arXiv subjects

Yue Hu

Publications and source records attributed to Yue Hu.

At least 37 records · Page 2Linked to original sources

arXiv2026

GLoRI: Closed-Loop Whole-Body Tracking with Global-Local Reference Interaction for Humanoid Loco-Manipulation

Humanoid loco-manipulation requires accurate whole-body motion tracking in the world frame for physical interaction. While local references preserve motion structure, they lack explicit constraints on absolute spatial placement, leading to accumulated global errors. Existing globally aware approaches augment teleoperation policies with global observations but do not explicitly integrate global correction with local motion guidance, limiting autonomous tracking accuracy. We present GLoRI, a closed-loop whole-body controller that integrates structured global reference and feedback with local motion guidance. Its GLoRI-Net uses Global-Local Cross Attention(GLCA) to refine local keypoint features with global target and pose-difference features, preserving motion structure while correcting world-frame placement. GLoRI achieves 100% completion and a g-MPJPE of 6.44cm on held-out HuMoTo motions. This accuracy remains robust under direct Isaac Gym-to-MuJoCo transfer without fine-tuning, demonstrating strong generalization. Furthermore, such accuracy and generalization enable autonomous loco-manipulation with a single policy on a real Unitree G1 interacting with diverse unseen objects, extending beyond prior systems that primarily rely on teleoperation or focus on single-object interactions.

cs.RO↗

arXiv2026

ACES VII. Compact Continuum Source Catalog of the Central Molecular Zone

The Central Molecular Zone (CMZ) resides in the inner few hundred parsecs of our Galaxy, and despite being the largest reservoir of dense molecular gas in the Milky Way, it has a relatively low present-day star formation rate (SFR) of $\sim0.08~M_{\odot}~\text{yr}^{-1}$. Continuum and spectral line observations from the Atacama Large Millimeter/submillimeter Array (ALMA) CMZ Exploration Survey (ACES) provide the first full-coverage, high-resolution map of the inner 200 parsecs of the CMZ at 3 mm. In this paper we present the ACES catalog of compact continuum sources, the most complete catalog of potential sites of star formation in the CMZ to date. Using an automated dendrogram-based source extraction procedure in combination with a by-eye morphological classification scheme, we produce a `full' catalog of 1735 detections in total. Additionally, we use spectral index measurements to generate a `filtered' catalog of 567 sources with minimal contamination from non-thermal filaments and extended free-free emission. We find that 359 ($\sim63\%$) of the filtered catalog sources are located at column densities $< 10^{23}$ cm$^{-2}$, outside of the densest molecular cloud regions, 195 of which have not been identified in previous surveys. After cross-referencing with various catalogs generated from data at different wavelengths, we consider it likely that many of these newly discovered detections are produced by pre/protostellar sources or compact HII regions.

astro-ph.GA↗

arXiv2026

Shuffle algebra realizations for modular Yangians

We study the shuffle algebra realization of positive modular Yangians of classical type over an algebraically closed field of characteristic $p>3$. We show that, unlike in characteristic zero, the natural shuffle homomorphism has a nontrivial kernel. Its image is characterized by a $p$-wheel condition, while its kernel is precisely the ideal generated by the $p$-th powers of the Lyndon root vectors. This identifies the corresponding quotient with the small Yangian arising from a $\mathbb Z[\frac12]$-integral form. As part of the construction, we establish a PBW basis for the integral form and obtain the PBW theorem and $p$-center results for the Drinfeld presentation of modular Yangians.

