arXiv subjects

Xi Liu

Publications and source records attributed to Xi Liu.

At least 55 records · Page 3Linked to original sources

arXiv2026

A diffuse-interface model for N-phase flows with liquid-solid phase change

In this work, we first propose a diffuse interface model for simulating N phase flows with solid liquid phase change. In this model, a phase field approach is adopted to capture multiphase fluid interfaces, and an enthalpy based formulation is used to describe the phase change. The volume changes resulting from density differences during phase change are incorporated by introducing a source term into the continuity equation. The method also satisfies the reduction consistent property, allowing it to rigorously degenerate to both the conservative phase field method for N phase flows and the classical enthalpy method for solid liquid phase change. Then a coupled lattice Boltzmann (LB) method is developed to solve this diffuse interface model. Some numerical tests, including film freezing, single droplet freezing, and compound droplet freezing are performed, and the results are in good agreement with the analytical solutions and data reported in the previous works. Furthermore, the proposed method is applied to study freezing dynamics of complex systems with insoluble impurities, capturing the interaction between the advancing freezing front and embedded impurities. It is found that the proposed diffuse interface method is accurate and efficient for studying N phase systems with phase change.

physics.flu-dyn↗

arXiv2026

Renormalization-Inspired Effective Field Neural Networks for Scalable Modeling of Classical and Quantum Many-Body Systems

We introduce Effective Field Neural Networks (EFNNs), a new architecture based on continued functions -- mathematical tools used in renormalization to handle divergent perturbative series. Our key insight is that neural networks can implement these continued functions directly, providing a principled approach to many-body interactions. Testing on three systems (a classical 3-spin infinite- range model, a continuous classical Heisenberg spin system, and a quantum double exchange model), we find that EFNN outperforms standard deep networks, ResNet, and DenseNet. Most striking is EFNN's generalization: trained on $10 \times 10$ lattices, it accurately predicts behavior on systems up to $40\times 40$ with no additional training -- and the accuracy improves with system size, with a computational time speed-up of $10^{3}$ compared to ED for $40\times 40$ lattice. This demonstrates that EFNN captures the underlying physics rather than merely fitting data, making it valuable beyond many-body problems to any field where renormalization ideas apply.

physics.comp-ph↗

arXiv2026

Cross-Domain Policy Optimization via Bellman Consistency and Hybrid Critics

Cross-domain reinforcement learning (CDRL) is meant to improve the data efficiency of RL by leveraging the data samples collected from a source domain to facilitate the learning in a similar target domain. Despite its potential, cross-domain transfer in RL is known to have two fundamental and intertwined challenges: (i) The source and target domains can have distinct state space or action space, and this makes direct transfer infeasible and thereby requires more sophisticated inter-domain mappings; (ii) The transferability of a source-domain model in RL is not easily identifiable a priori, and hence CDRL can be prone to negative effect during transfer. In this paper, we propose to jointly tackle these two challenges through the lens of \textit{cross-domain Bellman consistency} and \textit{hybrid critic}. Specifically, we first introduce the notion of cross-domain Bellman consistency as a way to measure transferability of a source-domain model. Then, we propose $Q$Avatar, which combines the Q functions from both the source and target domains with an adaptive hyperparameter-free weight function. Through this design, we characterize the convergence behavior of $Q$Avatar and show that $Q$Avatar achieves reliable transfer in the sense that it effectively leverages a source-domain Q function for knowledge transfer to the target domain. Through experiments, we demonstrate that $Q$Avatar achieves favorable transferability across various RL benchmark tasks, including locomotion and robot arm manipulation. Our code is available at https://rl-bandits-lab.github.io/Cross-Domain-RL/.

cs.LG↗

arXiv2026

Controlling Your Image via Simplified Vector Graphics

Recent advances in image generation have achieved remarkable visual quality, while a fundamental challenge remains: Can image generation be controlled at the element level, enabling intuitive modifications such as adjusting shapes, altering colors, or adding and removing objects? In this work, we address this challenge by introducing layer-wise controllable generation through simplified vector graphics (VGs). Our approach first efficiently parses images into hierarchical VG representations that are semantic-aligned and structurally coherent. Building on this representation, we design a novel image synthesis framework guided by VGs, allowing users to freely modify elements and seamlessly translate these edits into photorealistic outputs. By leveraging the structural and semantic features of VGs in conjunction with noise prediction, our method provides precise control over geometry, color, and object semantics. Extensive experiments demonstrate the effectiveness of our approach in diverse applications, including image editing, object-level manipulation, and fine-grained content creation, establishing a new paradigm for controllable image generation. Project page: https://guolanqing.github.io/Vec2Pix/

cs.CV↗

arXiv2025

Meta Lattice: Model Space Redesign for Cost-Effective Industry-Scale Ads Recommendations

