Xi Chen

Shuqi Lu, Lin Yao, Xi Chen et al. · microsoft-research

Structure-based drug design, i.e., finding molecules with high affinities to the target protein pocket, is one of the most critical tasks in drug discovery. Traditional solutions, like virtual screening, require exhaustively searching on a large molecular database, which are inefficient and cannot return novel molecules beyond the database. The pocket-based 3D molecular generation model, i.e., directly generating a molecule with a 3D structure and binding position in the pocket, is a new promising way to address this issue. Herein, we propose VD-Gen, a novel pocket-based 3D molecular generation pipeline. VD-Gen consists of several carefully designed stages to generate fine-grained 3D molecules with binding positions in the pocket cavity end-to-end. Rather than directly generating or sampling atoms with 3D positions in the pocket like in early attempts, in VD-Gen, we first randomly initialize many virtual particles in the pocket; then iteratively move these virtual particles, making the distribution of virtual particles approximate the distribution of molecular atoms. After virtual particles are stabilized in 3D space, we extract a 3D molecule from them. Finally, we further refine atoms in the extracted molecule by iterative movement again, to get a high-quality 3D molecule, and predict a confidence score for it. Extensive experiment results on pocket-based molecular generation demonstrate that VD-Gen can generate novel 3D molecules to fill the target pocket cavity with high binding affinities, significantly outperforming previous baselines.

Xi Chen, Anindya De, Yizhi Huang et al.

Several recent works [DHLNSY25, CPPS25a, CPPS25b] have studied a model of property testing of Boolean functions under a \emph{relative-error} criterion. In this model, the distance from a target function $f: \{0,1\}^n \to \{0,1\}$ that is being tested to a function $g$ is defined relative to the number of inputs $x$ for which $f(x)=1$; moreover, testing algorithms in this model have access both to a black-box oracle for $f$ and to independent uniform satisfying assignments of $f$. The motivation for this model is that it provides a natural framework for testing \emph{sparse} Boolean functions that have few satisfying assignments, analogous to well-studied models for property testing of sparse graphs. The main result of this paper is a lower bound for testing \emph{halfspaces} (i.e., linear threshold functions) in the relative error model: we show that $\tildeΩ(\log n)$ oracle calls are required for any relative-error halfspace testing algorithm over the Boolean hypercube $\{0,1\}^n$. This stands in sharp contrast both with the constant-query testability (independent of $n$) of halfspaces in the standard model [MORS10], and with the positive results for relative-error testing of many other classes given in [DHLNSY25, CPPS25a, CPPS25b]. Our lower bound for halfspaces gives the first example of a well-studied class of functions for which relative-error testing is provably more difficult than standard-model testing.

Lan Chen, Xi Chen, Shiyu Wu et al.

As a phenomenal large language model, ChatGPT has achieved unparalleled success in various real-world tasks and increasingly plays an important role in our daily lives and work. However, extensive concerns are also raised about the potential ethical issues, especially about whether ChatGPT-like artificial general intelligence (AGI) will replace human jobs. To this end, in this paper, we introduce a preliminary data-driven study on the future of ChatGPT-enabled labor market from the view of Human-AI Symbiosis instead of Human-AI Confrontation. To be specific, we first conduct an in-depth analysis of large-scale job posting data in BOSS Zhipin, the largest online recruitment platform in China. The results indicate that about 28% of occupations in the current labor market require ChatGPT-related skills. Furthermore, based on a large-scale occupation-centered knowledge graph, we develop a semantic information enhanced collaborative filtering algorithm to predict the future occupation-skill relations in the labor market. As a result, we find that additional 45% occupations in the future will require ChatGPT-related skills. In particular, industries related to technology, products, and operations are expected to have higher proficiency requirements for ChatGPT-related skills, while the manufacturing, services, education, and health science related industries will have lower requirements for ChatGPT-related skills.

Xi Chen, Anindya De, Yizhi Huang et al.

