{
  "generated": "2026-05-03",
  "author": "Tianchen Zhao",
  "topic_clusters": [
    {
      "id": 1,
      "name": "Deepfake Detection and Face Security",
      "papers": [
        "learning-self-consistency-for-deepfake-detection",
        "principles-of-designing-robust-remote-face-anti-spoofing-systems",
        "optimal-transport-guided-source-free-adaptation-for-face-anti-spoofing",
        "authguard-generalizable-deepfake-detection-via-language-guidance"
      ]
    },
    {
      "id": 2,
      "name": "Generative Modeling, Robustness, and Vision-Language Adaptation",
      "papers": [
        "diversity-sensitive-conditional-generative-adversarial-networks",
        "adversarial-defense-via-learning-to-generate-diverse-attacks",
        "salient-concept-aware-generative-data-augmentation",
        "model-diagnosis-and-correction-via-linguistic-and-implicit-attribute-editing",
        "decoupling-vision-and-language-codebook-anchored-visual-adaptation"
      ]
    },
    {
      "id": 3,
      "name": "Neural Quantum States and Scientific Machine Learning",
      "papers": [
        "scalable-neural-quantum-states-architecture-for-quantum-chemistry",
        "natural-evolution-strategies-and-variational-monte-carlo",
        "overcoming-barriers-to-scalability-in-variational-quantum-monte-carlo",
        "natural-evolution-strategies-and-quantum-approximate-optimization",
        "meta-variational-monte-carlo-arxiv",
        "meta-variational-quantum-monte-carlo",
        "neural-quantum-states-for-scientific-computing"
      ]
    },
    {
      "id": 4,
      "name": "Quantum-Inspired Scientific Computing and Finance",
      "papers": [
        "quantum-inspired-variational-algorithms-for-partial-differential-equations"
      ]
    }
  ],
  "papers": [
    {
      "title": "Learning Self-Consistency for Deepfake Detection",
      "authors": ["Tianchen Zhao", "Xiang Xu", "Mingze Xu", "Hui Ding", "Yuanjun Xiong", "Wei Xia"],
      "venue": "ICCV",
      "year": 2021,
      "citations": 538,
      "slug": "learning-self-consistency-for-deepfake-detection",
      "abstract": "We propose a new method to detect deepfake images using the cue of the source feature inconsistency within the forged images. It is based on the hypothesis that images' distinct source features can be preserved and extracted after going through state-of-the-art deepfake generation processes. We introduce a novel representation learning approach, called pair-wise self-consistency learning (PCL), for training ConvNets to extract these source features and detect deepfake images. It is accompanied by a new image synthesis approach, called inconsistency image generator (I2G), to provide richly annotated training data for PCL. Experimental results on seven popular datasets show that our models improve averaged AUC over the state of the art from 96.45% to 98.05% in the in-dataset evaluation and from 86.03% to 92.18% in the cross-dataset evaluation.",
      "one_sentence_summary": "Detects deepfakes by learning to extract and compare source-feature inconsistencies within forged images, achieving state-of-the-art cross-dataset generalization.",
      "key_contributions": [
        "Pair-wise self-consistency learning (PCL) for extracting source features from forged images",
        "Inconsistency image generator (I2G) for creating richly annotated training data",
        "State-of-the-art cross-dataset deepfake detection (86.03% -> 92.18% AUC)"
      ],
      "citation_identity": "Cite for: generalizable deepfake detection, self-consistency learning for face forgery detection, source feature inconsistency, patch-wise consistency learning, cross-dataset deepfake detection",
      "search_phrases": ["deepfake detection self-consistency", "generalizable deepfake detection", "source feature inconsistency", "patch-wise consistency learning", "cross-dataset face forgery detection", "robust deepfake detection"],
      "topic_cluster": 1,
      "related_papers": [
        "principles-of-designing-robust-remote-face-anti-spoofing-systems",
        "optimal-transport-guided-source-free-adaptation-for-face-anti-spoofing",
        "authguard-generalizable-deepfake-detection-via-language-guidance"
      ],
      "links": {
        "arxiv": "https://arxiv.org/abs/2012.09311",
        "pdf": "https://arxiv.org/pdf/2012.09311",
        "code": "TODO"
      },
      "bibtex_key": "zhao2021selfconsistency"
    },
    {
      "title": "Diversity-Sensitive Conditional Generative Adversarial Networks",
      "authors": ["Dingdong Yang", "Seunghoon Hong", "Yunseok Jang", "Tianchen Zhao", "Honglak Lee"],
      "venue": "ICLR",
      "year": 2019,
      "citations": 268,
      "slug": "diversity-sensitive-conditional-generative-adversarial-networks",
      "abstract": "We propose a simple yet highly effective method that addresses the mode-collapse problem in the Conditional Generative Adversarial Network (cGAN). Although conditional distributions are multi-modal (i.e., having many modes) in practice, most cGAN approaches tend to learn an overly simplified distribution where an input is always mapped to a single output regardless of variations in latent code. To address such issue, we propose to explicitly regularize the generator to produce diverse outputs depending on latent codes. The proposed regularization is simple, general, and can be easily integrated into most conditional GAN objectives. Additionally, explicit regularization on generator allows our method to control a balance between visual quality and diversity. We demonstrate the effectiveness of our method on three conditional generation tasks: image-to-image translation, image inpainting, and future video prediction. We show that simple addition of our regularization to existing models leads to surprisingly diverse generations, substantially outperforming the previous approaches for multi-modal conditional generation specifically designed in each individual task.",