math.QA↗

arXiv2026

Plasmoid-Mediated 2D Magnetic Reconnection in Partially Ionized Plasmas

Magnetic reconnection in partially ionized plasmas is an important channel for energy release. While the plasmoid instability is well characterized in 2D fully ionized plasmas, its behavior in the presence of neutral-dominated plasma remains poorly understood in the nonlinear, high-Lundquist-number ($S = 10^5$) regime. We present high-resolution ($16384 \times 4096$ cells) two-dimensional two-fluid (ion $+$ neutral) simulations of Harris-sheet reconnection with upstream plasma beta $β= 2$, comparing fully ionized and partially ionized (ionization fraction $ξ= 10^{-1}$ and $10^{-2}$) regimes. Neutral-ion decoupling accelerates the linear tearing stage and alters the plasmoid hierarchy: the large-scale ``monster'' plasmoid that dominates the fully ionized case is suppressed, and the sheet instead fragments into a dense chain of sub-scale plasmoids. Below the neutral-ion decoupling scale $\ell_{\rm dec}$, ions concentrate into the plasmoids, reaching peak overdensities $ρ_i/ρ_{i,0} \approx 10$ ($ξ= 10^{-1}$) and $3-5\times10^{3}$ ($ξ= 10^{-2}$), while the neutrals remain comparatively smooth. This local pile-up raises the ionization fraction and recouples the two fluids within the plasmoids. Measured from the out-of-plane electric field at the reconnection sites, the reconnection rate in the $ξ= 10^{-2}$ case achieves $R_{\rm rec}\approx0.01$, whereas the $ξ= 10^{-1}$ case rises to a rate $\approx0.02$ and further $0.035$ when apparent coalescence occurs. In the $ξ= 10^{-2}$ case, the ambipolar drift drives a rapid ion inflow $\sim0.5\,v_{A,0}$ into the layer at the same reconnection sites, far above the neutral inflow velocity $\sim0.1\,v_{A,0}$. Here, $v_{A,0}$ is the upstream total Alfvén speed.

astro-ph.HE↗

arXiv2026

Slow Cosmic-Ray Diffusion in Supersonic and Super-Alfvénic Turbulence

Extended TeV-PeV gamma-ray halo observations imply cosmic-ray (CR) diffusion that is both substantially slower than the Galactic mean and only weakly anisotropic, despite the presence of a large-scale Galactic magnetic field. We investigate whether such transport can arise in partially ionized source environments, where ion-neutral damping removes the small-scale fluctuations responsible for gyroresonant scattering. Using two-fluid magnetohydrodynamic turbulence simulations with relativistic test-particle tracking, we find that damping increases the parallel diffusion coefficient by approximately two orders of magnitude in trans-sonic turbulence, thereby strongly enhancing field-aligned transport. In supersonic turbulence, shock-associated magnetic fluctuations survive the damping and sustain non-resonant pitch-angle scattering, limiting the increase in parallel diffusion to a factor of a few to ten. At high Alfvénic Mach numbers, $M_A\gtrsim5$, the magnetic-field direction decorrelates over the Alfvén scale, driving parallel diffusion coefficient $D_\parallel$ and perpendicular diffusion coefficient $D_\perp$ toward equality. For $M_s\simeq10$, $M_A\simeq5$-$10$, and representative parameters for hundred-TeV gamma-ray source environments, we obtain comparable $D_\parallel$ and $D_\perp$, within the observationally inferred range of $\sim10^{27}$-$10^{28}\,\mathrm{cm^2\,s^{-1}}$. These results show that strongly supersonic, highly super-Alfvénic turbulence in source environments can sustain slow, nearly isotropic CR transport even when the ion-neutral damping effect is important.

astro-ph.HE↗

arXiv2026

VIDP: Variable Impedance Diffusion Policy for Compliant Robot Manipulation from Diverse Demonstrations

Contact-rich manipulation requires precise tracking and mechanical compliance, where variable impedance control can improve robustness in task success, whereas static compliance cannot adapt to varying contact constraints. Variable impedance skills can be learned from demonstrations, avoiding complex modeling, but compliance is a hidden variable in force-agnostic kinematic data. While existing methods infer compliance from trajectory variations, these variations may reflect geometric adaptation and not intentional compliance when subject to changing spatial layouts. Therefore, this letter introduces Variable Impedance Diffusion Policy (VIDP), an imitation learning-based variable impedance control framework leveraging a Task-Parameterized Directionality-Aware Mixture Model (TP-DAMM) to extract physically consistent trajectory distributions from diverse demonstrations. By mapping distributions to stiffness profiles, VIDP jointly predicts pose actions and task compliance without force sensors. Real-world experiments show that VIDP significantly outperforms fixed-impedance baselines in task success rate while reducing interaction forces with respect to high stiffness controllers and tracking errors with respect to low stiffness baselines.

cs.RO↗

arXiv2026

EndoVLM: An Endoscopy Vision-Language Pre-training Model via Anatomy-Guided Sparsity and Progressive Alignment