The rapidly evolving landscape of products, surfaces, policies, and regulations poses significant challenges for deploying state-of-the-art recommendation models at industry scale, primarily due to data fragmentation across domains and escalating infrastructure costs that hinder sustained quality improvements. To address this challenge, we propose Lattice, a recommendation framework centered around model space redesign that extends Multi-Domain, Multi-Objective (MDMO) learning beyond models and learning objectives. Lattice addresses these challenges through a comprehensive model space redesign that combines cross-domain knowledge sharing, data consolidation, model unification, distillation, and system optimizations to achieve significant improvements in both quality and cost-efficiency. Our deployment of Lattice at Meta has resulted in 10% revenue-driving top-line metrics gain, 11.5% user satisfaction improvement, 6% boost in conversion rate, with 20% capacity saving.

cs.IR↗

arXiv2025

Bezier Splatting for Fast and Differentiable Vector Graphics Rendering

Differentiable vector graphics (VGs) are widely used in image vectorization and vector synthesis, while existing representations are costly to optimize and struggle to achieve high-quality rendering results for high-resolution images. This work introduces a new differentiable VG representation, dubbed Bézier Splatting, that enables fast yet high-fidelity VG rasterization. Bézier Splatting samples 2D Gaussians along Bézier curves, which naturally provide positional gradients at object boundaries. Thanks to the efficient splatting-based differentiable rasterizer, Bézier Splatting achieves 30x and 150x faster per forward and backward rasterization step for open curves compared to DiffVG. Additionally, we introduce an adaptive pruning and densification strategy that dynamically adjusts the spatial distribution of curves to escape local minima, further improving VG quality. Furthermore, our new VG representation supports conversion to standard XML-based SVG format, enhancing interoperability with existing VG tools and pipelines. Experimental results show that Bézier Splatting significantly outperforms existing methods with better visual fidelity and significant optimization speedup.

cs.GR↗

arXiv2025

Diffuse-interface modeling and simulation of the freezing of binary fluids with the Marangoni effect

This paper proposes a diffuse-interface model for simulating gas-liquid-solid multiphase flows involving solid-liquid phase change, solute transport, and the Marangoni effect. In this model, a phase-field method is employed to capture the evolution of fluid-fluid interfaces, while an enthalpy-based approach is used to describe the temperature field and implicitly track the solid-liquid interface. Solute transport is modeled using a constrained scalar-transport model combined with a pseudo-potential concentration approach. The proposed diffuse-interface model satisfies the reduction consistency, and can degenerate to the conservative phase-field method for incompressible two-phase flow and the classical enthalpy method for binary material solidification in an appropriate way. Furthermore, the model not only can preserve the mass conservation, but also can capture the volume change induced by phase change. To solve the diffuse-interface model, a lattice Boltzmann (LB) method is then developed, and the numerical tests demonstrate that the method has a good performance in the study of the freezing process coupled with Marangoni flow, phase-change-induced volume change, and solute transport. Finally, the model is applied to investigate the freezing dynamics of a system containing an insoluble impurity, revealing the complex interaction between the advancing freezing front and the impurity. It is found that the numerical results are in good agreement with experimental data.

physics.flu-dyn↗

arXiv2025

InterFormer: Effective Heterogeneous Interaction Learning for Click-Through Rate Prediction

Click-through rate (CTR) prediction, which predicts the probability of a user clicking an ad, is a fundamental task in recommender systems. The emergence of heterogeneous information, such as user profile and behavior sequences, depicts user interests from different aspects. A mutually beneficial integration of heterogeneous information is the cornerstone towards the success of CTR prediction. However, most of the existing methods suffer from two fundamental limitations, including (1) insufficient inter-mode interaction due to the unidirectional information flow between modes, and (2) aggressive information aggregation caused by early summarization, resulting in excessive information loss. To address the above limitations, we propose a novel module named InterFormer to learn heterogeneous information interaction in an interleaving style. To achieve better interaction learning, InterFormer enables bidirectional information flow for mutually beneficial learning across different modes. To avoid aggressive information aggregation, we retain complete information in each data mode and use a separate bridging arch for effective information selection and summarization. Our proposed InterFormer achieves state-of-the-art performance on three public datasets and a large-scale industrial dataset.