      "one_sentence_summary": "Proposes a simple diversity-sensitive regularization for conditional GANs that explicitly encourages diverse outputs, solving mode collapse across image translation, inpainting, and video prediction.",
      "key_contributions": [
        "Diversity-sensitive regularization for conditional GANs that is simple and task-agnostic",
        "Controllable balance between visual quality and output diversity",
        "State-of-the-art multi-modal conditional generation on three tasks"
      ],
      "citation_identity": "Cite for: conditional GANs, diverse conditional image generation, mode diversity in generative models, diversity-sensitive objectives for GAN training",
      "search_phrases": ["diversity-sensitive conditional GAN", "diverse conditional image generation", "mode diversity GAN", "conditional generative adversarial networks", "diversity loss for GANs"],
      "topic_cluster": 2,
      "related_papers": [
        "adversarial-defense-via-learning-to-generate-diverse-attacks",
        "salient-concept-aware-generative-data-augmentation"
      ],
      "links": {
        "arxiv": "https://arxiv.org/abs/1901.09024",
        "pdf": "https://arxiv.org/pdf/1901.09024",
        "code": "TODO"
      },
      "bibtex_key": "yang2019diversitysensitive"
    },
    {
      "title": "Adversarial Defense via Learning to Generate Diverse Attacks",
      "authors": ["Yunseok Jang", "Tianchen Zhao", "Seunghoon Hong", "Honglak Lee"],
      "venue": "ICCV",
      "year": 2019,
      "citations": 108,
      "slug": "adversarial-defense-via-learning-to-generate-diverse-attacks",
      "abstract": "TODO: Full abstract unavailable due to Semantic Scholar rate limits. Partial: The paper proposes using a recursive and stochastic generator that produces strong and diverse adversarial perturbations that comprehensively reveal the vulnerability of the target classifier, improving robustness against various attack types on MNIST and CIFAR-10.",
      "one_sentence_summary": "Trains a stochastic generator to produce diverse adversarial attacks, yielding more robust defenses than single-attack adversarial training.",
      "key_contributions": [
        "Recursive and stochastic generator for diverse adversarial perturbations",
        "Comprehensive vulnerability exposure through attack diversity",
        "Improved robustness against various attack types"
      ],
      "citation_identity": "Cite for: adversarial defense, diverse adversarial attack generation, robust training, defenses that use learned attack generators",
      "search_phrases": ["adversarial defense diverse attacks", "learning to generate diverse attacks", "adversarial training attack diversity", "robustness via generated attacks", "diverse adversarial examples"],
      "topic_cluster": 2,
      "related_papers": [
        "diversity-sensitive-conditional-generative-adversarial-networks",
        "salient-concept-aware-generative-data-augmentation",
        "model-diagnosis-and-correction-via-linguistic-and-implicit-attribute-editing"
      ],
      "links": {
        "pdf": "TODO",
        "code": "TODO"
      },
      "bibtex_key": "jang2019adversarial"
    },
    {
      "title": "Scalable Neural Quantum States Architecture for Quantum Chemistry",
      "authors": ["Tianchen Zhao", "James Stokes", "Shravan Veerapaneni"],
      "venue": "Machine Learning: Science and Technology",
      "year": 2023,
      "citations": 65,
      "slug": "scalable-neural-quantum-states-architecture-for-quantum-chemistry",
      "abstract": "Variational optimization of neural-network representations of quantum states has been successfully applied to solve interacting fermionic problems. Despite rapid developments, significant scalability challenges arise when considering molecules of large scale, which correspond to non-locally interacting quantum spin Hamiltonians consisting of sums of thousands or even millions of Pauli operators. In this work, we introduce scalable parallelization strategies to improve neural-network-based variational quantum Monte Carlo calculations for ab-initio quantum chemistry applications. We establish GPU-supported local energy parallelism to compute the optimization objective for Hamiltonians of potentially complex molecules. Using autoregressive sampling techniques, we demonstrate systematic improvement in wall-clock timings required to achieve CCSD baseline target energies. The performance is further enhanced by accommodating the structure of resultant spin Hamiltonians into the autoregressive sampling ordering. The algorithm achieves promising performance in comparison with the classical approximate methods and exhibits both running time and scalability advantages over existing neural-network based methods.",
      "one_sentence_summary": "Introduces GPU-scalable parallelization strategies for neural-network variational Monte Carlo applied to ab-initio quantum chemistry, achieving systematic speedups over existing neural-network methods.",
      "key_contributions": [
        "GPU-supported local energy parallelism for large-scale quantum chemistry Hamiltonians",
        "Autoregressive sampling with structure-aware ordering for spin Hamiltonians",
        "Scalability and runtime advantages over existing neural-network VMC methods"
      ],
      "citation_identity": "Cite for: scalable neural quantum states, neural-network variational Monte Carlo for quantum chemistry, autoregressive neural wavefunctions, GPU-scaled quantum chemistry",
      "search_phrases": ["scalable neural quantum states", "neural quantum states quantum chemistry", "variational Monte Carlo quantum chemistry", "autoregressive neural quantum states", "GPU-scaled quantum chemistry", "ab initio neural quantum states"],
      "topic_cluster": 3,
      "related_papers": [
        "natural-evolution-strategies-and-variational-monte-carlo",
        "overcoming-barriers-to-scalability-in-variational-quantum-monte-carlo",
        "meta-variational-monte-carlo-arxiv",
        "meta-variational-quantum-monte-carlo",
        "neural-quantum-states-for-scientific-computing"
      ],
      "links": {