The development of foundation models (FMs) is crucial for advancing endoscopic image analysis. However, existing endoscopy FMs mainly rely on self-supervised learning from uni-modal images or videos, overlooking the rich semantic knowledge contained in clinical reports. Furthermore, effectively leveraging these records is hindered by a fundamental modality gap: structured anatomical descriptions are not naturally mapped to specific frames within the high-redundancy, uncurated visual streams. In this paper, we present EndoVLM, a novel vision-language FM pre-trained on over 348K endoscopic examinations, each pairing a clinical report with its corresponding image collection. An Anatomy-Guided Sparse Pooling mechanism utilizes textual descriptions as queries to drive sparse attention, efficiently aggregating semantically salient frames into anatomy-specific visual representations across redundant image-sets. Next, a Progressive Semantic-Aware Alignment strategy models clinical taxonomy (anatomy and pathological status) via structured soft targets, bridging the gap from global patient-level matching to fine-grained localized alignment. Finally, a Semantic-Concentrated Masked Autoencoder is applied exclusively to these semantic-rich frames, integrating low-level visual precision with robust high-level semantic representation. Extensive experiments across various downstream tasks demonstrate that EndoVLM outperforms existing foundation models and remains competitive with task-specific methods. Remarkably, EndoVLM also exhibits robust zero-shot generalization capabilities, highlighting its potential for broader clinical application.

cs.CV↗

arXiv2026

NavVerse: Benchmarking Indoor-to-Outdoor Embodied Navigation in Continuous Robot Simulation

Robots deployed in delivery, campus, and emergency-response settings often need to navigate from buildings to streets within a single continuous episode. Existing benchmarks usually evaluate indoor and outdoor navigation separately, and many abstract away robot execution, leaving exit finding, boundary traversal, adaptation, and kinodynamic failures underexplored. We introduce NavVerse, a physics-enabled benchmark for indoor-to-outdoor embodied navigation. NavVerse contains 100 indoor scenes, 50 urban outdoor scenes, and 50 indoor-to-outdoor scenes, and 10,000 episodes spanning Object Navigation, Vision-and-Language Navigation, and Place Navigation tasks, where agents search for semantic points of interest such as restaurants or banks. Agents are evaluated through executable robot interfaces using task-success, path-efficiency, and safety metrics. Zero-shot experiments with RL, VLA, and modular baselines show that current agents remain far from solving cross-context navigation: end-to-end VLAs obtain the highest zero-shot success, while the modular method provides the strongest safety profile. PlaceNav further reveals a clear drop from outdoor to indoor-to-outdoor scenes, indicating that adaptation remains major bottleneck.

cs.RO↗

arXiv2026

Model Checking for Regressions Based on Weighted Residual Processes with Diverging Number of Predictors

The integrated conditional moment (ICM) test is a classical and widely used method for assessing the adequacy of regression models. Although it performs well in fixed-dimension settings, its behavior changes dramatically when the predictor dimension diverges: in such regimes, the limiting null and alternative distributions of the ICM statistic degenerate to fixed constants. Moreover, when the number of predictors diverges, the commonly used wild bootstrap no longer approximates the null distribution of the ICM statistic well, leading to size distortion and substantial power loss. To address these challenges, we propose a new specification test based on weighted residual processes for evaluating the parametric form of the regression mean function in high-dimensional settings where the number of predictors increases with the sample size. We establish the asymptotic properties of the test statistic under the null hypothesis and under global and local alternatives. The proposed test maintains the nominal significance level and can detect local alternatives that deviate from the null hypothesis at the parametric rate $1/\sqrt{n}$. Furthermore, we propose a smooth residual bootstrap to approximate the limiting null distribution and establish its validity in high-dimensional settings. Two simulation studies and a real-data example are conducted to evaluate the finite-sample performance of the proposed test.

stat.ME↗

arXiv2026

An ultralow-loss integrated photonic platform for discrete-variable quantum information processing

