoDiff, LAD-VF iteratively refines prompts rather than model parameters. This yields three key benefits: (i) scalable adaptation without fine-tuning; (ii) compatibility with modular LLM architectures; and (iii) interpretable refinement via auditable prompts. Experiments in robot navigation and manipulation tasks demonstrate that LAD-VF substantially enhances specification compliance, improving success rates from 60% to over 90%. Our method thus presents a scalable and interpretable pathway toward trustworthy, formally-verified LLM-driven control systems.

cs.RO↗

arXiv2025

Stabilization of Ferroelectric Hafnia and Zirconia through Y2O3 doping

We investigate the possible stabilization of ferroelectricity in bulk Y2O3-doped hafnia and zirconia. We use density functional theory (DFT) with large random supercells of hafnia and zirconia and study the relative phase stability of the centrosymmetric cubic and monoclinic phases compared with the polar orthorhombic phase. We find that Y2O3-doping stabilizes the polar ferroelectric phase over the monoclinic baddeleyite phase in both hafnia and zirconia.

cond-mat.mtrl-sci↗

arXiv2025

Scaling Textual Gradients via Sampling-Based Momentum

LLM-based prompt optimization, which uses LLM-provided ``textual gradients'' (feedback) to refine prompts, has emerged as an effective method for automatic prompt engineering. However, its scalability and stability are unclear when using more data in training. We systematically investigate the potential and challenges of scaling training data in textual gradient descent. We show that naively scaling training examples is infeasible due to both explicit context-length limits and an implicit context wall, where long-context degradation yields diminishing returns. Inspired by prior wisdom in stochastic gradient descent, we propose Textual Stochastic Gradient Descent with Momentum (TSGD-M), which reweights updates through momentum sampling, using bootstrapped minibatch validation accuracy as importance weights over historical prompts. To stabilize TSGD and enable effective scaling within a limited context window, TSGD-M carries prior prompts information by \textit{dynamically} exploring the past top performing prompts without expanding input context length. TSGD-M integrates seamlessly into existing prompt optimization frameworks, including TextGrad, DSPy-COPRO, and AdalFlow, and achieves consistent gains across 6 benchmarks.

cs.CL↗

arXiv2025

TrackID3x3: A Dataset and Algorithm for Multi-Player Tracking with Identification and Pose Estimation in 3x3 Basketball Full-court Videos

Multi-object tracking, player identification, and pose estimation are fundamental components of sports analytics, essential for analyzing player movements, performance, and tactical strategies. However, existing datasets and methodologies primarily target mainstream team sports such as soccer and conventional 5-on-5 basketball, often overlooking scenarios involving fixed-camera setups commonly used at amateur levels, less mainstream sports, or datasets that explicitly incorporate pose annotations. In this paper, we propose the TrackID3x3 dataset, the first publicly available comprehensive dataset specifically designed for multi-player tracking, player identification, and pose estimation in 3x3 basketball scenarios. The dataset comprises three distinct subsets (Indoor fixed-camera, Outdoor fixed-camera, and Drone camera footage), capturing diverse full-court camera perspectives and environments. We also introduce the Track-ID task, a simplified variant of the game state reconstruction task that excludes field detection and focuses exclusively on fixed-camera scenarios. To evaluate performance, we propose a baseline algorithm called Track-ID algorithm, tailored to assess tracking and identification quality. Furthermore, our benchmark experiments, utilizing recent multi-object tracking algorithms (e.g., BoT-SORT-ReID) and top-down pose estimation methods (HRNet, RTMPose, and SwinPose), demonstrate robust results and highlight remaining challenges. Our dataset and evaluation benchmarks provide a solid foundation for advancing automated analytics in 3x3 basketball. Dataset and code will be available at https://github.com/open-starlab/TrackID3x3.

cs.CV↗

arXiv2025

On some properties of special functions involving $k$-gamma and $k$-digamma functions

Based on $k$-gamma and $k$-digamma functions, we show four series expansions to the Furdui-type integral related to Riemann zeta function and hypergeometric function, and also present some new identities, series expansions and inequalities on the Hadamard $k$-gamma function and the Nielsen $k$-beta function. Finally, we also pose an open problem.

