arXiv subjects

Xintong Li

Publications and source records attributed to Xintong Li.

At least 55 records · Page 3Linked to original sources

arXiv2025

From Selection to Generation: A Survey of LLM-based Active Learning

Active Learning (AL) has been a powerful paradigm for improving model efficiency and performance by selecting the most informative data points for labeling and training. In recent active learning frameworks, Large Language Models (LLMs) have been employed not only for selection but also for generating entirely new data instances and providing more cost-effective annotations. Motivated by the increasing importance of high-quality data and efficient model training in the era of LLMs, we present a comprehensive survey on LLM-based Active Learning. We introduce an intuitive taxonomy that categorizes these techniques and discuss the transformative roles LLMs can play in the active learning loop. We further examine the impact of AL on LLM learning paradigms and its applications across various domains. Finally, we identify open challenges and propose future research directions. This survey aims to serve as an up-to-date resource for researchers and practitioners seeking to gain an intuitive understanding of LLM-based AL techniques and deploy them to new applications.

cs.LG↗

arXiv2025

Facial Foundational Model Advances Early Warning of Coronary Artery Disease from Live Videos with DigitalShadow

Global population aging presents increasing challenges to healthcare systems, with coronary artery disease (CAD) responsible for approximately 17.8 million deaths annually, making it a leading cause of global mortality. As CAD is largely preventable, early detection and proactive management are essential. In this work, we introduce DigitalShadow, an advanced early warning system for CAD, powered by a fine-tuned facial foundation model. The system is pre-trained on 21 million facial images and subsequently fine-tuned into LiveCAD, a specialized CAD risk assessment model trained on 7,004 facial images from 1,751 subjects across four hospitals in China. DigitalShadow functions passively and contactlessly, extracting facial features from live video streams without requiring active user engagement. Integrated with a personalized database, it generates natural language risk reports and individualized health recommendations. With privacy as a core design principle, DigitalShadow supports local deployment to ensure secure handling of user data.

cs.CV↗

arXiv2025

ASRL:A robust loss function with potential for development

In this article, we proposed a partition:wise robust loss function based on the previous robust loss function. The characteristics of this loss function are that it achieves high robustness and a wide range of applicability through partition-wise design and adaptive parameter adjustment. Finally, the advantages and development potential of this loss function were verified by applying this loss function to the regression question and using five different datasets (with different dimensions, different sample numbers, and different fields) to compare with the other loss functions. The results of multiple experiments have proven the advantages of our loss function .

cs.LG↗

arXiv2025

A Survey on Personalized and Pluralistic Preference Alignment in Large Language Models

Personalized preference alignment for large language models (LLMs), the process of tailoring LLMs to individual users' preferences, is an emerging research direction spanning the area of NLP and personalization. In this survey, we present an analysis of works on personalized alignment and modeling for LLMs. We introduce a taxonomy of preference alignment techniques, including training time, inference time, and additionally, user-modeling based methods. We provide analysis and discussion on the strengths and limitations of each group of techniques and then cover evaluation, benchmarks, as well as open problems in the field.

cs.CL↗

arXiv2025

Toward Multi-Session Personalized Conversation: A Large-Scale Dataset and Hierarchical Tree Framework for Implicit Reasoning

There has been a surge in the use of large language models (LLM) conversational agents to generate responses based on long-term history from multiple sessions. However, existing long-term open-domain dialogue datasets lack complex, real-world personalization and fail to capture implicit reasoning-where relevant information is embedded in subtle, syntactic, or semantically distant connections rather than explicit statements. In such cases, traditional retrieval methods fail to capture relevant context, and long-context modeling also becomes inefficient due to numerous complicated persona-related details. To address this gap, we introduce ImplexConv, a large-scale long-term dataset with 2,500 examples, each containing approximately 100 conversation sessions, designed to study implicit reasoning in personalized dialogues. Additionally, we propose TaciTree, a novel hierarchical tree framework that structures conversation history into multiple levels of summarization. Instead of brute-force searching all data, TaciTree enables an efficient, level-based retrieval process where models refine their search by progressively selecting relevant details. Our experiments demonstrate that TaciTree significantly improves the ability of LLMs to reason over long-term conversations with implicit contextual dependencies.