Photonic integrated circuits offer a scalable and robust route toward quantum information technologies by consolidating photon sources and linear optical networks onto compact, wafer-manufacturable chips. Although silicon photonics has enabled diverse discrete-variable quantum breakthroughs -- spanning multiphoton entanglement, quantum networking, and photonic qubit fusion for quantum computing -- scaling these platforms beyond proof-of-principle demonstrations remains severely constrained by a critical system-level bottleneck. Optical loss compounds rapidly across photon generation, routing, and state analysis, causing multiphoton generation probabilities to plummet exponentially as circuit depth and complexity grow. Here we overcome this rate-loss barrier by demonstrating a monolithic, ultralow-loss silicon nitride (Si$_3$N$_4$) integrated photonic platform engineered for high-performance discrete-variable quantum information processing. Our architecture seamlessly integrates narrowband photon-pair sources with low-loss qubit-fusion circuits and reconfigurable state-analysis interferometers. The on-chip sources prepare Einstein-Podolsky-Rosen (EPR) states with a fidelity of 0.9875(3) and exhibit near-unity photon indistinguishability, yielding a heralded Hong-Ou-Mandel interference visibility of 0.990(6). By executing on-chip fusion of two EPR states, we synthesize and characterize four-photon Greenberger-Horne-Zeilinger states with a record fidelity of 0.943(8) and a fourfold count rate of 27 Hz -- more than two orders of magnitude higher than previous silicon-photonic implementations. Combined with standard CMOS-compatible fabrication on 150-mm-diameter wafers, these results establish ultralow-loss Si$_3$N$_4$ integrated photonics as a definitive, manufacturable platform for deployable, large-scale quantum information processors.

quant-ph↗

arXiv2026

LTM: Large-scale Terrain Model for Wildfire-prone Landscapes

Accurate 3D terrain maps are essential for emergency response when assessing wildfire hazards. However, wildfire-prone regions often span vast areas where conventional reconstruction methods underperform. Airborne LiDAR systems provide high-resolution terrain data, but they are expensive and infrequently updated. Image-based methods offer a lower-cost alternative, but struggle due to sparse visual features and limited image overlap. We propose a multi-modal reconstruction framework leveraging outdated Digital Elevation Models (DEMs) as geometric priors for image-based 3D reconstruction. Our key innovation is physics-based pixel-pixel alignment between images and DEM data, dramatically reducing computational complexity by eliminating expensive feature matching procedures. To validate our approach, we developed a large-terrain simulator based on a real wildfire-prone area, generating realistic images enabling a comprehensive evaluation. Given posed images and legacy DEMs, our method produces high-fidelity depth maps while maintaining real-time performance. We find significant improvements in reconstruction accuracy and computational efficiency over existing techniques, offering a scalable solution for wildfire response.

cs.CV↗

arXiv2026

Multimodal Spatiotemporal-Frequency Fusion with Peak Enhancement for Cellular Traffic Forecasting

Accurate forecasting of cellular network traffic is essential for network planning, resource allocation, and quality-of-service assurance in modern mobile communication systems. Real-world traffic often exhibits bursty endogenous dynamics and disturbances triggered by external urban events, which makes reliable prediction highly challenging. Most existing spatiotemporal traffic forecasting methods primarily focus on intrinsic traffic patterns or structural relationships within a single modality, and rarely model burst behavior together with exogenous contextual signals. To address this issue, we propose \textbf{MSPF-Net}, a multimodal cellular traffic forecasting framework that integrates external contextual information. Specifically, MSPF-Net consists of a Spatiotemporal-Frequency Traffic Encoder for capturing temporal, spatial, and spectral traffic patterns, a Peak Enhancement Module for extracting burst-aware representations of sudden spikes, a News Context Representation Module for encoding urban news streams into exogenous contextual embeddings, and a Dynamic Fusion Prediction Module for adaptively integrating these heterogeneous signals to generate forecasts. Experiments on the Milano, Trento, and LTE traffic datasets demonstrate that jointly modeling traffic dynamics, burst patterns, and news contextual signals can effectively improve forecasting performance.