        "arxiv": "https://arxiv.org/abs/2208.05637",
        "pdf": "https://arxiv.org/pdf/2208.05637",
        "code": "TODO"
      },
      "bibtex_key": "zhao2023scalable"
    },
    {
      "title": "Natural Evolution Strategies and Variational Monte Carlo",
      "authors": ["Tianchen Zhao", "Giuseppe Carleo", "James Stokes", "Shravan Veerapaneni"],
      "venue": "Machine Learning: Science and Technology",
      "year": 2021,
      "citations": 32,
      "slug": "natural-evolution-strategies-and-variational-monte-carlo",
      "abstract": "A notion of quantum natural evolution strategies is introduced, which provides a geometric synthesis of a number of known quantum/classical algorithms for performing classical black-box optimization. Recent work of Gomes et al. [2019] on heuristic combinatorial optimization using neural quantum states is pedagogically reviewed in this context, emphasizing the connection with natural evolution strategies. The algorithmic framework is illustrated for approximate combinatorial optimization problems, and a systematic strategy is found for improving the approximation ratios. In particular it is found that natural evolution strategies can achieve approximation ratios competitive with widely used heuristic algorithms for Max-Cut, at the expense of increased computation time.",
      "one_sentence_summary": "Introduces quantum natural evolution strategies as a geometric unification of quantum/classical black-box optimization algorithms, achieving competitive Max-Cut approximation ratios.",
      "key_contributions": [
        "Quantum natural evolution strategies as a geometric synthesis of optimization algorithms",
        "Systematic strategy for improving approximation ratios in combinatorial optimization",
        "Competitive Max-Cut results via neural quantum states with NES"
      ],
      "citation_identity": "Cite for: natural evolution strategies for VMC, gradient-free optimization of neural quantum states, optimization methods for quantum many-body wavefunctions",
      "search_phrases": ["natural evolution strategies variational Monte Carlo", "gradient-free variational Monte Carlo", "neural quantum states optimization", "quantum many-body optimization", "VMC natural evolution strategies"],
      "topic_cluster": 3,
      "related_papers": [
        "scalable-neural-quantum-states-architecture-for-quantum-chemistry",
        "overcoming-barriers-to-scalability-in-variational-quantum-monte-carlo",
        "natural-evolution-strategies-and-quantum-approximate-optimization",
        "meta-variational-monte-carlo-arxiv",
        "neural-quantum-states-for-scientific-computing"
      ],
      "links": {
        "arxiv": "https://arxiv.org/abs/2005.04447",
        "pdf": "https://arxiv.org/pdf/2005.04447",
        "code": "TODO"
      },
      "bibtex_key": "zhao2021nes"
    },
    {
      "title": "Overcoming Barriers to Scalability in Variational Quantum Monte Carlo",
      "authors": ["Tianchen Zhao", "Saibal De", "Brian Chen", "James Stokes", "Shravan Veerapaneni"],
      "venue": "SC (International Conference for High Performance Computing, Networking, Storage and Analysis)",
      "year": 2021,
      "citations": 27,
      "slug": "overcoming-barriers-to-scalability-in-variational-quantum-monte-carlo",
      "abstract": "The variational quantum Monte Carlo (VQMC) method received significant attention in the recent past because of its ability to overcome the curse of dimensionality inherent in many-body quantum systems. Close parallels exist between VQMC and the emerging hybrid quantum-classical computational paradigm of variational quantum algorithms. VQMC overcomes the curse of dimensionality by performing alternating steps of Monte Carlo sampling from a parametrized quantum state followed by gradient-based optimization. While VQMC has been applied to solve high-dimensional problems, it is known to be difficult to parallelize, primarily owing to the Markov Chain Monte Carlo (MCMC) sampling step. In this work, we explore the scalability of VQMC when autoregressive models, with exact sampling, are used in place of MCMC. This approach can exploit distributed-memory, shared-memory and/or GPU parallelism in the sampling task without any bottlenecks. In particular, we demonstrate the GPU-scalability of VQMC for solving up to ten-thousand dimensional combinatorial optimization problems.",
      "one_sentence_summary": "Replaces MCMC with autoregressive exact sampling in variational quantum Monte Carlo, enabling GPU-scalable parallelism for up to ten-thousand dimensional problems.",
      "key_contributions": [
        "Autoregressive exact sampling as a replacement for MCMC in VQMC",
        "Distributed-memory, shared-memory, and GPU parallelism without sampling bottlenecks",
        "Demonstration of GPU-scalable VQMC on 10,000-dimensional combinatorial optimization"
      ],
      "citation_identity": "Cite for: scalable VMC, high-performance computing for neural quantum states, distributed VMC, computational bottlenecks in VMC",
      "search_phrases": ["scalable variational quantum Monte Carlo", "HPC variational quantum Monte Carlo", "distributed neural quantum states", "scalability barriers VMC", "high performance quantum Monte Carlo"],
      "topic_cluster": 3,
      "related_papers": [
        "scalable-neural-quantum-states-architecture-for-quantum-chemistry",
        "natural-evolution-strategies-and-variational-monte-carlo",
        "natural-evolution-strategies-and-quantum-approximate-optimization",
        "meta-variational-monte-carlo-arxiv",
        "neural-quantum-states-for-scientific-computing"
      ],
      "links": {
        "arxiv": "https://arxiv.org/abs/2106.13308",
        "pdf": "https://arxiv.org/pdf/2106.13308",
        "code": "TODO"
      },
      "bibtex_key": "zhao2021scalability"
    },
    {
      "title": "Quantum-Inspired Variational Algorithms for Partial Differential Equations: Application to Financial Derivative Pricing",
      "authors": ["Tianchen Zhao", "Chuhao Sun", "Asaf Cohen", "James Stokes", "Shravan Veerapaneni"],
