arXiv subjects

Siqi Guo

Publications and source records attributed to Siqi Guo.

10 recordsLinked to original sources

arXiv2026

Adversarial Attacks for Good: A Survey of Proactive Protection across the Visual Content Lifecycle

Once visual content enters an AI pipeline, its owner often retains little technical control over how it is used. Legal and regulatory remedies can address misuse, but many technical interventions must be applied earlier, when content is released or accessed. This survey examines the protective paradigm that has grown around this intervention point, which we call \emph{adversarial attacks for good}. Perturbations and structured signals long studied as attacks on learned models are instead applied by data owners, creators, platforms, or auditors to disrupt unauthorized automation or support later accountability. Five research communities have arrived at this inversion largely independently, each addressing a different stage of a visual asset's lifecycle: privacy filters against unwanted recognition at sharing time, unlearnable examples against unauthorized training, generative safeguards against malicious editing or imitation, adversarial CAPTCHAs for access control against automated agents, and provenance mechanisms for post-circulation attribution. Although developed in separate venues with incompatible success criteria, many of these methods exploit persistent gaps between human perception, semantic interpretation, and machine inference, suggesting that the paradigm remains relevant as visual pipelines evolve toward multimodal models and autonomous agents. To make their claims comparable, we evaluate all five families along shared axes of transferability, adaptability, and deployment readiness. Across the lifecycle, we find that most protections are still validated mainly against static or weakly adaptive adversaries, while evidence beyond controlled benchmarks remains scarce. We close by consolidating cross-stage countermeasures and open problems for robust, composable, and deployable owner-side protection.

cs.CR↗

arXiv2026

Heterostructuring as Gateway to Electron Doping of Nickelate Superconductors

Despite enormous expenditures in the research field, the electron-doped side of nickelate superconductors remains uncharted territory. Substituting the trivalent rare-earth cations by a tetravalent one hitherto failed. Here, we demonstrate by first-principles calculations a disorder-free route to electron dope Ruddlesden-Popper nickelates. When intercalating wide-band-gap insulating layers such as La$X$O$_3$ ($X$=Al, Ga, Sc) into La$_2$NiO$_4$, the extra (LaO)$^+$ layers act as electron donors, releasing carriers into the Ni-3$d$ orbitals. This electron doping puts La$_2$NiO$_4$:La$_2$AlO$_4$ naturally in the optimal region for $d_{x^2-y^2}$-wave superconductivity with T$_c$ exceeding 50 K. The same concept also allows us to electron dope La$_3$Ni$_2$O$_7$, the superconductor in the limelight.

cond-mat.supr-con↗

arXiv2026

DRTriton: Large-Scale Synthetic Data Driven Reinforcement Learning for Triton Kernel Generation

Developing efficient CUDA kernels is a fundamental yet challenging task in the generative AI industry. Recent research leverages Large Language Models (LLMs) to automatically convert PyTorch reference implementations to CUDA kernels, significantly reducing engineering effort. State-of-the-art LLMs, such as GPT-5.2 and Claude-Sonnet-4.5, still struggle with this task. To address this challenge, we propose DRTriton, a scalable learning framework for training LLMs to convert PyTorch programs into highly optimized Triton kernels, which are then compiled to CUDA kernels at runtime. DRTriton consists of three key components: (i) a data synthetic algorithm CSP-DAG that guarantees full coverage and unbiased uniform sampling over the operator space with controlled difficulty; (ii) a curriculum RL framework with decoupled rewards that jointly optimizes conversion success rate and execution speed; and (iii) a test-time search algorithm that further improves the execution speed of the generated Triton kernels. With a warmup stage of SFT on limited PyTorch-Triton pairs curated using existing LLMs, DRTriton trained by RL on synthesized PyTorch programs generalizes effectively to real-world CUDA kernels that are challenging even for human experts. Experimental results show that DRTriton-7B achieves speedup over PyTorch on 92% of KernelBench Level 2 tasks, compared to 23% for GPT-5.2 and 19% for Claude-Sonnet-4.5.

cs.CL↗

arXiv2025

Discriminative Finetuning of Generative Large Language Models without Reward Models and Human Preference Data

