Xinyu Yuan

Pengyu Zhu, Lijun Li, Yaxing Lyu et al.

As LLMs are increasingly deployed as agents, reliable assessment of their agentic capabilities has become essential. However, reported benchmark scores often jointly reflect model capability and the implementation choices each benchmark is packaged with, making cross-benchmark results difficult to interpret as clean measurements of the underlying model. In this work, we present a unified framework for the fair evaluation of LLM agentic capabilities. Driven by a unified configuration system, the framework integrates diverse benchmarks into a standardized instruction--tool--environment format, executes agents through a fixed ReAct-style architecture within a controllable sandbox, and provides an optional offline setting that replaces volatile live environments with curated snapshots, so that framework effects and environment effects can be analyzed separately. Building on this, we unify the evaluation methodology under each benchmark's original task-success criteria, while introducing unified metrics for resource consumption and a taxonomy for decision- and execution-level failure attribution. Within this framework, we adapt 7 widely used benchmarks spanning 24 domains across single-agent, multi-agent, and safety-critical scenarios, and conduct a large-scale empirical analysis over 400K rollouts and 5B tokens on 15 models. The results show that scaffold choice and environmental volatility materially shift benchmark outcomes in both directions, allowing our framework to disentangle intrinsic LLM capabilities from framework- and environment-induced artifacts. We further demonstrate its extensibility as a secure testbed for safety-critical domains. Codes and benchmarks at are available at https://github.com/whfeLingYu/A-Unified-Framework-for-the-Evaluation-of-LLM-Agentic-Capabilities, https://huggingface.co/AgentFramework/Unified_Farmework.

Mikhail Galkin, Xinyu Yuan, Hesham Mostafa et al. · deepmind

Foundation models in language and vision have the ability to run inference on any textual and visual inputs thanks to the transferable representations such as a vocabulary of tokens in language. Knowledge graphs (KGs) have different entity and relation vocabularies that generally do not overlap. The key challenge of designing foundation models on KGs is to learn such transferable representations that enable inference on any graph with arbitrary entity and relation vocabularies. In this work, we make a step towards such foundation models and present ULTRA, an approach for learning universal and transferable graph representations. ULTRA builds relational representations as a function conditioned on their interactions. Such a conditioning strategy allows a pre-trained ULTRA model to inductively generalize to any unseen KG with any relation vocabulary and to be fine-tuned on any graph. Conducting link prediction experiments on 57 different KGs, we find that the zero-shot inductive inference performance of a single pre-trained ULTRA model on unseen graphs of various sizes is often on par or better than strong baselines trained on specific graphs. Fine-tuning further boosts the performance.

Zhaocheng Zhu, Xinyu Yuan, Mikhail Galkin et al. · deepmind

Reasoning on large-scale knowledge graphs has been long dominated by embedding methods. While path-based methods possess the inductive capacity that embeddings lack, their scalability is limited by the exponential number of paths. Here we present A*Net, a scalable path-based method for knowledge graph reasoning. Inspired by the A* algorithm for shortest path problems, our A*Net learns a priority function to select important nodes and edges at each iteration, to reduce time and memory footprint for both training and inference. The ratio of selected nodes and edges can be specified to trade off between performance and efficiency. Experiments on both transductive and inductive knowledge graph reasoning benchmarks show that A*Net achieves competitive performance with existing state-of-the-art path-based methods, while merely visiting 10% nodes and 10% edges at each iteration. On a million-scale dataset ogbl-wikikg2, A*Net not only achieves a new state-of-the-art result, but also converges faster than embedding methods. A*Net is the first path-based method for knowledge graph reasoning at such scale.

Xinyu Yuan, Xixian Liu, Jianan Zhao et al.