cs.CL↗

arXiv2025

Active Learning for Direct Preference Optimization

Direct preference optimization (DPO) is a form of reinforcement learning from human feedback (RLHF) where the policy is learned directly from preferential feedback. Although many models of human preferences exist, the critical task of selecting the most informative feedback for training them is under-explored. We propose an active learning framework for DPO, which can be applied to collect human feedback online or to choose the most informative subset of already collected feedback offline. We propose efficient algorithms for both settings. The key idea is to linearize the DPO objective at the last layer of the neural network representation of the optimized policy and then compute the D-optimal design to collect preferential feedback. We prove that the errors in our DPO logit estimates diminish with more feedback. We show the effectiveness of our algorithms empirically in the setting that matches our theory and also on large language models.

cs.LG↗

arXiv2024

FAP-CD: Fairness-Driven Age-Friendly Community Planning via Conditional Diffusion Generation

As global populations age rapidly, incorporating age-specific considerations into urban planning has become essential to addressing the urgent demand for age-friendly built environments and ensuring sustainable urban development. However, current practices often overlook these considerations, resulting in inadequate and unevenly distributed elderly services in cities. There is a pressing need for equitable and optimized urban renewal strategies to support effective age-friendly planning. To address this challenge, we propose a novel framework, Fairness-driven Age-friendly community Planning via Conditional Diffusion generation (FAP-CD). FAP-CD leverages a conditioned graph denoising diffusion probabilistic model to learn the joint probability distribution of aging facilities and their spatial relationships at a fine-grained regional level. Our framework generates optimized facility distributions by iteratively refining noisy graphs, conditioned on the needs of the elderly during the diffusion process. Key innovations include a demand-fairness pre-training module that integrates community demand features and facility characteristics using an attention mechanism and min-max optimization, ensuring equitable service distribution across regions. Additionally, a discrete graph structure captures walkable accessibility within regional road networks, guiding model sampling. To enhance information integration, we design a graph denoising network with an attribute augmentation module and a hybrid graph message aggregation module, combining local and global node and edge information. Empirical results across multiple metrics demonstrate the effectiveness of FAP-CD in balancing age-friendly needs with regional equity, achieving an average improvement of 41% over competitive baseline models.

cs.AI↗

arXiv2024

OCEAN: Offline Chain-of-thought Evaluation and Alignment in Large Language Models

Offline evaluation of LLMs is crucial in understanding their capacities, though current methods remain underexplored in existing research. In this work, we focus on the offline evaluation of the chain-of-thought capabilities and show how to optimize LLMs based on the proposed evaluation method. To enable offline feedback with rich knowledge and reasoning paths, we use knowledge graphs (e.g., Wikidata5m) to provide feedback on the generated chain of thoughts. Due to the heterogeneity between LLM reasoning and KG structures, direct interaction and feedback from KGs on LLM behavior are challenging, as they require accurate entity linking and grounding of LLM-generated chains of thought in the KG. To address the above challenge, we propose an offline chain-of-thought evaluation framework, OCEAN, which models chain-of-thought reasoning in LLMs as an MDP and evaluate the policy's alignment with KG preference modeling. To overcome the reasoning heterogeneity and grounding problems, we leverage on-policy KG exploration and RL to model a KG policy that generates token-level likelihood distributions for LLM-generated chain-of-thought reasoning paths, simulating KG reasoning preference. Then we incorporate the knowledge-graph feedback on the validity and alignment of the generated reasoning paths into inverse propensity scores and propose KG-IPS estimator. Theoretically, we prove the unbiasedness of the proposed KG-IPS estimator and provide a lower bound on its variance. With the off-policy evaluated value function, we can directly enable off-policy optimization to further enhance chain-of-thought alignment. Our empirical study shows that OCEAN can be efficiently optimized for generating chain-of-thought reasoning paths with higher estimated values without affecting LLMs' general abilities in downstream tasks or their internal knowledge.