cs.IR↗

arXiv2025

Electrohydrodynamics of a pair of leaky dielectric droplets on the solid substrate: A lattice Boltzmann study

In this work, the electrohydrodynamics of a pair of leaky dielectric droplets on a solid substrate is investigated by the phase-field-based lattice Boltzmann method. Different from a pair of suspended droplets that may coalesce or separate, two leaky dielectric droplets on the substrate exhibit more complex modes due to the effects of wettability and electric force. The results show that when a horizontal electric field is applied, five different modes with electrostatic attractive force are observed, including attraction without coalescence, attraction with coalescence, coalescence with bubble entrapment, coalescence followed by suspension, and suspension followed by coalescence. Particularly, if the droplets are in a hydrophilic state, the coalescence mode is usually observed, while for droplets in a neutral or hydrophobic state, the permittivity ratio has an important effect on the droplet modes. Additionally, during the coalescence process, two droplets in a hydrophobic state not only capture bubbles, but may also exhibit suspension at a large permittivity ratio or contact angle. On the other hand, when a vertical electric field is applied, there are three different modes with repulsive electrostatic force, including non-coalescence, coalescence, and suspension followed by repulsion. Specially, a small permittivity ratio or a large contact angle can suppress the horizontal deformation of droplets, preventing their coalescence. Moreover, under superhydrophobic conditions, both horizontal and vertical electric fields suspend the droplets. However, the vertical electric field induces repulsion between the suspended droplets, driving them apart, whereas the horizontal electric field promotes their coalescence.

physics.flu-dyn↗

arXiv2025

External Large Foundation Model: How to Efficiently Serve Trillions of Parameters for Online Ads Recommendation

Ads recommendation is a prominent service of online advertising systems and has been actively studied. Recent studies indicate that scaling-up and advanced design of the recommendation model can bring significant performance improvement. However, with a larger model scale, such prior studies have a significantly increasing gap from industry as they often neglect two fundamental challenges in industrial-scale applications. First, training and inference budgets are restricted for the model to be served, exceeding which may incur latency and impair user experience. Second, large-volume data arrive in a streaming mode with data distributions dynamically shifting, as new users/ads join and existing users/ads leave the system. We propose the External Large Foundation Model (ExFM) framework to address the overlooked challenges. Specifically, we develop external distillation and a data augmentation system (DAS) to control the computational cost of training/inference while maintaining high performance. We design the teacher in a way like a foundation model (FM) that can serve multiple students as vertical models (VMs) to amortize its building cost. We propose Auxiliary Head and Student Adapter to mitigate the data distribution gap between FM and VMs caused by the streaming data issue. Comprehensive experiments on internal industrial-scale applications and public datasets demonstrate significant performance gain by ExFM.

cs.IR↗

arXiv2025

ARIG: Autoregressive Interactive Head Generation for Real-time Conversations

Face-to-face communication, as a common human activity, motivates the research on interactive head generation. A virtual agent can generate motion responses with both listening and speaking capabilities based on the audio or motion signals of the other user and itself. However, previous clip-wise generation paradigm or explicit listener/speaker generator-switching methods have limitations in future signal acquisition, contextual behavioral understanding, and switching smoothness, making it challenging to be real-time and realistic. In this paper, we propose an autoregressive (AR) based frame-wise framework called ARIG to realize the real-time generation with better interaction realism. To achieve real-time generation, we model motion prediction as a non-vector-quantized AR process. Unlike discrete codebook-index prediction, we represent motion distribution using diffusion procedure, achieving more accurate predictions in continuous space. To improve interaction realism, we emphasize interactive behavior understanding (IBU) and detailed conversational state understanding (CSU). In IBU, based on dual-track dual-modal signals, we summarize short-range behaviors through bidirectional-integrated learning and perform contextual understanding over long ranges. In CSU, we use voice activity signals and context features of IBU to understand the various states (interruption, feedback, pause, etc.) that exist in actual conversations. These serve as conditions for the final progressive motion prediction. Extensive experiments have verified the effectiveness of our model.