Perturbation experiments are central to understanding cellular mechanisms, but remain costly and sparse, motivating prediction of gene expression responses for unobserved conditions. A promising recent direction leverages large language models (LLMs) as "virtual cell" simulators-using stepwise, knowledge-grounded mechanistic reasoning to infer differential expression-pointing toward an interpretable, knowledge-driven paradigm that transcends purely data-driven approaches. However, we find that plausibility is not prediction: despite producing biologically plausible explanations, these methods fail to capture perturbation-specific effects: systematically overestimating differential expression, often underperforming a simple gene-frequency baseline in aggregate evaluations, and collapsing to chance-level performance at the per-gene level. This reveals a reliance on intrinsic gene response tendencies rather than true perturbation reasoning. We trace this failure to how evidence is presented: existing methods evaluate perturbation-gene pairs in isolation, without exposing how related perturbations differ in their effects on the same gene. To address this limitation, we introduce CORE (Contrastive Organization of Relational Evidence), which reframes prediction as a comparison task by organizing evidence into positive and negative outcomes from related perturbations. Using a biomedical knowledge graph for evidence retrieval, CORE improves calibration and substantially boosts perturbation-specific prediction in both LLM-based and non-LLM settings: for example, on drug-perturbation data, CORE-Reasoning improves Qwen3.5-9B aggregate metrics by up to 28.6%, while on generic perturbation data, CORE-Voting raises macro-per-gene AUROC from chance to 0.703 in average across four cell lines. This highlights contrastive evidence organization as essential to reliable LLM-based perturbation reasoning

Minghao Xu, Xinyu Yuan, Santiago Miret et al.

Current protein language models (PLMs) learn protein representations mainly based on their sequences, thereby well capturing co-evolutionary information, but they are unable to explicitly acquire protein functions, which is the end goal of protein representation learning. Fortunately, for many proteins, their textual property descriptions are available, where their various functions are also described. Motivated by this fact, we first build the ProtDescribe dataset to augment protein sequences with text descriptions of their functions and other important properties. Based on this dataset, we propose the ProtST framework to enhance Protein Sequence pre-training and understanding by biomedical Texts. During pre-training, we design three types of tasks, i.e., unimodal mask prediction, multimodal representation alignment and multimodal mask prediction, to enhance a PLM with protein property information with different granularities and, at the same time, preserve the PLM's original representation power. On downstream tasks, ProtST enables both supervised learning and zero-shot prediction. We verify the superiority of ProtST-induced PLMs over previous ones on diverse representation learning benchmarks. Under the zero-shot setting, we show the effectiveness of ProtST on zero-shot protein classification, and ProtST also enables functional protein retrieval from a large-scale database without any function annotation.

Xinyu Yuan, Xixian Liu, Ya Shi Zhang et al.

Building Virtual Cells that can accurately simulate cellular responses to perturbations is a long-standing goal in systems biology. A fundamental challenge is that high-throughput single-cell sequencing is destructive: the same cell cannot be observed both before and after a perturbation. Thus, perturbation prediction requires mapping unpaired control and perturbed populations. Existing models address this by learning maps between distributions, but typically assume a single fixed response distribution when conditioned on observed cellular context (e.g., cell type) and the perturbation type. In reality, responses vary systematically due to unobservable latent factors such as microenvironmental fluctuations and complex batch effects, forming a manifold of possible distributions for the same observed conditions. To account for this variability, we introduce PerturbDiff, which shifts modeling from individual cells to entire distributions. By embedding distributions as points in a Hilbert space, we define a diffusion-based generative process operating directly over probability distributions. This allows PerturbDiff to capture population-level response shifts across hidden factors. Benchmarks on established datasets show that PerturbDiff achieves state-of-the-art performance in single-cell response prediction and generalizes substantially better to unseen perturbations. See our project page (https://katarinayuan.github.io/PerturbDiff-ProjectPage/), where code and data will be made publicly available (https://github.com/DeepGraphLearning/PerturbDiff).

Jianan Zhao, Xixian Liu, Zhihao Zhan et al.