cs.LG↗

arXiv2024

Visual Prompting in Multimodal Large Language Models: A Survey

Multimodal large language models (MLLMs) equip pre-trained large-language models (LLMs) with visual capabilities. While textual prompting in LLMs has been widely studied, visual prompting has emerged for more fine-grained and free-form visual instructions. This paper presents the first comprehensive survey on visual prompting methods in MLLMs, focusing on visual prompting, prompt generation, compositional reasoning, and prompt learning. We categorize existing visual prompts and discuss generative methods for automatic prompt annotations on the images. We also examine visual prompting methods that enable better alignment between visual encoders and backbone LLMs, concerning MLLM's visual grounding, object referring, and compositional reasoning abilities. In addition, we provide a summary of model training and in-context learning methods to improve MLLM's perception and understanding of visual prompts. This paper examines visual prompting methods developed in MLLMs and provides a vision of the future of these methods.

cs.LG↗

arXiv2024

Negligible Normal Fluid in Superconducting State of Heavily Overdoped Bi$_2$Sr$_2$CaCu$_2$O$_{8+δ}$ Detected by Ultra-Low Temperature Angle-Resolved Photoemission Spectroscopy

In high temperature cuprate superconductors, it was found that in the overdoped region the superfluid density decreases with the increase of hole doping. One natural question is whether there exists normal fluid in the superconducting state in the overdoped region. In this paper, we have carried out high-resolution ultra-low temperature laser-based angle-resolved photoemission measurements on a heavily overdoped Bi2212 sample with a $T_{\mathrm{c}}$ of 48 K. We find that this heavily overdoped Bi2212 remains in the strong coupling regime with $2 \mathitΔ_0 / k_{\mathrm{B}} T_{\mathrm{c}}=5.8$. The single-particle scattering rate is very small along the nodal direction ($\sim$5 meV) and increases as the momentum moves from the nodal to the antinodal regions. A hard superconducting gap opening is observed near the antinodal region with the spectral weight at the Fermi level fully suppressed to zero. The normal fluid is found to be negligibly small in the superconducting state of this heavily overdoped Bi2212. These results provide key information to understand the high $T_\mathrm{c}$ mechanism in the cuprate superconductors.

cond-mat.supr-con↗

arXiv2024

Open-world Multi-label Text Classification with Extremely Weak Supervision

We study open-world multi-label text classification under extremely weak supervision (XWS), where the user only provides a brief description for classification objectives without any labels or ground-truth label space. Similar single-label XWS settings have been explored recently, however, these methods cannot be easily adapted for multi-label. We observe that (1) most documents have a dominant class covering the majority of content and (2) long-tail labels would appear in some documents as a dominant class. Therefore, we first utilize the user description to prompt a large language model (LLM) for dominant keyphrases of a subset of raw documents, and then construct a (initial) label space via clustering. We further apply a zero-shot multi-label classifier to locate the documents with small top predicted scores, so we can revisit their dominant keyphrases for more long-tail labels. We iterate this process to discover a comprehensive label space and construct a multi-label classifier as a novel method, X-MLClass. X-MLClass exhibits a remarkable increase in ground-truth label space coverage on various datasets, for example, a 40% improvement on the AAPD dataset over topic modeling and keyword extraction methods. Moreover, X-MLClass achieves the best end-to-end multi-label classification accuracy.

cs.CL↗

arXiv2024

Probing critical spin fluctuations with a composite magnetoelectric method: A case study on a Kitaev spin liquid candidate Na$_3$Co$_2$SbO$_6$