The problem of super-resolution, roughly speaking, is to reconstruct an unknown signal to high accuracy, given (potentially noisy) information about its low-degree Fourier coefficients. Prior results on super-resolution have imposed strong modeling assumptions on the signal, typically requiring that it is a linear combination of spatially separated point sources. In this work we analyze a very general version of the super-resolution problem by considering completely general non-negative signals (equivalently, distributions) over the $d$-dimensional torus $[0,1)^d$; we do not assume any spatial separation between point sources, or even that the distribution is a finite linear combination of point sources. The question naturally arises: what can be said about super-resolution in such a general setting? - As a warm-up, we first give a set of results for reconstructing distributions under the Wasserstein distance. We establish essentially matching upper and lower bounds on the cutoff frequency $T$ and the magnitude $κ$ of the noise for which accurate reconstruction is possible: we show that for $d$-dimensional distributions, estimates of $\approx \exp(d)$ many Fourier coefficients are both necessary and sufficient for accurate Wasserstein reconstruction. - As our main result, we define a new notion of "heavy hitter" reconstruction for distributions, which essentially amounts to achieving high-accuracy reconstruction of all "sufficiently dense" regions of the distribution. We give essentially matching upper and lower bounds on the cutoff frequency $T$ and the magnitude $κ$ of the noise for which accurate reconstruction is possible under this notion. Our results show that (in sharp contrast with Wasserstein reconstruction) accurate estimates of only $\approx \exp(\sqrt{d})$ many Fourier coefficients are both necessary and sufficient for heavy hitter reconstruction.

Siwei Zhang, Yun Xiong, Yao Zhang et al.

Temporal Graph Networks (TGNs) have shown remarkable performance in learning representation for continuous-time dynamic graphs. However, real-world dynamic graphs typically contain diverse and intricate noise. Noise can significantly degrade the quality of representation generation, impeding the effectiveness of TGNs in downstream tasks. Though structure learning is widely applied to mitigate noise in static graphs, its adaptation to dynamic graph settings poses two significant challenges. i) Noise dynamics. Existing structure learning methods are ill-equipped to address the temporal aspect of noise, hampering their effectiveness in such dynamic and ever-changing noise patterns. ii) More severe noise. Noise may be introduced along with multiple interactions between two nodes, leading to the re-pollution of these nodes and consequently causing more severe noise compared to static graphs. In this paper, we present RDGSL, a representation learning method in continuous-time dynamic graphs. Meanwhile, we propose dynamic graph structure learning, a novel supervisory signal that empowers RDGSL with the ability to effectively combat noise in dynamic graphs. To address the noise dynamics issue, we introduce the Dynamic Graph Filter, where we innovatively propose a dynamic noise function that dynamically captures both current and historical noise, enabling us to assess the temporal aspect of noise and generate a denoised graph. We further propose the Temporal Embedding Learner to tackle the challenge of more severe noise, which utilizes an attention mechanism to selectively turn a blind eye to noisy edges and hence focus on normal edges, enhancing the expressiveness for representation generation that remains resilient to noise. Our method demonstrates robustness towards downstream tasks, resulting in up to 5.1% absolute AUC improvement in evolving classification versus the second-best baseline.

Xi Chen, Wei Hu, Jingru Yu et al.

Urban growth sometimes leads to rigid infrastructure that struggles to adapt to changing demand. This paper introduces a novel approach, aiming to enable cities to evolve and respond more effectively to such dynamic demand. It identifies the limitations arising from the complexity and inflexibility of existing urban systems. A framework is presented for enhancing the city's adaptability perception through advanced sensing technologies, conducting parallel simulation via graph-based techniques, and facilitating autonomous decision-making across domains through decentralized and autonomous organization and operation. Notably, a symbiotic mechanism is employed to implement these technologies practically, thereby making urban management more agile and responsive. In the case study, we explore how this approach can optimize traffic flow by adjusting lane allocations. This case not only enhances traffic efficiency but also reduces emissions. The proposed evolutionary city offers a new perspective on sustainable urban development, highliting the importance of integrated intelligence within urban systems.

Xiao Chen, Zhongjing Du, Jiazhen Huang et al.