Supervised fine-tuning (SFT) has become a crucial step for aligning pretrained large language models (LLMs) using supervised datasets of input-output pairs. However, despite being supervised, SFT is inherently limited by its generative training objective. To address its limitations, the existing common strategy is to follow SFT with a separate phase of preference optimization (PO), which relies on either human-labeled preference data or a strong reward model to guide the learning process. In this paper, we address the limitations of SFT by exploring one of the most successful techniques in conventional supervised learning: discriminative learning. We introduce Discriminative Fine-Tuning (DFT), an improved variant of SFT, which mitigates the burden of collecting human-labeled preference data or training strong reward models. Unlike SFT that employs a generative approach and overlooks negative data, DFT adopts a discriminative paradigm that increases the probability of positive answers while suppressing potentially negative ones, aiming for data prediction instead of token prediction. Our contributions include: (i) a discriminative probabilistic framework for fine-tuning LLMs by explicitly modeling the discriminative likelihood of an answer among all possible outputs given an input; (ii) efficient algorithms to optimize this discriminative likelihood; and (iii) extensive experiments demonstrating DFT's effectiveness, achieving performance better than SFT and comparable to if not better than SFT$\rightarrow$PO. The code can be found at https://github.com/Optimization-AI/DFT.

cs.CL↗

arXiv2025

CityBench: Evaluating the Capabilities of Large Language Models for Urban Tasks

As large language models (LLMs) continue to advance and gain widespread use, establishing systematic and reliable evaluation methodologies for LLMs and vision-language models (VLMs) has become essential to ensure their real-world effectiveness and reliability. There have been some early explorations about the usability of LLMs for limited urban tasks, but a systematic and scalable evaluation benchmark is still lacking. The challenge in constructing a systematic evaluation benchmark for urban research lies in the diversity of urban data, the complexity of application scenarios and the highly dynamic nature of the urban environment. In this paper, we design \textit{CityBench}, an interactive simulator based evaluation platform, as the first systematic benchmark for evaluating the capabilities of LLMs for diverse tasks in urban research. First, we build \textit{CityData} to integrate the diverse urban data and \textit{CitySimu} to simulate fine-grained urban dynamics. Based on \textit{CityData} and \textit{CitySimu}, we design 8 representative urban tasks in 2 categories of perception-understanding and decision-making as the \textit{CityBench}. With extensive results from 30 well-known LLMs and VLMs in 13 cities around the world, we find that advanced LLMs and VLMs can achieve competitive performance in diverse urban tasks requiring commonsense and semantic understanding abilities, e.g., understanding the human dynamics and semantic inference of urban images. Meanwhile, they fail to solve the challenging urban tasks requiring professional knowledge and high-level numerical abilities, e.g., geospatial prediction and traffic control task.

cs.AI↗

arXiv2025

CityGPT: Empowering Urban Spatial Cognition of Large Language Models

Large language models(LLMs), with their powerful language generation and reasoning capabilities, have already achieved notable success in many domains, e.g., math and code generation. However, they often fall short when tackling real-life geospatial tasks within urban environments. This limitation stems from a lack of physical world knowledge and relevant data during training. To address this gap, we propose \textit{CityGPT}, a systematic framework designed to enhance LLMs' understanding of urban space and improve their ability to solve the related urban tasks by integrating a city-scale `world model' into the model. Firstly, we construct a diverse instruction tuning dataset, \textit{CityInstruction}, for injecting urban knowledge into LLMs and effectively boosting their spatial reasoning capabilities. Using a combination of \textit{CityInstruction} and open source general instruction data, we introduce a novel and easy-to-use self-weighted fine-tuning method (\textit{SWFT}) to train various LLMs (including ChatGLM3-6B, Llama3-8B, and Qwen2.5-7B) to enhance their urban spatial capabilities without compromising, or even improving, their general abilities. Finally, to validate the effectiveness of our proposed framework, we develop a comprehensive text-based spatial benchmark \textit{CityEval} for evaluating the performance of LLMs across a wide range of urban scenarios and geospatial tasks. Extensive evaluation results demonstrate that smaller LLMs trained with \textit{CityInstruction} by \textit{SWFT} method can achieve performance that is competitive with, and in some cases superior to, proprietary LLMs when assessed using \textit{CityEval}.

cs.AI↗

arXiv2024

Bidirectional Optimization onto Thermoelectric Performance via Hydrostatic-Pressure in Chalcopyrite AgXTe2 (X=In, Ga)

Pressure tuning has emerged as a powerful strategy for manipulating the thermoelectric properties of materials by inducing structural and electronic modifications. Herein, we systematically investigate the transport properties and thermoelectric performance concerning lattice distortions induced by hydrostatic pressure in Ag-based chalcopyrite AgXTe2 (X=In, Ga). The findings reveal that the lattice distortion in AgXTe2 exhibits distinct behaviors under lattice compression, diverging from trends observed at ambient pressure. Importantly, the hydrostatic pressure breaks the phenomenally negative correlation between thermal conductivity and lattice distortion. Pressure-induced softening of low-frequency acoustic phonons broadens the low-energy phonon spectrum, enhancing interactions between acoustic and optical phonons. Such broadening substantially increases the number of available three-phonon scattering channels, resulting in a marked reduction in thermal conductivity. Meanwhile, we establish a macroscopic connection between metavalent bonding and anharmonicity, providing an indirect explanation for lattice anharmonicity through pressure-driven transferred charge. Additionally, the applied pressure achieves a notable net increase in the power factor despite the strong coupling of electrical transport parameters, which underscores the potential for bidirectional optimization of transport properties in AgXTe2. As a result, the maximum ZT value of AgInTe2 is nearly doubled, demonstrating that pressure modulation is a powerful strategy for enhancing thermoelectric performance. Our work not only establishes the link between pressure, lattice dynamics, and thermoelectric properties within chalcopyrite AgXTe2, but also inspires the exploration of pressure-related optimization strategies for conventional thermoelectric materials.