Long-context DNA models are limited by token-mixing cost and by how compression allocates representational budget across the genome. Existing approaches operate close to base-pair resolution, apply fixed downsampling, or learn content-dependent chunks without an explicit genomic budget, making long-context pretraining expensive and difficult to control. We introduce GeneZip, a region-aware DNA compression framework that combines H-Net-style dynamic routing with a Region-Aware Ratio (RAR) objective and bounded routing. GeneZip uses static gene-structure annotations during compression training to specify region-wise base-pairs-per-token (BPT) targets; at inference time, it compresses raw unseen DNA without annotations. GeneZip provides three main benefits. First, it is effective: GeneZip variants achieve the best validation PPL among encoder-based compressors, with GeneZip-70M operating at 137.6 BPT, and across four reproducible DNALongBench tasks--contact map prediction, eQTL prediction, enhancer-target gene prediction, and transcription-initiation signal prediction--GeneZip obtains the best average rank among compared sequence models. Second, it is redundancy-aware: a post-hoc RepeatMasker/TRF analysis shows that, without repeat supervision, GeneZip assigns higher local BPT to TE-derived interspersed repeats and tandem repeats, two major classes of repetitive DNA sequence redundancy. Third, it is efficient: by reducing the effective token-mixing length, GeneZip enables longer-context and larger-capacity pretraining, including 128K-context and 636M-parameter variants on a single A100 80GB GPU, and fine-tunes the eQTL task 50.4x faster than JanusDNA (50 vs. 2520 minutes). These results establish GeneZip as an effective, redundancy-aware, and efficient compression interface for long-context DNA modeling.

Xinyu Yuan, Zhihao Zhan, Zuobai Zhang et al.

Transcriptome foundation models TFMs hold great promises of deciphering the transcriptomic language that dictate diverse cell functions by self-supervised learning on large-scale single-cell gene expression data, and ultimately unraveling the complex mechanisms of human diseases. However, current TFMs treat cells as independent samples and ignore the taxonomic relationships between cell types, which are available in cell ontology graphs. We argue that effectively leveraging this ontology information during the TFM pre-training can improve learning biologically meaningful gene co-expression patterns while preserving TFM as a general purpose foundation model for downstream zero-shot and fine-tuning tasks. To this end, we present single cell, Cell-ontology guided TFM scCello. We introduce cell-type coherence loss and ontology alignment loss, which are minimized along with the masked gene expression prediction loss during the pre-training. The novel loss component guide scCello to learn the cell-type-specific representation and the structural relation between cell types from the cell ontology graph, respectively. We pre-trained scCello on 22 million cells from CellxGene database leveraging their cell-type labels mapped to the cell ontology graph from Open Biological and Biomedical Ontology Foundry. Our TFM demonstrates competitive generalization and transferability performance over the existing TFMs on biologically important tasks including identifying novel cell types of unseen cells, prediction of cell-type-specific marker genes, and cancer drug responses.

Hung-Min Hsu, Xinyu Yuan, Baohua Zhu et al.

Package theft detection has been a challenging task mainly due to lack of training data and a wide variety of package theft cases in reality. In this paper, we propose a new Global and Local Fusion Package Theft Detection Embedding (GLF-PTDE) framework to generate package theft scores for each segment within a video to fulfill the real-world requirements on package theft detection. Moreover, we construct a novel Package Theft Detection dataset to facilitate the research on this task. Our method achieves 80% AUC performance on the newly proposed dataset, showing the effectiveness of the proposed GLF-PTDE framework and its robustness in different real scenes for package theft detection.

Xinyu Yuan, Zichen Wang, Marcus Collins et al.

Recent years have witnessed a surge in the development of protein structural tokenization methods, which chunk protein 3D structures into discrete or continuous representations. Structure tokenization enables the direct application of powerful techniques like language modeling for protein structures, and large multimodal models to integrate structures with protein sequences and functional texts. Despite the progress, the capabilities and limitations of these methods remain poorly understood due to the lack of a unified evaluation framework. We first introduce StructTokenBench, a framework that comprehensively evaluates the quality and efficiency of structure tokenizers, focusing on fine-grained local substructures rather than global structures, as typical in existing benchmarks. Our evaluations reveal that no single model dominates all benchmarking perspectives. Observations of codebook under-utilization led us to develop AminoAseed, a simple yet effective strategy that enhances codebook gradient updates and optimally balances codebook size and dimension for improved tokenizer utilization and quality. Compared to the leading model ESM3, our method achieves an average of 6.31% performance improvement across 24 supervised tasks, with sensitivity and utilization rates increased by 12.83% and 124.03%, respectively. Source code and model weights are available at https://github.com/KatarinaYuan/StructTokenBench

Xinyu Yuan, Yan Qiao, Zonghui Wang et al.