math.CA↗

arXiv2025

LLM-AutoDiff: Auto-Differentiate Any LLM Workflow

Large Language Models (LLMs) have reshaped natural language processing, powering applications from multi-hop retrieval and question answering to autonomous agent workflows. Yet, prompt engineering -- the task of crafting textual inputs to effectively direct LLMs -- remains difficult and labor-intensive, particularly for complex pipelines that combine multiple LLM calls with functional operations like retrieval and data formatting. We introduce LLM-AutoDiff: a novel framework for Automatic Prompt Engineering (APE) that extends textual gradient-based methods (such as Text-Grad) to multi-component, potentially cyclic LLM architectures. Implemented within the AdalFlow library, LLM-AutoDiff treats each textual input as a trainable parameter and uses a frozen backward engine LLM to generate feedback-akin to textual gradients -- that guide iterative prompt updates. Unlike prior single-node approaches, LLM-AutoDiff inherently accommodates functional nodes, preserves time-sequential behavior in repeated calls (e.g., multi-hop loops), and combats the "lost-in-the-middle" problem by isolating distinct sub-prompts (instructions, formats, or few-shot examples). It further boosts training efficiency by focusing on error-prone samples through selective gradient computation. Across diverse tasks, including single-step classification, multi-hop retrieval-based QA, and agent-driven pipelines, LLM-AutoDiff consistently outperforms existing textual gradient baselines in both accuracy and training cost. By unifying prompt optimization through a graph-centric lens, LLM-AutoDiff offers a powerful new paradigm for scaling and automating LLM workflows - mirroring the transformative role that automatic differentiation libraries have long played in neural network research.

cs.CL↗

arXiv2025

Anomalous temperature-dependent magnetization in the nearly collinear antiferromagnet Y$_2$Co$_3$

Y$_2$Co$_3$ is a newly discovered antiferromagnetic (AFM) compound with distorted kagome layers. Previous investigations via bulk magnetization measurements suggested a complex noncollinear magnetic behavior, with magnetic moments primarily anti-aligned along the $b$ axis and some canting towards the $ac$ plane. In this study, we report the magnetic structure of Y$_2$Co$_3$ to be an A-type AFM structure with ferromagnetic (FM) interactions within the distorted kagome plane and an interplane antiferromagnetic interaction, as determined by single-crystal neutron diffraction. The magnetic moments align along the $b$ axis, with minimal canting towards the $c$ axis, at odds with the previous interpretation of bulk magnetization measurements. The magnetic moments on the two distinct Co sites are [0, -0.68(3), 0] $\mu_B$ and [0, 1.25(4), 0.07(1)] $\mu_B$. We attribute the previously reported "noncollinear" behavior to the considerable temperature dependence of itinerant AFM exchange interactions, induced by thermal contraction along the $b$ axis. Additionally, our examination of lattice constants through pressure studies reveals compensating effects on FM and AFM interactions, resulting in negligible pressure dependence of $T_\textrm{N}$.

cond-mat.str-el↗

arXiv2024

Enhanced Multi-Object Tracking Using Pose-based Virtual Markers in 3x3 Basketball

Multi-object tracking (MOT) is crucial for various multi-agent analyses such as evaluating team sports tactics and player movements and performance. While pedestrian tracking has advanced with Tracking-by-Detection MOT, team sports like basketball pose unique challenges. These challenges include players' unpredictable movements, frequent close interactions, and visual similarities that complicate pose labeling and lead to significant occlusions, frequent ID switches, and high manual annotation costs. To address these challenges, we propose a novel pose-based virtual marker (VM) MOT method for team sports, named Sports-vmTracking. This method builds on the vmTracking approach developed for multi-animal tracking with active learning. First, we constructed a 3x3 basketball pose dataset for VMs and applied active learning to enhance model performance in generating VMs. Then, we overlaid the VMs on video to identify players, extract their poses with unique IDs, and convert these into bounding boxes for comparison with automated MOT methods. Using our 3x3 basketball dataset, we demonstrated that our VM configuration has been highly effective, and reduced the need for manual corrections and labeling during pose model training while maintaining high accuracy. Our approach achieved an average HOTA score of 72.3%, over 10 points higher than other state-of-the-art methods without VM, and resulted in 0 ID switches. Beyond improving performance in handling occlusions and minimizing ID switches, our framework could substantially increase the time and cost efficiency compared to traditional manual annotation.

