arXiv subjects

Jingyuan Li

Publications and source records attributed to Jingyuan Li.

At least 37 records · Page 2Linked to original sources

arXiv2026

LabBuilder: Protocol-Grounded 3D Layout Generation for Interactable and Safe Laboratory

Automated laboratories hold the promise of accelerating scientific discovery, yet their deployment is bottlenecked by the difficulty of designing safe and executable environments. While simulator-based design offers scalability, existing 3D scene generation methods are primarily tailored for household settings, optimizing for visual plausibility while neglecting the protocol grounding and layout-level safety constraints essential for scientific experimentation. We present LabBuilder, an end-to-end system that generates and verifies 3D laboratory layouts from concise textual specifications. It operates through three tightly coupled components: LabForge first curates a meta-dataset of annotated assets and chemical knowledge, translating natural language specifications into structured protocols; building on these protocols, LabGen synthesizes laboratory layouts via an iterative, constraint-aware optimization strategy; finally, LabTouchstone evaluates the resulting layouts as a unified benchmark. Extensive experiments demonstrate that LabBuilder significantly outperforms existing state-of-the-art methods, producing laboratory environments that are realistic and valid under modeled geometric, chemical-safety, and navigation constraints.

cs.CV↗

arXiv2026

A cross-species neural foundation model for end-to-end speech decoding

Speech brain-computer interfaces (BCIs) aim to restore communication for people with paralysis by translating neural activity into text. Most systems use cascaded frameworks that decode phonemes before assembling sentences with an n-gram language model (LM), preventing joint optimization of all stages simultaneously. Here, we introduce an end-to-end BraIn-to-Text (BIT) framework that translates neural activity into coherent sentences using a single differentiable neural network. Central to our approach is a cross-task, cross-species pretrained neural encoder, whose representations transfer to both attempted and imagined speech. In a cascaded setting with an n-gram LM, the pretrained encoder establishes a new state-of-the-art (SOTA) on the Brain-to-Text '24 and '25 benchmarks. Integrated end-to-end with audio large language models (LLMs) and trained with contrastive learning for cross-modal alignment, BIT reduces the word error rate (WER) of the prior end-to-end method from 24.69% to 10.22%. Notably, we find that small-scale audio LLMs markedly improve end-to-end decoding. Beyond record-setting performance, BIT aligns attempted and imagined speech embeddings to enable cross-task generalization. Altogether, our approach advances the integration of large, diverse neural datasets, paving the way for an end-to-end decoding framework that supports seamless, differentiable optimization.

cs.CL↗

arXiv2026

Neural Continuous-Time Markov Chain: Discrete Diffusion via Decoupled Jump Timing and Direction

Discrete diffusion models based on continuous-time Markov chains (CTMCs) have shown strong performance on language and discrete data generation, yet existing approaches typically parameterize the reverse rate matrix monolithically -- through proxies such as concrete scores (SEDD) or clean-data predictions (MDLM, GIDD) -- rather than aligning the parameterization with the intrinsic CTMC decomposition into jump timing and jump direction. We propose \textbf{Neural CTMC}, which exploits the underlying Poisson structure of CTMC dynamics by separately parameterizing the reverse process through an \emph{exit rate} (when to jump) and a \emph{jump distribution} (where to jump) via two dedicated network heads. We show that the evidence lower bound (ELBO) reduces to a path-space KL divergence between the true and learned reverse processes that factorizes into a Poisson KL for timing and a categorical KL for direction, and admits a tractable, gradient-equivalent and consistent loss. Experimentally, scored by Gemma2-9B, our pure-uniform Neural CTMC achieves $16.36$ generative perplexity on TinyStories (vs.\ GIDD $37.60$ and MDLM $42.66$). On OpenWebText, it attains the best perplexity at the same training-token budget across 16--128 sampling steps among the methods we compare (e.g., at 128 steps: Neural CTMC $183.6$ vs.\ MDLM $210.5$ and GIDD $249.8$). To facilitate reproducibility, we release our pretrained weights at https://huggingface.co/Jiangxy1117/Neural-CTMC.