      "venue": "Quantitative Finance",
      "year": 2024,
      "citations": 16,
      "slug": "quantum-inspired-variational-algorithms-for-partial-differential-equations",
      "abstract": "Variational quantum Monte Carlo (VMC) combined with neural-network quantum states offers a novel angle of attack on the curse-of-dimensionality encountered in a particular class of partial differential equations (PDEs); namely, the real- and imaginary time-dependent Schr\u00f6dinger equation. In this paper, we present a simple generalization of VMC applicable to arbitrary time-dependent PDEs, showcasing the technique in the multi-asset Black-Scholes PDE for pricing European options contingent on many correlated underlying assets.",
      "one_sentence_summary": "Generalizes variational quantum Monte Carlo with neural-network quantum states to solve arbitrary time-dependent PDEs, demonstrated on multi-asset Black-Scholes option pricing.",
      "key_contributions": [
        "Generalization of VMC from Schrodinger equation to arbitrary time-dependent PDEs",
        "Application to multi-asset Black-Scholes PDE for European option pricing",
        "Neural-network quantum states for financial derivative pricing"
      ],
      "citation_identity": "Cite for: quantum-inspired variational PDE solvers, variational algorithms for financial derivative pricing, scientific machine learning methods for quantitative finance",
      "search_phrases": ["quantum-inspired variational algorithms PDE", "financial derivative pricing variational algorithm", "variational PDE solver finance", "quantum-inspired finance", "scientific machine learning derivative pricing"],
      "topic_cluster": 4,
      "related_papers": [
        "scalable-neural-quantum-states-architecture-for-quantum-chemistry",
        "natural-evolution-strategies-and-variational-monte-carlo",
        "overcoming-barriers-to-scalability-in-variational-quantum-monte-carlo",
        "neural-quantum-states-for-scientific-computing"
      ],
      "links": {
        "arxiv": "https://arxiv.org/abs/2207.10838",
        "pdf": "https://arxiv.org/pdf/2207.10838",
        "code": "TODO"
      },
      "bibtex_key": "zhao2024quantum"
    },
    {
      "title": "Principles of Designing Robust Remote Face Anti-Spoofing Systems",
      "authors": ["Xiang Xu", "Tianchen Zhao", "Zheng Zhang", "Zhihua Li", "Jon Wu", "Alessandro Achille", "Mani Srivastava"],
      "venue": "arXiv",
      "year": 2024,
      "citations": 10,
      "slug": "principles-of-designing-robust-remote-face-anti-spoofing-systems",
      "abstract": "Protecting digital identities of human face from various attack vectors is paramount, and face anti-spoofing plays a crucial role in this endeavor. Current approaches primarily focus on detecting spoofing attempts within individual frames to detect presentation attacks. However, the emergence of hyper-realistic generative models capable of real-time operation has heightened the risk of digitally generated attacks. In light of these evolving threats, this paper aims to address two key aspects. First, it sheds light on the vulnerabilities of state-of-the-art face anti-spoofing methods against digital attacks. Second, it presents a comprehensive taxonomy of common threats encountered in face anti-spoofing systems. Through a series of experiments, we demonstrate the limitations of current face anti-spoofing detection techniques and their failure to generalize to novel digital attack scenarios. Notably, the existing models struggle with digital injection attacks including adversarial noise, realistic deepfake attacks, and digital replay attacks. To aid in the design and implementation of robust face anti-spoofing systems resilient to these emerging vulnerabilities, the paper proposes key design principles from model accuracy and robustness to pipeline robustness and even platform robustness. Especially, we suggest to implement the proactive face anti-spoofing system using active sensors to significant reduce the risks for unseen attack vectors and improve the user experience.",
      "one_sentence_summary": "Presents a comprehensive taxonomy of face anti-spoofing threats and proposes design principles for building robust systems resilient to digital injection, deepfake, and replay attacks.",
      "key_contributions": [
        "Comprehensive taxonomy of threats to face anti-spoofing systems",
        "Experimental demonstration of vulnerabilities against digital attacks",
        "Design principles from model accuracy to pipeline and platform robustness",
        "Recommendation for proactive face anti-spoofing with active sensors"
      ],
      "citation_identity": "Cite for: remote face anti-spoofing, robust face liveness detection, presentation attack detection, secure remote biometric authentication systems",
      "search_phrases": ["remote face anti-spoofing", "robust face anti-spoofing systems", "face liveness detection", "presentation attack detection", "remote biometric authentication", "face security system design"],
      "topic_cluster": 1,
      "related_papers": [
        "learning-self-consistency-for-deepfake-detection",
        "optimal-transport-guided-source-free-adaptation-for-face-anti-spoofing",
        "authguard-generalizable-deepfake-detection-via-language-guidance"
      ],
      "links": {
        "arxiv": "https://arxiv.org/abs/2406.03684",
        "pdf": "https://arxiv.org/pdf/2406.03684",
        "code": "TODO"
      },
      "bibtex_key": "xu2024principles"
    },
    {
      "title": "Optimal Transport-Guided Source-Free Adaptation for Face Anti-Spoofing",
      "authors": ["Zhuowei Li", "Tianchen Zhao", "Xiang Xu", "Zheng Zhang", "Zhihua Li", "Xuanbai Chen", "Qin Zhang", "Alessandro Bergamo", "Anil K. Jain", "Yifan Xing"],
      "venue": "CVPR",
      "year": 2025,
      "citations": 8,
      "slug": "optimal-transport-guided-source-free-adaptation-for-face-anti-spoofing",