Sketch techniques have been extensively studied in recent years and are especially well-suited to data streaming scenarios, where the sketch summary is updated quickly and compactly. However, it is challenging to recover the current state from these summaries in a way that is accurate, fast, and real. In this paper, we seek a solution that reconciles this tension, aiming for near-perfect recovery with lightweight computational procedures. Focusing on linear sketching problems of the form $\boldsymbolΦf \rightarrow f$, our study proceeds in three stages. First, we dissect existing techniques and show the root cause of the sketching dilemma: an orthogonal information loss. Second, we examine how generative priors can be leveraged to bridge the information gap. Third, we propose FLORE, a novel generative sketching framework that embraces these analyses to achieve the best of all worlds. More importantly, FLORE can be trained without access to ground-truth data. Comprehensive evaluations demonstrate FLORE's ability to provide high-quality recovery, and support summary with low computing overhead, outperforming previous methods by up to 1000 times in error reduction and 100 times in processing speed compared to learning-based solutions.

Xinyu Yuan, Yan Qiao

Denoising diffusion probabilistic models (DDPMs) are becoming the leading paradigm for generative models. It has recently shown breakthroughs in audio synthesis, time series imputation and forecasting. In this paper, we propose Diffusion-TS, a novel diffusion-based framework that generates multivariate time series samples of high quality by using an encoder-decoder transformer with disentangled temporal representations, in which the decomposition technique guides Diffusion-TS to capture the semantic meaning of time series while transformers mine detailed sequential information from the noisy model input. Different from existing diffusion-based approaches, we train the model to directly reconstruct the sample instead of the noise in each diffusion step, combining a Fourier-based loss term. Diffusion-TS is expected to generate time series satisfying both interpretablity and realness. In addition, it is shown that the proposed Diffusion-TS can be easily extended to conditional generation tasks, such as forecasting and imputation, without any model changes. This also motivates us to further explore the performance of Diffusion-TS under irregular settings. Finally, through qualitative and quantitative experiments, results show that Diffusion-TS achieves the state-of-the-art results on various realistic analyses of time series.

Xinyu Yuan, Yan Qiao, Zonghui Wang et al.

The multi-commodity flow (MCF) problem is a fundamental topic in network flow and combinatorial optimization, with broad applications in transportation, communication, and logistics, etc. Nowadays, the rapid expansion of allocation systems has posed challenges for existing optimization engines in balancing optimality and tractability. In this paper, we present Pram, the first ML-based method that leverages the reasoning power of multimodal language models (MLMs) for addressing the trade-off dilemma -- a great need of service providers. As part of our proposal, Pram (i) quickly computes high-quality allocations by dividing the original problem into local subproblems, which are then resolved by an MLM-powered "agent", and (ii) ensures global consistency by harmonizing these subproblems via a multi-agent reinforcement learning algorithm. Theoretically, we show that Pram, which learns to perform gradient descent in context, provably converges to the optimum within the family of MCF problems. Empirically, on real-world datasets and public topologies, Pram achieves performance comparable to, and in some cases even surpassing, linear programming solvers (very close to the optimal solution), and substantially lower runtimes (1 to 2 orders of magnitude faster). Moreover, Pram exhibits strong robustness (<10\% performance degradation under link failures or flow bursts), demonstrating MLM's generalization ability to unforeseen events. Pram is objective-agnostic and seamlessly integrates with mainstream allocation systems, providing a practical and scalable solution for future networks.

Xinyu Yuan, Yan Qiao, Meng Li et al.