cs.CV↗

arXiv2025

DistDNAS: Search Efficient Feature Interactions within 2 Hours

Search efficiency and serving efficiency are two major axes in building feature interactions and expediting the model development process in recommender systems. On large-scale benchmarks, searching for the optimal feature interaction design requires extensive cost due to the sequential workflow on the large volume of data. In addition, fusing interactions of various sources, orders, and mathematical operations introduces potential conflicts and additional redundancy toward recommender models, leading to sub-optimal trade-offs in performance and serving cost. In this paper, we present DistDNAS as a neat solution to brew swift and efficient feature interaction design. DistDNAS proposes a supernet to incorporate interaction modules of varying orders and types as a search space. To optimize search efficiency, DistDNAS distributes the search and aggregates the choice of optimal interaction modules on varying data dates, achieving over 25x speed-up and reducing search cost from 2 days to 2 hours. To optimize serving efficiency, DistDNAS introduces a differentiable cost-aware loss to penalize the selection of redundant interaction modules, enhancing the efficiency of discovered feature interactions in serving. We extensively evaluate the best models crafted by DistDNAS on a 1TB Criteo Terabyte dataset. Experimental evaluations demonstrate 0.001 AUC improvement and 60% FLOPs saving over current state-of-the-art CTR models.

cs.IR↗

arXiv2025

APOLLO: SGD-like Memory, AdamW-level Performance

Large language models (LLMs) are notoriously memory-intensive during training, particularly with the popular AdamW optimizer. This memory burden necessitates using more or higher-end GPUs or reducing batch sizes, limiting training scalability and throughput. To address this, various memory-efficient optimizers have been proposed to reduce optimizer memory usage. However, they face critical challenges: (i) reliance on costly SVD operations; (ii) significant performance trade-offs compared to AdamW; and (iii) still substantial optimizer memory overhead to maintain competitive performance. In this work, we identify that AdamW's learning rate adaptation rule can be effectively coarsened as a structured learning rate update. Based on this insight, we propose Approximated Gradient Scaling for Memory-Efficient LLM Optimization (APOLLO), which approximates learning rate scaling using an auxiliary low-rank optimizer state based on pure random projection. This structured learning rate update rule makes APOLLO highly tolerant to further memory reductions while delivering comparable pre-training performance. Even its rank-1 variant, APOLLO-Mini, achieves superior pre-training performance compared to AdamW with SGD-level memory costs. Extensive experiments demonstrate that the APOLLO series performs on-par with or better than AdamW, while achieving greater memory savings by nearly eliminating the optimization states of AdamW. These savings provide significant system-level benefits: (1) Enhanced Throughput: 3x throughput on an 8xA100-80GB setup compared to AdamW by supporting 4x larger batch sizes. (2) Improved Model Scalability: Pre-training LLaMA-13B with naive DDP on A100-80GB GPUs without system-level optimizations. (3) Low-End GPU Friendly Pre-training: Pre-training LLaMA-7B on a single GPU using less than 12 GB of memory with weight quantization.

cs.LG↗

arXiv2025

Enhancing Offline Model-Based RL via Active Model Selection: A Bayesian Optimization Perspective

Offline model-based reinforcement learning (MBRL) serves as a competitive framework that can learn well-performing policies solely from pre-collected data with the help of learned dynamics models. To fully unleash the power of offline MBRL, model selection plays a pivotal role in determining the dynamics model utilized for downstream policy learning. However, offline MBRL conventionally relies on validation or off-policy evaluation, which are rather inaccurate due to the inherent distribution shift in offline RL. To tackle this, we propose BOMS, an active model selection framework that enhances model selection in offline MBRL with only a small online interaction budget, through the lens of Bayesian optimization (BO). Specifically, we recast model selection as BO and enable probabilistic inference in BOMS by proposing a novel model-induced kernel, which is theoretically grounded and computationally efficient. Through extensive experiments, we show that BOMS improves over the baseline methods with a small amount of online interaction comparable to only $1\%$-$2.5\%$ of offline training data on various RL tasks.