cs.LG↗

arXiv2026

A Good Initialization is All You Need for Faithful Visual Attribution

Faithful visual attribution identifies which image regions support a model prediction. Search-based perturbation methods lead the insertion--deletion faithfulness frontier by masking regions and measuring score changes, but they usually output a complete ordering of all regions. Many applications, especially MLLM attribution and repair, only need a compact top-$k$ evidence mask. We study this mask-first attribution problem. An exactly $k$-region mask is combinatorial: useful evidence can depend on interactions among fine regions. Coarse grouping can stabilize early search but aggregates redundant content, whereas one-step scoring can miss high-value combinations. We introduce two forward-only methods. \textsc{CoPAIR} uses a PhaseWin--Greedy gap diagnosis to construct coarse singleton/pair candidates that warm-start full-ordering search. \textsc{TRACE} directly searches fixed-cardinality fine-region masks with cross-entropy sampling, elite retention, and distribution updates, with a finite-budget recovery analysis. The resulting evidence set can be returned as a compact attribution mask or used to initialize Greedy or PhaseWin when a complete ranking is required. Across ImageNet classification with CLIP ViT-L/14, CLIP RN101, and ResNet-101, our initialized search methods establish a new state-of-the-art frontier for faithful full-ordering attribution under inclusive forward-call accounting. On POPE and RePOPE with Qwen2.5-VL-3B-Instruct and LLaVA-v1.5-7B, \textsc{TRACE}+Greedy gives the strongest search-based MLLM attribution results. Direct \textsc{TRACE} masks further achieve single-point RePOPE repair rates of $94.44\%$ and $96.00\%$, showing that compact evidence masks can be actionable attribution outputs, not merely prefixes of full rankings.

cs.CV↗

arXiv2026

Galactic Dust Polarization in Turbulent Multiphase ISM: On the Origin of the $EE/BB$ Asymmetry

Polarized thermal emission from Galactic dust is the dominant foreground for CMB polarization measurements at high frequencies, with its statistical properties shaped by the interplay between turbulence and magnetic fields in the multiphase interstellar medium (ISM). Variations in turbulence regime and density-magnetic-field alignment across the warm (WNM), unstable (UNM), and cold (CNM) neutral media should imprint distinct signatures on the power spectra and $EE/BB$ power ratio, yet the relative contributions of these phases remain poorly constrained. Using high-resolution 3D magnetohydrodynamic simulations of a turbulent multiphase ISM coupled with synthetic dust polarization maps, we quantify phase-dependent turbulence, anisotropy, and alignment properties. We find that the trans-Alfvénic and transonic WNM and UNM are strongly anisotropic, exhibiting tight alignment of density and velocity structures with the local magnetic field. In contrast, the super-Alfvénic and supersonic CNM displays reduced anisotropy and weak alignment. These dynamical differences are reflected in the statistical scaling of fluctuations: the square root of the second-order velocity structure function exhibits a slope near $1/3$ in the WNM, near $1/2$ in the CNM, and intermediate in the UNM. Comparing our synthetic polarization power spectra with \textit{Planck} measurements, we find that polarization from UNM dust yields spectral slopes closest to the \textit{Planck}-inferred values, whereas WNM and CNM dust produce steeper and shallower spectra, respectively. The WNM yields $EE/BB>2$, the UNM gives $EE/BB\sim2$, and the CNM yields $EE/BB\approx1$. These results suggest that UNM dust may be an important contributor to the polarized foreground under typical high-latitude ISM conditions. We present predictions at 150 GHz to inform foreground modeling and separation.

astro-ph.GA↗

arXiv2026

Seeing Through Occlusion: Deterministic Arm Kinematic Correction for Robot Teleoperation

Markerless, single-RGB-D-camera motion capture provides a low-cost and non-invasive alternative to conventional marker-based systems for robot teleoperation; however, depth estimation often degrades in the presence of self-occlusion, particularly during upper-limb motion. This paper presents an Arm Kinematic Correction (AKC) method that improves depth estimation by enforcing geometric constraints based on constant arm lengths. The proposed approach reconstructs occluded joint depths by leveraging wrist positions and predefined arm lengths via a deterministic formulation based on the Pythagorean theorem, thereby avoiding the need for complex probabilistic modeling or parameter tuning. Experimental validation against a Vicon reference system demonstrates reliable performance for both static and dynamic joint motions, evaluated using root-mean-square error (RMSE) and Pearson correlation. Furthermore, motion-mapping teleoperation is successfully demonstrated in both simulated and physical robot environments. The results show that AKC enhances robustness and preserves anatomical consistency under long-duration, severe self-occlusion, even when paired with less reliable temporal filters, highlighting its practicality for real-time applications such as robot teleoperation and human-robot interaction.