cs.CV↗

arXiv2024

Quantum Monte Carlo and density functional theory study of strain and magnetism in 2D 1T-VSe$_2$ with charge density wave states

Two-dimensional (2D) 1T-VSe$_2$ has prompted significant interest due to the discrepancies regarding alleged ferromagnetism (FM) at room temperature, charge density wave (CDW) states and the interplay between the two. We employed a combined Diffusion Monte Carlo (DMC) and density functional theory (DFT) approach to accurately investigate the magnetic properties, CDW states, and their response to strain in monolayer 1T-VSe$_2$. Our calculations show the delicate competition between various phases, revealing critical insights into the relationship between their energetic and structural properties. We performed classical Monte Carlo simulations informed by our DMC and DFT results, and found the magnetic transition temperature ($T_c$) of the undistorted (non-CDW) FM phase to be 228 K and the distorted (CDW) phase to be 68 K. Additionally, we studied the response of biaxial strain on the energetic stability and magnetic properties of various phases of 2D 1T-VSe$_2$ and found that small amounts of strain can increase the $T_c$, suggesting a promising route for engineering and enhancing magnetic behavior. Finally, we synthesized 1T-VSe$_2$ and performed Raman spectroscopy measurements, which were in close agreement with our calculated results, validating our computational approach. Our work emphasizes the role of highly accurate DMC methods in advancing the understanding of monolayer 1T-VSe$_2$ and provides a robust framework for future studies of 2D magnetic materials.

cond-mat.mtrl-sci↗

arXiv2024

Exploring control of the emergent exciton insulator state in 1T-TiSe$_2$ monolayer by state-of-the-art theory models

The layered transition metal dichalcogenide 1T-TiSe$_2$ is of great research interest, having intriguing properties of charge density waves (CDW) and superconductivity under doping or pressurizing. The monolayer form of 1T-TiSe$_2$ also shows a CDW with a higher transition temperature T_c than the bulk, indicating a stronger CDW interaction. By using the meta-generalized gradient approximation (metaGGA)-based model Bethe-Salpeter Equation (BSE) and many-body perturbation GW+BSE methods, we calculate the exciton binding energies and electron energy loss spectrum (EELS) for the 1T-TiSe$_2$ monolayer under different in-plane biaxial strains. We find that even without strain the 1T-TiSe$_2$ monolayer can have negative exciton energies at the Brillouin zone boundary point M, with a binding energy larger than the gap. The calculated EELS reinforces this picture, indicating EI (exciton insulator) states in 1T-TiSe$_2$ monolayer even without strain. The Wannier-Mott formula calculations of exciton binding energy corroborate results from GW+BSE. Small compressive strains enhance the EI state, and for tensile strains slightly less than 3%, the EI state in this monolayer persists. At large tensile strains, the material makes a transition to a normal semiconductor. Our results provide important information for understanding the quantum nature of this two-dimensional (2D) material. Our results from the standard G0W0@PBE+SOC+U+BSE approach are not qualitatively different from those of a more computationally efficient metaGGA-based SCAN+SOC+U+mBSE+$f_{xc}^{loc}$ approach that employs a model BSE.

physics.comp-ph↗

arXiv2024

Inside the Working Mechanism of Meta-generalized Gradient Density Functional Approximations: The Example of Quantum Spin-Hall Insulator 1T`-WTe2