In correlated quantum materials, divergent critical fluctuations near the quantum critical point are often closely associated with exotic quantum phases of matter, such as unconventional superconductivity and quantum spin liquids. Here we present a simple yet highly sensitive composite magnetoelectric (ME) method for detecting the critical spin fluctuations in quantum magnets. The ME signal is proportional the magnetostriction coefficient, which directly probes the product of magnetization and spin-spin correlation. As a demonstration, the composite ME method is applied to a Kitaev quantum spin liquid candidate Na$_3$Co$_2$SbO$_6$, which shows signs of magnetic field-induced quantum criticality. Notably, the ME signal prominently diverges at the magnetic field-induced tricritical points, particularly at a tricritical point that lies in close proximity to a zero-temperature quantum critical point. A crucial aspect of these tricritical points is their tunability through the modification of the in-plane magnetic field's direction. The direction of magnetic field can thus serve as a handful yet important tuning parameter, alongside pressure and chemical doping, for searching quantum critical points in quantum magnets with pronounced magnetic anisotropy.

cond-mat.str-el↗

arXiv2024

Development of a Chinese Human-Automation Trust Scale

The development of a reliable and valid assessment tool of human-automation trust is an important topic. This study aimed to develop a Chinese version of human-automation trust scale (C-HATS) with reasonable reliability and validity based on Lee and See (2004)'s trust model. After three phases of assessments including exploratory factor analysis, item analysis, and confirmatory factor analysis, different dimensions and items were considered for initial and posttask human-automation trust. For post-task trust, the scale had three dimensions and 11 items and reflected Lee and See (2004)'s model, whereas different from Lee and See (2004)'s model, the final scale had 14 items but only two dimensions for initial trust. Nevertheless, for both initial and post-task trust, reasonable reliability and validity of the scale were verified with various consumer automation products. Although further verification is still necessary, the developed C-HATS could be used to effectively assess human-automation trust in the Chinese context.

cs.HC↗

arXiv2024

Field-induced phase transitions and quantum criticality in a honeycomb antiferromagnet Na3Co2SbO6

We performed 23Na NMR measurements on a single-domain crystal of the Kitaev material Na3Co2SbO6, with magnetic field applied along the crystalline a axis. A positive Curie-Weiss constant is obtained from the NMR Knight shift, which suggests the existence of ferromagnetic exchange couplings. The antiferromagnetic ordering is found to be suppressed at a field of 1.9 T. Inside the ordered phase, our data reveal two additional phase transitions. At 1.9 T, the spin-lattice relaxation rate 1/23T1 establishes a quantum critical behavior at high temperatures. However, at low temperatures, a gapped behavior is observed at the critical field, which suggests a weakly first-order transition instead and a possible field-induced quantum spin liquid. Our results reveal complex microscopic interactions in the system, which may help to search for possible quantum spin liquids.

cond-mat.str-el↗

arXiv2023

AutoWS-Bench-101: Benchmarking Automated Weak Supervision with 100 Labels

Weak supervision (WS) is a powerful method to build labeled datasets for training supervised models in the face of little-to-no labeled data. It replaces hand-labeling data with aggregating multiple noisy-but-cheap label estimates expressed by labeling functions (LFs). While it has been used successfully in many domains, weak supervision's application scope is limited by the difficulty of constructing labeling functions for domains with complex or high-dimensional features. To address this, a handful of methods have proposed automating the LF design process using a small set of ground truth labels. In this work, we introduce AutoWS-Bench-101: a framework for evaluating automated WS (AutoWS) techniques in challenging WS settings -- a set of diverse application domains on which it has been previously difficult or impossible to apply traditional WS techniques. While AutoWS is a promising direction toward expanding the application-scope of WS, the emergence of powerful methods such as zero-shot foundation models reveals the need to understand how AutoWS techniques compare or cooperate with modern zero-shot or few-shot learners. This informs the central question of AutoWS-Bench-101: given an initial set of 100 labels for each task, we ask whether a practitioner should use an AutoWS method to generate additional labels or use some simpler baseline, such as zero-shot predictions from a foundation model or supervised learning. We observe that in many settings, it is necessary for AutoWS methods to incorporate signal from foundation models if they are to outperform simple few-shot baselines, and AutoWS-Bench-101 promotes future research in this direction. We conclude with a thorough ablation study of AutoWS methods.