cs.LG↗

arXiv2025

Latent Radiance Fields with 3D-aware 2D Representations

Latent 3D reconstruction has shown great promise in empowering 3D semantic understanding and 3D generation by distilling 2D features into the 3D space. However, existing approaches struggle with the domain gap between 2D feature space and 3D representations, resulting in degraded rendering performance. To address this challenge, we propose a novel framework that integrates 3D awareness into the 2D latent space. The framework consists of three stages: (1) a correspondence-aware autoencoding method that enhances the 3D consistency of 2D latent representations, (2) a latent radiance field (LRF) that lifts these 3D-aware 2D representations into 3D space, and (3) a VAE-Radiance Field (VAE-RF) alignment strategy that improves image decoding from the rendered 2D representations. Extensive experiments demonstrate that our method outperforms the state-of-the-art latent 3D reconstruction approaches in terms of synthesis performance and cross-dataset generalizability across diverse indoor and outdoor scenes. To our knowledge, this is the first work showing the radiance field representations constructed from 2D latent representations can yield photorealistic 3D reconstruction performance.

cs.CV↗

arXiv2025

The GeV $γ$-ray emission from the composite SNR CTB 87

We report the GeV $γ$-ray emission around the composite supernova remnant (SNR) CTB 87 with more than 16 yrs PASS 8 data recorded by the Fermi Large Area Telescope. Two separate point sources with the different GeV spectra are identified in this region: one has a soft $γ$-ray spectrum, likely due to interactions between the SNR shock and molecular clouds (MCs); and another source with a hard GeV $γ$-ray spectrum aligns with the TeV spectrum of VER J2016+371, suggesting it as the GeV counterpart. Considering the observations of CTB 87 in the radio and X-ray bands, VER J2016+371 is proposed to originate from the pulsar wind nebula (PWN) associated with PSR J2016+3711. A leptonic model with a broken power-law electron distribution could explain the multi-wavelength data of VER J2016+371, with fitted parameters matching typical $γ$-ray PWNe. Deeper searching for the SNR shock of CTB 87 in other bands and the future TeV observations by LHAASO and CTA are crucial to reveal the nature of CTB 87.

astro-ph.HE↗

arXiv2024

Detection of the extended $γ$-ray emission from the new supernova remnant G321.3-3.9 with Fermi-LAT

With the 15 yrs of Pass 8 data recorded by the {\em Fermi} Large Area Telescope, we report the detection of an extended gigaelectronvolt emission component with a 68\% containment radius of $0^{\circ}\!.85$, which is spatially associated with the newly identified supernova remnant (SNR) G321.3-3.9. The $γ$-ray spectrum is best described by a log-parabola model in the energy range of 100 MeV - 1 TeV, which shows a significant spectral curvature at $\sim$ 1 GeV. Either a leptonic or a hadronic model could explain the multi-wavelength data of G321.3-3.9, while the leptonic model predicts a too low strength of magnetic field. Also considering the flat radio spectrum of G321.3-3.9 and the $γ$-ray upper limit in the low energy band, the hadronic model is favored. The spatial coincidence between the $γ$-ray morphology and the diffuse thermal X-ray emission of G321.3-3.9 and the curved gigaelectronvolt $γ$-ray spectrum of it make G321.3-3.9 to be similar to the typical middle-aged SNRs interacting with molecular clouds. Such characteristics provide another evidence of the potential hadronic origin for its $γ$-ray emission. While there is no molecular cloud detected around G321.3-3.9, which challenges the hadronic model.

astro-ph.HE↗

arXiv2024

3DGS-Enhancer: Enhancing Unbounded 3D Gaussian Splatting with View-consistent 2D Diffusion Priors

Novel-view synthesis aims to generate novel views of a scene from multiple input images or videos, and recent advancements like 3D Gaussian splatting (3DGS) have achieved notable success in producing photorealistic renderings with efficient pipelines. However, generating high-quality novel views under challenging settings, such as sparse input views, remains difficult due to insufficient information in under-sampled areas, often resulting in noticeable artifacts. This paper presents 3DGS-Enhancer, a novel pipeline for enhancing the representation quality of 3DGS representations. We leverage 2D video diffusion priors to address the challenging 3D view consistency problem, reformulating it as achieving temporal consistency within a video generation process. 3DGS-Enhancer restores view-consistent latent features of rendered novel views and integrates them with the input views through a spatial-temporal decoder. The enhanced views are then used to fine-tune the initial 3DGS model, significantly improving its rendering performance. Extensive experiments on large-scale datasets of unbounded scenes demonstrate that 3DGS-Enhancer yields superior reconstruction performance and high-fidelity rendering results compared to state-of-the-art methods. The project webpage is https://xiliu8006.github.io/3DGS-Enhancer-project .

cs.CV↗