Test-Time Adaptation (TTA) enhances model robustness to out-of-distribution (OOD) data by updating the model online during inference, yet existing methods lack theoretical insights into the fundamental causes of performance degradation under domain shifts. Recently, Neural Collapse (NC) has been proposed as an emergent geometric property of deep neural networks (DNNs), providing valuable insights for TTA. In this work, we extend NC to the sample-wise level and discover a novel phenomenon termed Sample-wise Alignment Collapse (NC3+), demonstrating that a sample's feature embedding, obtained by a trained model, aligns closely with the corresponding classifier weight. Building on NC3+, we identify that the performance degradation stems from sample-wise misalignment in adaptation which exacerbates under larger distribution shifts. This indicates the necessity of realigning the feature embeddings with their corresponding classifier weights. However, the misalignment makes pseudo-labels unreliable under domain shifts. To address this challenge, we propose NCTTA, a novel feature-classifier alignment method with hybrid targets to mitigate the impact of unreliable pseudo-labels, which blends geometric proximity with predictive confidence. Extensive experiments demonstrate the effectiveness of NCTTA in enhancing robustness to domain shifts. For example, NCTTA outperforms Tent by 14.52% on ImageNet-C.

Xiao Fan, Jingyan Jiang, Zhaoru Chen et al.

Test-Time adaptation (TTA) has proven effective in mitigating performance drops under single-domain distribution shifts by updating model parameters during inference. However, real-world deployments often involve mixed distribution shifts, where test samples are affected by diverse and potentially conflicting domain factors, posing significant challenges even for SOTA TTA methods. A key limitation in existing approaches is their reliance on a unified adaptation path, which fails to account for the fact that optimal gradient directions can vary significantly across different domains. Moreover, current benchmarks focus only on synthetic or homogeneous shifts, failing to capture the complexity of real-world heterogeneous mixed distribution shifts. To address this, we propose MoETTA, a novel entropy-based TTA framework that integrates the Mixture-of-Experts (MoE) architecture. Rather than enforcing a single parameter update rule for all test samples, MoETTA introduces a set of structurally decoupled experts, enabling adaptation along diverse gradient directions. This design allows the model to better accommodate heterogeneous shifts through flexible and disentangled parameter updates. To simulate realistic deployment conditions, we introduce two new benchmarks: potpourri and potpourri+. While classical settings focus solely on synthetic corruptions, potpourri encompasses a broader range of domain shifts--including natural, artistic, and adversarial distortions--capturing more realistic deployment challenges. Additionally, potpourri+ further includes source-domain samples to evaluate robustness against catastrophic forgetting. Extensive experiments across three mixed distribution shifts settings show that MoETTA consistently outperforms strong baselines, establishing SOTA performance and highlighting the benefit of modeling multiple adaptation directions via expert-level diversity.

Xi Chen, Qin Li, Haibin Cai et al.

Federated Learning (FL) empowers multiple clients to collaboratively train machine learning models without sharing local data, making it highly applicable in heterogeneous Internet of Things (IoT) environments. However, intrinsic heterogeneity in clients' model architectures and computing capabilities often results in model accuracy loss and the intractable straggler problem, which significantly impairs training effectiveness. To tackle these challenges, this paper proposes a novel Heterogeneity-aware Personalized Federated Learning method, named HAPFL, via multi-level Reinforcement Learning (RL) mechanisms. HAPFL optimizes the training process by incorporating three strategic components: 1) An RL-based heterogeneous model allocation mechanism. The parameter server employs a Proximal Policy Optimization (PPO)-based RL agent to adaptively allocate appropriately sized, differentiated models to clients based on their performance, effectively mitigating performance disparities. 2) An RL-based training intensity adjustment scheme. The parameter server leverages another PPO-based RL agent to dynamically fine-tune the training intensity for each client to further enhance training efficiency and reduce straggling latency. 3) A knowledge distillation-based mutual learning mechanism. Each client deploys both a heterogeneous local model and a homogeneous lightweight model named LiteModel, where these models undergo mutual learning through knowledge distillation. This uniform LiteModel plays a pivotal role in aggregating and sharing global knowledge, significantly enhancing the effectiveness of personalized local training. Experimental results across multiple benchmark datasets demonstrate that HAPFL not only achieves high accuracy but also substantially reduces the overall training time by 20.9%-40.4% and decreases straggling latency by 19.0%-48.0% compared to existing solutions.

Xiao Chen, Jiazhen Huang, Qinting Jiang et al.