      "abstract": "Developing a face anti-spoofing model that meets the security requirements of clients worldwide is challenging due to the domain gap between training datasets and diverse end-user test data. Moreover, for security and privacy reasons, it is undesirable for clients to share a large amount of their face data with service providers. In this work, we introduce a novel method in which the face anti-spoofing model can be adapted by the client itself to a target domain at test time using only a small sample of data while keeping model parameters and training data inaccessible to the client. Specifically, we develop a prototype-based base model and an optimal transport-guided adaptor that enables adaptation in either a lightweight training or training-free fashion, without updating base model's parameters. Furthermore, we propose geodesic mixup, an optimal transport-based synthesis method that generates augmented training data along the geodesic path between source prototypes and target data distribution. This allows training a lightweight classifier to effectively adapt to target-specific characteristics while retaining essential knowledge learned from the source domain. In cross-domain and cross-attack settings, compared with recent methods, our method achieves average relative improvements of 19.17% in HTER and 8.58% in AUC, respectively.",
      "one_sentence_summary": "Enables privacy-preserving test-time adaptation of face anti-spoofing models via optimal transport-guided prototype matching and geodesic mixup augmentation.",
      "key_contributions": [
        "Prototype-based model with optimal transport-guided test-time adaptation",
        "Geodesic mixup for synthesizing augmented training data along OT paths",
        "Privacy-preserving adaptation without sharing model parameters or training data",
        "19.17% HTER improvement and 8.58% AUC improvement over state-of-the-art"
      ],
      "citation_identity": "Cite for: source-free domain adaptation for face anti-spoofing, optimal transport for biometric security, domain adaptation without source data in presentation attack detection",
      "search_phrases": ["source-free adaptation face anti-spoofing", "optimal transport face anti-spoofing", "source-free domain adaptation biometrics", "presentation attack detection domain adaptation", "face anti-spoofing without source data"],
      "topic_cluster": 1,
      "related_papers": [
        "learning-self-consistency-for-deepfake-detection",
        "principles-of-designing-robust-remote-face-anti-spoofing-systems",
        "authguard-generalizable-deepfake-detection-via-language-guidance"
      ],
      "links": {
        "arxiv": "https://arxiv.org/abs/2503.22984",
        "pdf": "https://arxiv.org/pdf/2503.22984",
        "code": "TODO"
      },
      "bibtex_key": "li2025optimaltransport"
    },
    {
      "title": "Natural Evolution Strategies and Quantum Approximate Optimization",
      "authors": ["Tianchen Zhao", "Giuseppe Carleo", "James Stokes", "Shravan Veerapaneni"],
      "venue": "arXiv",
      "year": 2020,
      "citations": 8,
      "slug": "natural-evolution-strategies-and-quantum-approximate-optimization",
      "abstract": "TODO: Paper #10 and #5 share arXiv ID 2005.04447. Google Scholar may list the arXiv preprint (2020) and the ML:ST published version (2021) as separate entries with slightly different titles. The abstract is the same as paper #5. Verify whether these are truly separate works or the same paper at different stages.",
      "one_sentence_summary": "Applies natural evolution strategies to quantum approximate optimization, achieving competitive Max-Cut approximation ratios via neural quantum states.",
      "key_contributions": [
        "Application of NES to quantum approximate optimization (QAOA) problems",
        "Systematic strategy for improving approximation ratios",
        "Competitive Max-Cut results"
      ],
      "citation_identity": "Cite for: natural evolution strategies for QAOA, black-box optimization for QAOA, gradient-free optimization of variational quantum algorithms",
      "search_phrases": ["natural evolution strategies QAOA", "quantum approximate optimization natural evolution strategies", "gradient-free QAOA", "variational quantum algorithms optimization", "black-box optimization quantum circuits"],
      "topic_cluster": 3,
      "related_papers": [
        "natural-evolution-strategies-and-variational-monte-carlo",
        "overcoming-barriers-to-scalability-in-variational-quantum-monte-carlo",
        "meta-variational-monte-carlo-arxiv",
        "neural-quantum-states-for-scientific-computing"
      ],
      "links": {
        "arxiv": "https://arxiv.org/abs/2005.04447",
        "pdf": "https://arxiv.org/pdf/2005.04447",
        "code": "TODO"
      },
      "bibtex_key": "zhao2020nesqaoa"
    },
    {
      "title": "Meta Variational Monte Carlo",
      "authors": ["Tianchen Zhao", "James Stokes", "Oliver Knitter", "Brian Chen", "Shravan Veerapaneni"],
      "venue": "arXiv",
      "year": 2020,
      "citations": 4,
      "slug": "meta-variational-monte-carlo-arxiv",
      "abstract": "An identification is found between meta-learning and the problem of determining the ground state of a randomly generated Hamiltonian drawn from a known ensemble. A model-agnostic meta-learning approach is proposed to solve the associated learning problem and a preliminary experimental study of random Max-Cut problems indicates that the resulting Meta Variational Monte Carlo accelerates training and improves convergence.",
      "one_sentence_summary": "Connects meta-learning to quantum ground-state problems, proposing Meta Variational Monte Carlo that accelerates training and improves convergence on random Max-Cut instances.",
      "key_contributions": [
        "Identification between meta-learning and random Hamiltonian ground-state problems",
        "Model-agnostic meta-learning approach for variational Monte Carlo",
        "Accelerated training and improved convergence on Max-Cut"
      ],