cs.LG↗

arXiv2026

Coarse-Grained Dynamics with Spatial Disorder and Non-Markovian Memory

We introduce the spatial disorder-generalized Langevin equation (SD-GLE), a data-driven method for constructing coarse-grained (CG) dynamics in heterogeneous systems. Unlike conventional CG approaches that rely on a mean-field potential, SD-GLE utilizes a variational Bayesian framework with a random field prior to explicitly disentangle static spatial disorder from viscoelastic friction. Numerical results demonstrate the limits of standard GLEs, whereas SD-GLE accurately extrapolates long-time dynamics to capture the anomalous diffusion crossover from short trajectories and recover the ensemble statistical properties inherent to the disordered nature of these systems.

physics.comp-ph↗

arXiv2026

Reasoning as Gradient: Scaling MLE Agents Beyond Tree Search

LLM-based agents for machine learning engineering (MLE) predominantly rely on tree search, a form of gradient-free optimization that uses scalar validation scores to rank candidates. As LLM reasoning capabilities improve, exhaustive enumeration becomes increasingly inefficient compared to directed updates, analogous to how accurate gradients enable efficient descent over random search. We introduce Gome, an MLE agent that operationalizes gradient-based optimization. Gome maps structured diagnostic reasoning to gradient computation, success memory to momentum, and multi-trace execution to distributed optimization. Under a closed-world protocol that isolates architectural effects from external knowledge, Gome achieves a state-of-the-art 35.1\% any-medal rate on MLE-Bench with a restricted 12-hour budget on a single V100 GPU. Scaling experiments across 10 models reveal a critical crossover: with weaker models, tree search retains advantages by compensating for unreliable reasoning through exhaustive exploration; as reasoning capability strengthens, gradient-based optimization progressively outperforms, with the gap widening at frontier-tier models. Given the rapid advancement of reasoning-oriented LLMs, this positions gradient-based optimization as an increasingly favorable paradigm. We release our codebase and GPT-5 traces at https://github.com/microsoft/RD-Agent.

cs.LG↗

arXiv2026

Discovery of Interpretable Physical Laws in Materials via Language-Model-Guided Symbolic Regression

Discovering interpretable physical laws from high-dimensional data is a fundamental challenge in scientific research. Traditional methods, such as symbolic regression, often produce complex, unphysical formulas when searching a vast space of possible forms. We introduce a framework that guides the search process by leveraging the embedded scientific knowledge of large language models, enabling efficient identification of physical laws in the data. We validate our approach by modeling key properties of perovskite materials. Our method mitigates the combinatorial explosion commonly encountered in traditional symbolic regression, reducing the effective search space by a factor of approximately $10^5$. A set of novel formulas for bulk modulus, band gap, and oxygen evolution reaction activity are identified, which not only provide meaningful physical insights but also outperform previous formulas in accuracy and simplicity.

physics.comp-ph↗

arXiv2026

Online Coreset Selection for Learning Dynamic Systems

With the increasing availability of streaming data in dynamic systems, a critical challenge in data-driven modeling for control is how to efficiently select informative data to characterize system dynamics. In this work, we develop an online coreset selection method for set-membership identification in the presence of process disturbances, improving data efficiency while preserving convergence guarantees. Specifically, we derive a stacked polyhedral representation that over-approximates the feasible parameter set. Based on this representation, we propose a geometric selection criterion that retains a data point only if it induces a sufficient contraction of the feasible set. Theoretically, the feasible-set volume is shown to converge to zero almost surely under persistently exciting data and a tight disturbance bound. When the disturbance bound is mismatched, an explicit Hausdorff-distance bound is derived to quantify the resulting identification error. In addition, an upper bound on the expected coreset size is established and extensions to nonlinear systems with linear-in-the-parameter structures and to bounded measurement noise are discussed. The effectiveness of the proposed method is demonstrated through comprehensive simulation studies.

eess.SY↗

arXiv2026

The Interdecadal Bipolar Oscillation: An Atmospheric Water Vapor Mode Driving Asynchronous Polar Climate Change