Estimating the frequency of items on the high-volume, fast data stream has been extensively studied in many areas, such as database and network measurement. Traditional sketches provide only coarse estimates under strict memory constraints. Although some learning-augmented methods have emerged recently, they typically rely on offline training with real frequencies or/and labels, which are often unavailable. Moreover, these methods suffer from slow update speeds, limiting their suitability for real-time processing despite offering only marginal accuracy improvements. To overcome these challenges, we propose UCL-sketch, a practical learning-based paradigm for per-key frequency estimation. Our design introduces two key innovations: (i) an online training mechanism based on equivalent learning that requires no ground truth (GT), and (ii) a highly scalable architecture leveraging logically structured estimation buckets to scale to real-world data stream. The UCL-sketch, which utilizes compressive sensing (CS), converges to an estimator that provably yields a error bound far lower than that of prior works, without sacrificing the speed of processing. Extensive experiments on both real-world and synthetic datasets demonstrate that our approach outperforms previously proposed approaches regarding per-key accuracy and distribution. Notably, under extremely tight memory budgets, its quality almost matches that of an (infeasible) omniscient oracle. Moreover, compared to the existing equation-based sketch, UCL-sketch achieves an average decoding speedup of nearly 500 times. To help further research and development, our code is publicly available at https://github.com/Y-debug-sys/UCL-sketch.

Xinyu Yuan, Yan Qiao, Pei Zhao et al.

The traffic matrix estimation (TME) problem has been widely researched for decades of years. Recent progresses in deep generative models offer new opportunities to tackle TME problems in a more advanced way. In this paper, we leverage the powerful ability of denoising diffusion probabilistic models (DDPMs) on distribution learning, and for the first time adopt DDPM to address the TME problem. To ensure a good performance of DDPM on learning the distributions of TMs, we design a preprocessing module to reduce the dimensions of TMs while keeping the data variety of each OD flow. To improve the estimation accuracy, we parameterize the noise factors in DDPM and transform the TME problem into a gradient-descent optimization problem. Finally, we compared our method with the state-of-the-art TME methods using two real-world TM datasets, the experimental results strongly demonstrate the superiority of our method on both TM synthesis and TM estimation.

Xinyu Yuan, Yan Qiao, Zhenchun Wei et al.

Due to network operation and maintenance relying heavily on network traffic monitoring, traffic matrix analysis has been one of the most crucial issues for network management related tasks. However, it is challenging to reliably obtain the precise measurement in computer networks because of the high measurement cost, and the unavoidable transmission loss. Although some methods proposed in recent years allowed estimating network traffic from partial flow-level or link-level measurements, they often perform poorly for traffic matrix estimation nowadays. Despite strong assumptions like low-rank structure and the prior distribution, existing techniques are usually task-specific and tend to be significantly worse as modern network communication is extremely complicated and dynamic. To address the dilemma, this paper proposed a diffusion-based traffic matrix analysis framework named Diffusion-TM, which leverages problem-agnostic diffusion to notably elevate the estimation performance in both traffic distribution and accuracy. The novel framework not only takes advantage of the powerful generative ability of diffusion models to produce realistic network traffic, but also leverages the denoising process to unbiasedly estimate all end-to-end traffic in a plug-and-play manner under theoretical guarantee. Moreover, taking into account that compiling an intact traffic dataset is usually infeasible, we also propose a two-stage training scheme to make our framework be insensitive to missing values in the dataset. With extensive experiments with real-world datasets, we illustrate the effectiveness of Diffusion-TM on several tasks. Moreover, the results also demonstrate that our method can obtain promising results even with $5\%$ known values left in the datasets.

Zhaocheng Zhu, Chence Shi, Zuobai Zhang et al.

Machine learning has huge potential to revolutionize the field of drug discovery and is attracting increasing attention in recent years. However, lacking domain knowledge (e.g., which tasks to work on), standard benchmarks and data preprocessing pipelines are the main obstacles for machine learning researchers to work in this domain. To facilitate the progress of machine learning for drug discovery, we develop TorchDrug, a powerful and flexible machine learning platform for drug discovery built on top of PyTorch. TorchDrug benchmarks a variety of important tasks in drug discovery, including molecular property prediction, pretrained molecular representations, de novo molecular design and optimization, retrosynthsis prediction, and biomedical knowledge graph reasoning. State-of-the-art techniques based on geometric deep learning (or graph machine learning), deep generative models, reinforcement learning and knowledge graph reasoning are implemented for these tasks. TorchDrug features a hierarchical interface that facilitates customization from both novices and experts in this domain. Tutorials, benchmark results and documentation are available at https://torchdrug.ai. Code is released under Apache License 2.0.