Test-time adaptation (TTA) has emerged as a critical technique for enhancing the generalization capability of vision-language models (VLMs) during inference. However, existing approaches often incur substantial computational costs and exhibit poor scalability, primarily due to sample-wise adaptation granularity and reliance on costly auxiliary designs such as data augmentation. To address these limitations, we introduce SAIL (Small Aid, Big Leap), a novel adapter-based TTA framework that leverages a lightweight, learnable AdaptNet to enable efficient and scalable model adaptation. As SAIL's core, a frozen pre-trained VLM collaborates with AdaptNet through a confidence-based interpolation weight, generating robust predictions during inference. These predictions serve as self-supervised targets to align AdaptNet's outputs through efficient batch-wise processing, dramatically reducing computational costs without modifying the VLM or requiring memory caches. To mitigate catastrophic forgetting during continual adaptation, we propose a gradient-aware reset strategy driven by a gradient drift indicator (GDI), which dynamically detects domain transitions and strategically resets AdaptNet for stable adaptation. Extensive experiments across diverse benchmarks on two scenarios demonstrate that SAIL achieves state-of-the-art performance while maintaining low computational costs. These results highlight SAIL's effectiveness, efficiency and scalability for real-world deployment. The code will be released upon acceptance.

Frank Ban, Xi Chen, Rocco A. Servedio et al.

Several recent works have considered the \emph{trace reconstruction problem}, in which an unknown source string $x\in\{0,1\}^n$ is transmitted through a probabilistic channel which may randomly delete coordinates or insert random bits, resulting in a \emph{trace} of $x$. The goal is to reconstruct the original string~$x$ from independent traces of $x$. While the best algorithms known for worst-case strings use $\exp(O(n^{1/3}))$ traces \cite{DOS17,NazarovPeres17}, highly efficient algorithms are known \cite{PZ17,HPP18} for the \emph{average-case} version, in which $x$ is uniformly random. We consider a generalization of this average-case trace reconstruction problem, which we call \emph{average-case population recovery in the presence of insertions and deletions}. In this problem, there is an unknown distribution $\cal{D}$ over $s$ unknown source strings $x^1,\dots,x^s \in \{0,1\}^n$, and each sample is independently generated by drawing some $x^i$ from $\cal{D}$ and returning an independent trace of $x^i$. Building on \cite{PZ17} and \cite{HPP18}, we give an efficient algorithm for this problem. For any support size $s \leq \smash{\exp(Θ(n^{1/3}))}$, for a $1-o(1)$ fraction of all $s$-element support sets $\{x^1,\dots,x^s\} \subset \{0,1\}^n$, for every distribution $\cal{D}$ supported on $\{x^1,\dots,x^s\}$, our algorithm efficiently recovers ${\cal D}$ up to total variation distance $ε$ with high probability, given access to independent traces of independent draws from $\cal{D}$. The algorithm runs in time poly$(n,s,1/ε)$ and its sample complexity is poly$(s,1/ε,\exp(\log^{1/3}n)).$ This polynomial dependence on the support size $s$ is in sharp contrast with the \emph{worst-case} version (when $x^1,\dots,x^s$ may be any strings in $\{0,1\}^n$), in which the sample complexity of the most efficient known algorithm \cite{BCFSS19} is doubly exponential in $s$.

Xi Chen, Qihang Lin, Seyoung Kim et al.

We study the problem of learning high dimensional regression models regularized by a structured-sparsity-inducing penalty that encodes prior structural information on either input or output sides. We consider two widely adopted types of such penalties as our motivating examples: 1) overlapping group lasso penalty, based on the l1/l2 mixed-norm penalty, and 2) graph-guided fusion penalty. For both types of penalties, due to their non-separability, developing an efficient optimization method has remained a challenging problem. In this paper, we propose a general optimization approach, called smoothing proximal gradient method, which can solve the structured sparse regression problems with a smooth convex loss and a wide spectrum of structured-sparsity-inducing penalties. Our approach is based on a general smoothing technique of Nesterov. It achieves a convergence rate faster than the standard first-order method, subgradient method, and is much more scalable than the most widely used interior-point method. Numerical results are reported to demonstrate the efficiency and scalability of the proposed method.