Climate change is progressing asynchronously between the Arctic and Antarctic, with important implications for global climate dynamics. While the Arctic has experienced rapid warming and pronounced amplification, the Antarctic has exhibited a delayed and heterogeneous response. Here, we identify an Interdecadal Bipolar Oscillation (IBO) in atmospheric water vapor, a coherent internal mode of variability that connects the two polar regions and helps explain their divergent climate trajectories over the past eight decades. Using reanalysis data alongside historical and pre-industrial control simulations from CMIP6, we demonstrate that the IBO is a robust internal variability mode with a quasi-period of 60 ~ 80 years. This oscillation modulates the background warming signal, with a phase shift in the late 1980s amplifying moistening and warming in the Arctic while concurrently suppressing changes in the Antarctic. Crucially, model projections suggest a possible phase reversal, which could slow water vapor increases in the Arctic while accelerating them in the Antarctic, potentially marking the start of rapid Antarctic climate change. While uncertainties remain in how climate models capture polar water vapor, our findings highlight the IBO as a pivotal driver of polar climate evolution and a potential contributor to emerging climate risks in the Southern Hemisphere.

physics.ao-ph↗

arXiv2026

PLANET v2.0: A comprehensive Protein-Ligand Affinity Prediction Model Based on Mixture Density Network

Drug discovery represents a time-consuming and financially intensive process, and virtual screening can accelerate it. Scoring functions, as one of the tools guiding virtual screening, have their precision closely tied to screening efficiency. In our previous study, we developed a graph neural network model called PLANET (Protein-Ligand Affinity prediction NETwork), but it suffers from the defect in representing protein-ligand contact maps. Incorrect binding modes inevitably lead to poor affinity predictions, so accurate prediction of the protein-ligand contact map is desired to improve PLANET. In this study, we have proposed PLANET v2.0 as an upgraded version. The model is trained via multi-objective training strategy and incorporates the Mixture Density Network to predict binding modes. Except for the probability density distributions of non-covalent interactions, we innovatively employ another Gaussian mixture model to describe the relationship between distance and energy of each interaction pair and predict protein-ligand affinity like calculating the mathematical expectation. As on the CASF-2016 benchmark, PLANET v2.0 demonstrates excellent scoring power, ranking power, and docking power. The screening power of PLANET v2.0 gets notably improved compared to PLANET and Glide SP and it demonstrates robust validation on a commercial ultra-large-scale dataset. Given its efficiency and accuracy, PLANET v2.0 can hopefully become one of the practical tools for virtual screening workflows. PLANET v2.0 is freely available at https://www.pdbbind-plus.org.cn/planetv2.

cs.LG↗

arXiv2025

Phase transition revealed by eigen microstate entropy

We introduce the eigen microstate entropy ($S_{\text{EM}}$), a novel metric of complexity derived from the probabilities of statistically independent eigen microstates. After establishing its scaling behavior in equilibrium systems and demonstrating its utility in critical phenomena (mean spherical, Ising, and Potts models), we apply $S_{\text{EM}}$ to non-equilibrium complex systems. Our analysis reveals a consistent precursor signal: a significant increase in $S_{\text{EM}}$ precedes major phase transitions. Specifically, we observe this entropy rise before biomolecular condensate formation in liquid-liquid phase separation in living cells and months ahead of El Niño events. These findings position $S_{\text{EM}}$ as a general framework for detecting and interpreting phase transitions in non-equilibrium systems.

cond-mat.stat-mech↗

arXiv2025

STELLAR-koff: A Transfer Learning Model for Protein-Ligand Dissociation Rate Constant Prediction Based on Interaction Landscape