cs.LG↗

arXiv2023

Visualizing the Zhang-Rice singlet, molecular orbitals and pair formation in cuprate

The parent compound of cuprates is a charge-transfer-type Mott insulator with strong hybridization between the Cu $3d_{\mathrm x^2-y^2}$ and O $2p$ orbitals. A key question concerning the pairing mechanism is the behavior of doped holes in the antiferromagnetic (AF) Mott insulator background, which is a prototypical quantum many-body problem. It was proposed that doped hole on the O site tends to form a singlet, known as Zhang-Rice singlet (ZRS), with the unpaired Cu spin. But experimentally little is known about the properties of a single hole and the interplay between them that leads to superconductivity. Here we use scanning tunneling microscopy to visualize the electronic states in hole-doped $\mathrm{Ca_2CuO_2Cl_2}$, aiming to establish the atomic-scale local basis for pair formation. A single doped hole is shown to have an in-gap state and a clover-shaped spatial distribution that can be attributed to a localized ZRS. When the dopants are close enough, they develop delocalized molecular orbitals with characteristic stripe- and ladder-shaped patterns, accompanied by the opening of a small gap around the Fermi level ($E_{\mathrm F}$). With increasing doping, the molecular orbitals proliferate in space and gradually form densely packed plaquettes, but the stripe and ladder patterns remain nearly the same. The low-energy electronic states of the molecular orbitals are intimately related to the local pairing properties, thus play a vitally important role in the emergence of superconductivity. We propose that the Cooper pair is formed by two holes occupying the stripe-like molecular orbital, while the attractive interaction is mediated by the AF spin background.

cond-mat.supr-con↗

arXiv2023

Easy-plane multi-$\mathbf{q}$ magnetic ground state of Na$_3$Co$_2$SbO$_6$

Na$_3$Co$_2$SbO$_6$ is a potential Kitaev magnet with a monoclinic layered crystal structure. Recent investigations of the $C_3$-symmetric sister compound Na$_2$Co$_2$TeO$_6$ have uncovered a unique triple-$\mathbf{q}$ magnetic ground state, as opposed to a single-$\mathbf{q}$ (zigzag) one, prompting us to examine the influence of the reduced structural symmetry of Na$_3$Co$_2$SbO$_6$ on its ground state. Neutron diffraction data obtained on a twin-free crystal reveal that the ground state remains a multi-$\mathbf{q}$ state, despite the system's strong in-plane anisotropy. This robustness of multi-$\mathbf{q}$ orders suggests that they are driven by a common mechanism in the honeycomb cobaltates, such as higher-order magnetic interactions. Spin-polarized neutron diffraction results show that the ordered moments are entirely in-plane, with each staggered component orthogonal to the propagating wave vector. The inferred ground state favors a so-called XXZ easy-plane anisotropic starting point for the microscopic model over a Kitaev one, and features unequal ordered moments reduced by strong quantum fluctuations.

cond-mat.str-el↗

arXiv2023

Emergent normal fluid in the superconducting ground state of overdoped cuprates

The microscopic mechanism for the disappearance of superconductivity in overdoped cuprates is still under heated debate. Here we use scanning tunneling spectroscopy to investigate the evolution of quasiparticle interference phenomenon in $\rm Bi_2Sr_2CuO_{6+δ}$ over a wide range of hole densities. We find that when the system enters the overdoped regime, a peculiar quasiparticle interference wavevector with quarter-circle pattern starts to emerge even at zero bias, and its intensity grows with increasing doping level. Its energy dispersion is incompatible with the octet model for d-wave superconductivity, but is highly consistent with the scattering interference of gapless normal carriers. The weight of the gapless quasiparticle interference is mainly located at the antinodes and is independent of temperature. We propose that the normal fluid emerges from the pair-breaking scattering between flat antinodal bands in the quantum ground state, which is the primary cause for the reduction of superfluid density and suppression of superconductivity in overdoped cuprates.

cond-mat.supr-con↗