      "citation_identity": "Cite for: meta-learning for variational Monte Carlo, transferable neural quantum state optimization, learning optimization strategies across quantum systems",
      "search_phrases": ["meta variational Monte Carlo", "meta-learning variational Monte Carlo", "transferable neural quantum states", "learning to optimize VMC", "meta-learning quantum many-body systems", "neural quantum state transfer"],
      "topic_cluster": 3,
      "related_papers": [
        "scalable-neural-quantum-states-architecture-for-quantum-chemistry",
        "natural-evolution-strategies-and-variational-monte-carlo",
        "overcoming-barriers-to-scalability-in-variational-quantum-monte-carlo",
        "meta-variational-quantum-monte-carlo",
        "neural-quantum-states-for-scientific-computing"
      ],
      "links": {
        "arxiv": "https://arxiv.org/abs/2011.10614",
        "pdf": "https://arxiv.org/pdf/2011.10614",
        "code": "TODO"
      },
      "bibtex_key": "zhao2020metavmc"
    },
    {
      "title": "AuthGuard: Generalizable Deepfake Detection via Language Guidance",
      "authors": ["Guangyu Shen", "Zhihua Li", "Xiang Xu", "Tianchen Zhao", "Zheng Zhang", "Dongsheng An", "Zhuowen Tu", "Yifan Xing", "Qin Zhang"],
      "venue": "WACV",
      "year": 2025,
      "citations": 2,
      "slug": "authguard-generalizable-deepfake-detection-via-language-guidance",
      "abstract": "Existing deepfake detection techniques struggle to keep-up with the ever-evolving novel, unseen forgeries methods. This limitation stems from their reliance on statistical artifacts learned during training, which are often tied to specific generation processes that may not be representative of samples from new, unseen deepfake generation methods encountered at test time. We propose that incorporating language guidance can improve deepfake detection generalization by integrating human-like commonsense reasoning -- such as recognizing logical inconsistencies and perceptual anomalies -- alongside statistical cues. To achieve this, we train an expert deepfake vision encoder by combining discriminative classification with image-text contrastive learning, where the text is generated by generalist MLLMs using few-shot prompting. This allows the encoder to extract both language-describable, commonsense deepfake artifacts and statistical forgery artifacts from pixel-level distributions. To further enhance robustness, we integrate data uncertainty learning into vision-language contrastive learning, mitigating noise in image-text supervision. Our expert vision encoder seamlessly interfaces with an LLM, further enabling more generalized and interpretable deepfake detection while also boosting accuracy. The resulting framework, AuthGuard, achieves state-of-the-art deepfake detection accuracy in both in-distribution and out-of-distribution settings, achieving AUC gains of 6.15% on the DFDC dataset and 16.68% on the DF40 dataset. Additionally, AuthGuard significantly enhances deepfake reasoning, improving performance by 24.69% on the DDVQA dataset.",
      "one_sentence_summary": "Integrates language guidance and commonsense reasoning into deepfake detection via vision-language contrastive learning, achieving state-of-the-art generalization across unseen forgery methods.",
      "key_contributions": [
        "Expert deepfake vision encoder combining discriminative and contrastive learning with MLLM-generated text",
        "Data uncertainty learning for robust vision-language contrastive training",
        "State-of-the-art OOD deepfake detection with interpretable reasoning",
        "AUC gains of 6.15% on DFDC and 16.68% on DF40"
      ],
      "citation_identity": "Cite for: language-guided deepfake detection, generalizable face forgery detection, vision-language guidance for biometric security",
      "search_phrases": ["AuthGuard deepfake detection", "language-guided deepfake detection", "generalizable deepfake detection language guidance", "vision-language face forgery detection", "robust face forgery detection"],
      "topic_cluster": 1,
      "related_papers": [
        "learning-self-consistency-for-deepfake-detection",
        "principles-of-designing-robust-remote-face-anti-spoofing-systems",
        "optimal-transport-guided-source-free-adaptation-for-face-anti-spoofing"
      ],
      "links": {
        "arxiv": "https://arxiv.org/abs/2506.04501",
        "pdf": "https://arxiv.org/pdf/2506.04501",
        "code": "TODO"
      },
      "bibtex_key": "shen2025authguard"
    },
    {
      "title": "Salient Concept-Aware Generative Data Augmentation",
      "authors": ["Tianchen Zhao", "Xuanbai Chen", "Zhihua Li", "Jun Fang", "Dongsheng An", "Xiang Xu", "Zhuowen Tu", "Yifan Xing"],
      "venue": "arXiv",
      "year": 2025,
      "citations": 1,
      "slug": "salient-concept-aware-generative-data-augmentation",
      "abstract": "Recent generative data augmentation methods conditioned on both image and text prompts struggle to balance between fidelity and diversity, as it is challenging to preserve essential image details while aligning with varied text prompts. This challenge arises because representations in the synthesis process often become entangled with non-essential input image attributes such as environmental contexts, creating conflicts with text prompts intended to modify these elements. To address this, we propose a personalized image generation framework that uses a salient concept-aware image embedding model to reduce the influence of irrelevant visual details during the synthesis process, thereby maintaining intuitive alignment between image and text inputs. By generating images that better preserve class-discriminative features with additional controlled variations, our framework effectively enhances the diversity of training datasets and thereby improves the robustness of downstream models. Our approach demonstrates superior performance across eight fine-grained vision datasets, outperforming state-of-the-art augmentation methods with averaged classification accuracy improvements by 0.73% and 6.5% under conventional and long-tail settings, respectively.",