cond-mat.mtrl-sci↗

arXiv2024

To Cool or not to Cool? Temperature Network Meets Large Foundation Models via DRO

The temperature parameter plays a profound role during training and/or inference with large foundation models (LFMs) such as large language models (LLMs) and CLIP models. Particularly, it adjusts the logits in the softmax function in LLMs, which is crucial for next token generation, and it scales the similarities in the contrastive loss for training CLIP models. A significant question remains: Is it viable to learn a neural network to predict a personalized temperature of any input data for enhancing LFMs"? In this paper, we present a principled framework for learning a small yet generalizable temperature prediction network (TempNet) to improve LFMs. Our solution is composed of a novel learning framework with a robust loss underpinned by constrained distributionally robust optimization (DRO), and a properly designed TempNet with theoretical inspiration. TempNet can be trained together with a large foundation model from scratch or learned separately given a pretrained foundation model. It is not only useful for predicting personalized temperature to promote the training of LFMs but also generalizable and transferable to new tasks. Our experiments on LLMs and CLIP models demonstrate that TempNet greatly improves the performance of existing solutions or models, e.g. Table 1. The code to reproduce the experimental results in this paper can be found at https://github.com/zhqiu/TempNet.

cs.LG↗

arXiv2024

Ultra-low glassy thermal conductivity and controllable, promising thermoelectric properties in crystalline o-CsCu5S3

We thoroughly investigate the microscopic mechanisms of the thermal transport in orthorhombic \textit{o}-CsCu$_5$S$_3$ by integrating the first-principles-based self-consistent phonon calculations (SCP) with the linearized Wigner transport equation (LWTE). Our methodology takes into account contributions to phonon energy shifts and phonon scattering rates from both three- and four-phonon processes. Additionally, it incorporates the off-diagonal terms of heat flux operators to calculate the total thermal conductivity. The predicted $κ_\mathrm{L}$ with an extremely weak temperature dependence following $\sim T^{-0.33}$, in good agreement with experimental values along with the parallel to the Bridgman growth direction. Such nonstandard temperature dependence of $κ_\mathrm{L}$ can be traced back to the dual particlelike-wavelike behavior exhibited by thermal phonons. Specifically, the coexistence of the stochastic oscillation of Cs atoms and metavalent bonding among interlayer Cu-S atoms limits the particle-like phonon propagation and enhances the wave-like tunneling of phonons. Simultaneously, the electrical transport properties are determined by employing a precise momentum relaxation-time approximation (MRTA) within the framework of the linearized Boltzmann transport equation (LBTE). By properly adjusting the carrier concentration, excellent thermoelectric performance is achieved, with a maximum thermoelectric conversion efficiency of 18.4$\%$ observed at 800 K in \textit{p}-type \textit{o}-CsCu$_5$S$_3$.} Our work not only elucidates the anomalous thermal transport behavior in the copper-based chalcogenide \textit{o}-CsCu$_5$S$_3$ but also provides insights for manipulating its thermal and electronic properties for potential thermoelectric applications.

cond-mat.mtrl-sci↗

arXiv2022

Tactical fleet assignment and crew pairing problem with crew flight time allocation

Aircraft and crew are two major resources that ensure the smooth operations of airlines. However, with the anticipated growth in the aviation industry, the crew resource is predicted to be insufficient worldwide and has been one of the bottlenecks in fast-developing airlines. In addition, the mismatch between aircraft and crew has influenced the airlines' operation, and limited the full usage of all resources. To resolve this problem, we propose a tactical fleet assignment and crew pairing problem with crew flight time allocation (TFACPP). The basic integrated model is reformulated by the Benders decomposition, where the Benders master problem (BMP) poses the most significant computational barrier. To efficiently solve the BMP, we propose a column generation algorithm. The TFACPP can provide superior solutions compared to the equal allocation of crew resources. In addition, we provide a quantitative method for evaluating the scarcities of crew and aircraft resources and the matching degree between crew and aircraft based on the shadow prices of the proposed model. These information can provide rich managerial insights regarding the acquisition, replacement, and transition of crew and aircraft.

math.OC↗