cs.RO↗

arXiv2026

MASLab: A Unified and Comprehensive Codebase for LLM-based Multi-Agent Systems

LLM-based multi-agent systems (MAS) have demonstrated significant potential in enhancing single LLMs to address complex and diverse tasks in practical applications. Despite considerable advancements, the field lacks a unified codebase that consolidates existing methods, resulting in redundant re-implementation efforts, unfair comparisons, and high entry barriers for researchers. To address these challenges, we introduce MASLab, a unified, comprehensive, and research-friendly codebase for LLM-based MAS. (1) MASLab integrates over 20 established methods across multiple domains, each rigorously validated by comparing step-by-step outputs with its official implementation. (2) MASLab provides a unified environment with various benchmarks for fair comparisons among methods, ensuring consistent inputs and standardized evaluation protocols. (3) MASLab implements methods within a shared streamlined structure, lowering the barriers for understanding and extension. Building on MASLab, we conduct extensive experiments covering 10+ benchmarks and 8 models, offering researchers a clear and comprehensive view of the current landscape of MAS methods. MASLab will continue to evolve, tracking the latest developments in the field, and invite contributions from the broader open-source community.

cs.CL↗

arXiv2026

EasyNano: rapid epitope-targeted nanobody CDR design via differentiable distogram optimization with ESMFold2

Computational design of nanobodies that bind user-specified protein epitopes could transform therapeutic development, but current methods either rely on stochastic sampling requiring days of GPU computation or inverse folding approaches unable to target epitopes directly. Here we present EasyNano, a practical pipeline for rapid, epitope-targeted nanobody complementarity-determining region (CDR) design that operates in approximately 10-20 minutes on a high-end personal workstation. EasyNano optimizes CDR residue logits via gradient descent through the ESMFold2 pairwise distance distogram, using the lightweight ESMFold2-Fast model (721M) as a differentiable oracle guided by a composite loss including a dedicated epitope proximity term. A full ESMFold2 (1.3B) CA-coordinate structure prior prevents framework pose drift. The wild-type logit initialization bias emerges as a critical practical parameter controlling CDR mutability. Across six target-framework pairs spanning self-recovery and de novo design scenarios, EasyNano improves ipTM by up to +0.559 -- from 0.143 to 0.702 (Ty1/RBD) -- and achieves a 4.6-fold improvement (ipTM 0.117 to 0.538) on a manually docked AQP4-targeting framework, while preserving ipTM on already-strong binders. Random CDR baselines (n=30 per target) confirm statistical significance (5.7 sigma above random mean for Ty1). Multi-seed analysis reveals diverse local minima, underscoring the importance of replicate runs. Kabsch cross-validation against crystal structures confirms that designed CDRs preserve the framework pose basin. EasyNano demonstrates that ESMFold2-based differentiable optimization provides a fast, practical, and epitope-specific approach to nanobody CDR design.

q-bio.QM↗

arXiv2026

Interpretable enzyme function prediction via sparse autoencoder features of ESMC across the microbial protein universe

Microbial genomes and metagenomes contain millions of proteins whose enzymatic functions remain unknown, the enzyme dark matter. While deep learning has improved protein function prediction, most methods are black boxes relying on sequence or structural similarity, limiting discovery of novel catalytic activities. The ESMC-6B protein language model and its sparse autoencoder with a 16,384-dimensional codebook of interpretable biological concepts, each annotated by GPT-5, creates a new opportunity: using these features directly as semantic signatures for enzyme function. Here, we show that ESMC-SAE features enable accurate and interpretable enzyme commission (EC) number prediction without task-specific training or GPU-intensive computation. On a balanced benchmark of 4,868 microbial SwissProt enzymes across 161 EC3 subclasses, ESMC-SAE binary features achieve 78.9% top-1 and 88.5% top-5 accuracy, 37.6% higher than 3-mer baselines (57.3%). In leave-one-EC3-class-out evaluation simulating discovery of novel enzyme classes, SAE features recover the EC1 superclass in 47.7% of cases (3.3x random, 14.3%), versus 26.6% for sequence methods. Discriminative features correspond to mechanistically interpretable concepts: catalytic triad geometry for hydrolases, NAD(P)H-binding Rossmann folds for oxidoreductases, phosphate-binding P-loops for transferases. We also survey the ESM Atlas of 7.7 million clusters and identify 169,859 dark enzyme-like candidates across all major microbial phyla. Our results establish a paradigm for enzyme function discovery in microbial dark matter: interpretable by design, scalable without GPU clusters, and applicable to the billions of proteins in the ESM Atlas.

q-bio.QM↗