      "one_sentence_summary": "Proposes a salient concept-aware generative augmentation framework that disentangles class-discriminative features from irrelevant context, improving fine-grained recognition under both conventional and long-tail settings.",
      "key_contributions": [
        "Salient concept-aware image embedding that reduces irrelevant visual detail influence",
        "Personalized generation preserving class-discriminative features with controlled variation",
        "Superior performance on 8 fine-grained datasets, especially under long-tail settings (+6.5%)"
      ],
      "citation_identity": "Cite for: generative data augmentation, concept-aware augmentation, salient concept modeling, synthetic data generation for visual recognition",
      "search_phrases": ["salient concept-aware generative data augmentation", "concept-aware data augmentation", "generative data augmentation computer vision", "salient concept augmentation", "synthetic data generation visual recognition"],
      "topic_cluster": 2,
      "related_papers": [
        "diversity-sensitive-conditional-generative-adversarial-networks",
        "adversarial-defense-via-learning-to-generate-diverse-attacks",
        "model-diagnosis-and-correction-via-linguistic-and-implicit-attribute-editing",
        "decoupling-vision-and-language-codebook-anchored-visual-adaptation"
      ],
      "links": {
        "arxiv": "https://arxiv.org/abs/2510.15194",
        "pdf": "https://arxiv.org/pdf/2510.15194",
        "code": "TODO"
      },
      "bibtex_key": "zhao2025salient"
    },
    {
      "title": "Model Diagnosis and Correction via Linguistic and Implicit Attribute Editing",
      "authors": ["Xuanbai Chen", "Xiang Xu", "Zhihua Li", "Tianchen Zhao", "Pietro Perona", "Qin Zhang", "Yifan Xing"],
      "venue": "CVPR",
      "year": 2025,
      "citations": 1,
      "slug": "model-diagnosis-and-correction-via-linguistic-and-implicit-attribute-editing",
      "abstract": "How can we troubleshoot a deep visual model, i.e. understand why it makes certain mistakes and further take action to correct its behavior? We design a Model Diagnosis and Correction system (MDC), an automated framework that analyzes the pattern of errors, proposes candidate causes of attributes, conducts hypothesis testing via attribute editing, and ultimately generates counterfactual training samples to improve the performance of the model. Unlike previous methods, in addition to the linguistic attributes, our method also incorporates the analysis for implicit causal attributes, those cannot to be accurately described by natural language. To achieve this, we propose an image editing module capable of leveraging both implicit and linguistic attributes to generate counterfactual images depicting error patterns and further experimentally validate causality relationships. Lastly, we enrich the training set with synthetic samples depicting verified causal attributes and retrain the model, further boosting accuracy and robustness. Extensive experiments on both generalized and specialized domains demonstrate the superiority of MDC in model diagnosis and correction. Specifically, we achieve an average relative improvement of 62.01% in HTER for face security application over state-of-the-art methods.",
      "one_sentence_summary": "Automates model diagnosis and correction by analyzing error patterns, testing causal hypotheses via attribute editing, and generating counterfactual training samples for improved robustness.",
      "key_contributions": [
        "Automated MDC framework for error pattern analysis and causal hypothesis testing",
        "Image editing module for both linguistic and implicit causal attributes",
        "Counterfactual training sample generation for model correction",
        "62.01% relative HTER improvement for face security applications"
      ],
      "citation_identity": "Cite for: model diagnosis, attribute editing, linguistic attribute control, correcting model behavior through interpretable attributes",
      "search_phrases": ["model diagnosis attribute editing", "linguistic attribute editing", "implicit attribute editing", "model correction visual attributes", "interpretable model correction"],
      "topic_cluster": 2,
      "related_papers": [
        "adversarial-defense-via-learning-to-generate-diverse-attacks",
        "salient-concept-aware-generative-data-augmentation",
        "decoupling-vision-and-language-codebook-anchored-visual-adaptation"
      ],
      "links": {
        "pdf": "TODO",
        "code": "TODO"
      },
      "bibtex_key": "chen2025mdc"
    },
    {
      "title": "Decoupling Vision and Language: Codebook Anchored Visual Adaptation",
      "authors": ["Jason Wu", "Tianchen Zhao", "Chang Liu", "Jiarui Cai", "Zheng Zhang", "Zhuowei Li", "Aaditya Singh", "Xiang Xu", "Mani Srivastava", "Jonathan Wu"],
      "venue": "arXiv",
      "year": 2026,
      "citations": 0,
      "slug": "decoupling-vision-and-language-codebook-anchored-visual-adaptation",
      "abstract": "Large Vision-Language Models (LVLMs) use their vision encoders to translate images into representations for downstream reasoning, but the encoders often underperform in domain-specific visual tasks such as medical image diagnosis or fine-grained classification, where representation errors can cascade through the language model, leading to incorrect responses. Existing adaptation methods modify the continuous feature interface between encoder and language model through projector tuning or other parameter-efficient updates, which still couples the two components and requires re-alignment whenever the encoder changes. We introduce CRAFT (Codebook RegulAted Fine-Tuning), a lightweight method that fine-tunes the encoder using a discrete codebook that anchors visual representations to a stable token space, achieving domain adaptation without modifying other parts of the model. This decoupled design allows the adapted encoder to seamlessly boost the performance of LVLMs with different language architectures, as long as they share the same codebook. Empirically, CRAFT achieves an average gain of 13.51% across 10 domain-specific benchmarks such as VQARAD and PlantVillage, while preserving the LLM's linguistic capabilities and outperforming peer methods that operate on continuous tokens.",