Quantum spin Hall (QSH) insulators have attracted intensive experimental and theoretical studies due to their beneficial applications in spintronic devices. Density functional theory (DFT) meets challenges when describing the electronic structure of QSH materials. Only the Heyd-Scuseria-Ernzerhof (HSE06) with spin-orbit coupling (SOC) is effective in revealing the band opening in the typical QSH 1T`-WTe2, but with increased computational demands. Here, using DFT, Wannier function simulations, the screened hybrid HSE06 functional, and first-principles-based many body perturbation theory GW, we investigate the sensitive electronic structure in monolayer 1T`-WTe2, with advanced meta-generalized gradient (meta-GGA) density functional approximations. The success of the recent SCAN and r2SCAN meta-GGAs left their predecessor meta-GGA made very simple (MVS) ignored by the scientific community. Largely unnoticed were the increased band gaps of MVS compared to any semilocal approximation including SCAN. We find that the non-empirical MVS approximation yields a positive fundamental band gap, without any help from exact exchange, Hubbard U, or SOC correction. We explain the success of the meta-GGA MVS for the band gap in 1T`-WTe2 by presenting two working mechanisms in meta-GGA approximations. Besides, we point out the difficulty of using G0W0 for 1T`-WTe2. Although the single shot GW correction with an MVS reference yields a smaller band gap than GW with PBE, the G0W0@MVS is still not suitable for simulating 1T`-WTe2, due to its negative band gap. These DFT and beyond DFT results highlight the importance of meta-GGAs and novel construction schemes with enhanced kinetic energy density dependence. The MVS approximation re-appears as an appealing alternative for accurately describing 1T`-WTe2, paving an efficient way for exploring other two-dimensional QSH materials in high-throughput calculations.

cond-mat.mtrl-sci↗

arXiv2024

A Pathway to Efficient Simulations of Charge Density Waves in Transition Metal Dichalcogenides: A Case Study for TiSe2

Charge density waves (CDWs) in transition metal dichalcogenides are the subject of growing scientific interest due to their rich interplay with exotic phases of matter and their potential technological applications. Here, using density functional theory with advanced meta-generalized gradient approximations (meta-GGAs) and linear response time-dependent density functional theory (TDDFT) with state-of-the-art exchange-correlation kernels, we investigate the electronic, vibrational, and optical properties in 1T-TiSe2 with and without CDW. In both bulk and monolayer TiSe2, the electronic bands and phonon dispersions in either normal (semi-metallic) or CDW (semiconducting) phase are described well via meta-GGAs, which separate the valence and conduction bands just as HSE06 does but with significantly more computational feasibility. Instead of the underestimated gap with standard exchange-correlation approximations and the overestimated gap with screened hybrid functional HSE06, the band gap of the monolayer TiSe2 CDW phase calculated by the meta-GGA MVS (151 meV) is consistent with the angle-resolved photoemission spectroscopy (ARPES) gap of 153 meV measured at 10 K. In addition, the gap of bulk TiSe2 CDW phase reaches 67 meV within the TASK approximation, close to the ARPES gap of 82 meV. Regarding excitations of many-body nature, for bulk TiSe2 in normal and CDW phases, the experimentally observed humps of electron energy loss spectroscopy and plasmon peak are successfully reproduced in TDDFT, without an obvious kernel dependence. To unleash the full scientific and technological potential of CDWs in transition metal dichalcogenides, the chemical doping, heterostructure engineering, and pump-probe techniques are needed. Our study opens the door to simulating these complexities in CDW compounds from first principles by revealing meta-GGAs as an accurate low-cost alternative to HSE06.

cond-mat.mtrl-sci↗

arXiv2024

A diffusion MRI tractography atlas for concurrent white matter mapping across Eastern and Western populations

The study of brain differences across Eastern and Western populations provides vital insights for understanding potential cultural and genetic influences on cognition and mental health. Diffusion MRI (dMRI) tractography is an important tool in assessing white matter (WM) connectivity and brain tissue microstructure across different populations. However, a comprehensive investigation into WM fiber tracts between Eastern and Western populations is challenged due to the lack of a cross-population WM atlas and the large site-specific variability of dMRI data. This study presents a dMRI tractography atlas, namely the East-West WM Atlas, for concurrent WM mapping between Eastern and Western populations and creates a large, harmonized dMRI dataset (n=306) based on the Human Connectome Project and the Chinese Human Connectome Project. The curated WM atlas, as well as subject-specific data including the harmonized dMRI data, the whole brain tractography data, and parcellated WM fiber tracts and their diffusion measures, are publicly released. This resource is a valuable addition to facilitating the exploration of brain commonalities and differences across diverse cultural backgrounds.

q-bio.NC↗