      "one_sentence_summary": "Introduces CRAFT, a discrete-codebook method for adapting vision encoders in LVLMs to domain-specific tasks without modifying the language model, enabling portable encoder adaptation across architectures.",
      "key_contributions": [
        "CRAFT: codebook-regulated fine-tuning that decouples vision encoder adaptation from the language model",
        "Discrete codebook anchoring for stable, portable visual representations",
        "13.51% average gain across 10 domain-specific benchmarks while preserving linguistic capabilities"
      ],
      "citation_identity": "Cite for: vision-language adaptation, codebook-anchored visual adaptation, decoupling visual and language representations",
      "search_phrases": ["codebook anchored visual adaptation", "decoupling vision and language", "vision-language model adaptation", "visual adaptation codebook", "efficient vision-language adaptation"],
      "topic_cluster": 2,
      "related_papers": [
        "salient-concept-aware-generative-data-augmentation",
        "model-diagnosis-and-correction-via-linguistic-and-implicit-attribute-editing"
      ],
      "links": {
        "arxiv": "https://arxiv.org/abs/2602.19449",
        "pdf": "https://arxiv.org/pdf/2602.19449",
        "code": "TODO"
      },
      "bibtex_key": "wu2026craft"
    },
    {
      "title": "Meta-Variational Quantum Monte Carlo",
      "authors": ["Tianchen Zhao", "James Stokes", "Oliver Knitter", "Brian Chen", "Shravan Veerapaneni"],
      "venue": "Quantum Machine Intelligence",
      "year": 2023,
      "citations": 0,
      "slug": "meta-variational-quantum-monte-carlo",
      "abstract": "TODO: Journal version of arXiv paper #11 (2011.10614). The abstract is expected to be the same or a revised version of: An identification is found between meta-learning and the problem of determining the ground state of a randomly generated Hamiltonian drawn from a known ensemble. A model-agnostic meta-learning approach is proposed to solve the associated learning problem and a preliminary experimental study of random Max-Cut problems indicates that the resulting Meta Variational Monte Carlo accelerates training and improves convergence.",
      "one_sentence_summary": "Journal version presenting Meta Variational Monte Carlo, which applies meta-learning to accelerate neural quantum state optimization across random Hamiltonian ensembles.",
      "key_contributions": [
        "Meta-learning framework for variational Monte Carlo optimization",
        "Transferable initialization across random Hamiltonian instances",
        "Accelerated training and improved convergence on Max-Cut problems"
      ],
      "citation_identity": "Cite for: meta-learning for variational Monte Carlo, transferable neural quantum state optimization, learning optimization strategies across quantum systems",
      "search_phrases": ["meta variational Monte Carlo", "meta-learning variational Monte Carlo", "transferable neural quantum states", "learning to optimize VMC", "meta-learning quantum many-body systems", "neural quantum state transfer"],
      "topic_cluster": 3,
      "related_papers": [
        "scalable-neural-quantum-states-architecture-for-quantum-chemistry",
        "natural-evolution-strategies-and-variational-monte-carlo",
        "overcoming-barriers-to-scalability-in-variational-quantum-monte-carlo",
        "meta-variational-monte-carlo-arxiv",
        "neural-quantum-states-for-scientific-computing"
      ],
      "links": {
        "pdf": "TODO",
        "code": "TODO"
      },
      "bibtex_key": "zhao2023metavmc"
    },
    {
      "title": "Neural Quantum States for Scientific Computing",
      "authors": ["Tianchen Zhao"],
      "venue": "PhD Thesis, University of Michigan",
      "year": 2022,
      "citations": 0,
      "slug": "neural-quantum-states-for-scientific-computing",
      "abstract": "TODO: Fetch from University of Michigan Deep Blue repository. The thesis covers neural quantum states applied to computational chemistry and variational Monte Carlo methods, synthesizing the work from papers #4, #5, #6, #7, #10, #11.",
      "one_sentence_summary": "PhD thesis synthesizing neural quantum states for scientific computing, covering variational Monte Carlo for quantum chemistry, combinatorial optimization, and financial derivative pricing.",
      "key_contributions": [
        "Comprehensive treatment of neural quantum states for scientific computing",
        "Unified framework spanning quantum chemistry, combinatorial optimization, and PDEs",
        "Scalability analysis and GPU parallelization strategies for VMC"
      ],
      "citation_identity": "Cite for: neural quantum states for scientific computing, applications to computational chemistry, variational methods for finance",
      "search_phrases": ["neural quantum states scientific computing", "neural quantum states computational chemistry", "neural quantum states finance", "variational Monte Carlo chemistry finance", "scientific computing neural wavefunctions"],
      "topic_cluster": 3,
      "related_papers": [
        "scalable-neural-quantum-states-architecture-for-quantum-chemistry",
        "natural-evolution-strategies-and-variational-monte-carlo",
        "overcoming-barriers-to-scalability-in-variational-quantum-monte-carlo",
        "quantum-inspired-variational-algorithms-for-partial-differential-equations",
        "natural-evolution-strategies-and-quantum-approximate-optimization",
        "meta-variational-monte-carlo-arxiv",
        "meta-variational-quantum-monte-carlo"
      ],
      "links": {
        "pdf": "TODO: Check UMich Deep Blue repository",
        "code": "TODO"
      },
      "bibtex_key": "zhao2022thesis"
    }
  ]
}