The key to successful drug design lies in the correct comprehension of protein-ligand interactions. Within the current knowledge paragm, these interactions can be described from both thermodynamic and kinetic perspectives. In recent years, many deep learning models have emerged for predicting the thermodynamic properties of protein-ligand interactions. However, there is currently no mature model for predicting kinetic properties, primarily due to lack of kinetic data. To tackle this problem, we have developed a graph neural network model called STELLAR-koff (Structure-based TransfEr Learning for Ligand Activity Regression) to predict protein-ligand dissociation rate constant. Unlike traditional protein-ligand property prediction models, which typically use a single complex conformation as input, STELLAR-koff employs transfer learning to transform multiple ligand conformations within the protein into a protein ligand interaction landscape, and uses this landscape as the primary input for the model. In addition, we expanded the PDBbind koff dataset from 680 to 1,197 entries and employed the augmented dataset for model training and testing. When tested through five-fold cross-validation, STELLAR-koff achieved Pearson correlation coefficient of 0.729 surpassing or being on pair with most of the published prediction methods. Tested on external set, STELLAR-koff demonstrated strong predictive performance on unseen protein, achieving a Pearson of 0.838 on the focal adhesion kinase in particular. Experimental validation on cyclin-dependent kinase also demonstrated the effectiveness of STELLAR-koff in real drug discovering scenarios. We believe this study provides an effective tool for predicting protein-ligand dissociation rate constant and offers new insight for the future development of this field.

q-bio.QM↗

arXiv2025

Incentivizing Consistent, Effective and Scalable Reasoning Capability in Audio LLMs via Reasoning Process Rewards

The role of reasoning in Audio Large Language Models remains widely underexplored, as introducing a reasoning process often degrades rather than improves performance during inference, a phenomenon we term test-time inverse scaling, where longer reasoning chains yield progressively worse results. We demonstrate that this stems not from fundamental limitations of reasoning itself, but from inadequate training: models without proper guidance for the reasoning process produce hallucinatory, inconsistent reasoning that accumulates errors over longer chains. To address these challenges, we introduce CESAR (Consistent, Effective, and Scalable Audio Reasoners), shifting from outcome verification to rewarding the reasoning process. Our online reinforcement learning framework employs Group Relative Policy Optimization with a multi-faceted reward suite that incentivizes not only correctness and format but also consistency, structured analytical patterns, causal reasoning, domain-knowledge integration, and calibrated reasoning depth. CESAR resolves test-time inverse scaling, transforming reasoning from detriments into gains while revealing model-specific ``reasoning sweet spots", where performance peaks during test-time scaling. We achieve state-of-the-art results on MMAU Test-mini, substantially outperforming Gemini 2.5 Pro and GPT-4o Audio, and near-human-level performance on MMSU reasoning tasks. Through AI-as-judge evaluations and qualitative comparisons, we provide both quantitative and qualitative validation of our improved reasoning quality. Importantly, enhanced reasoning creates synergistic effects, simultaneously improving multimodal reasoning and perception capabilities. Overall, CESAR establishes a principled method for developing robust and scalable reasoning in Audio LLMs.

cs.LG↗

arXiv2025

Evolution of the superconductivity in pressurized La3-xSmxNi2O7

Motivated by the discovery of superconductivity in bilayer La$_3$Ni$_2$O$_7$ at 80 K and the increased superconducting transition temperature, $T_\text{c}$, up to 92 K in single crystals of La$_2$SmNi$_2$O$_7$ under pressure, we systematically study the effect of Sm doping on the superconductivity and structure of La$_{3-x}$Sm$_x$Ni$_2$O$_7$ (0 $\leq$ x $\leq$ 1.5) under pressure. Experimental investigations in polycrystalline samples reveal that Sm doping monotonically decreases the lattice constants $c$ and $a$, thereby enhancing crystal structure distortion and leading to an evolution of the metallic ground state in La$_3$Ni$_2$O$_7$ to an insulating state in La$_{1.5}$Sm$_{1.5}$Ni$_2$O$_7$. The maximum onset $T_\text{c}$ in compounds $x=0.9$ and 1.5 is 89 K, while the pressure that drives the emergence of superconductivity is higher for higher doping levels. The results suggest that the enhancement of $T_\text{c}$ in La$_{3-x}$Sm$_x$Ni$_2$O$_7$ is mainly affected by the compressed $c$ lattice before saturation, and the structure transition is critical for the emergence of superconductivity. Our experimental results provide insight into the influence of elemental substitution on nickelate superconductors, offering a means to increase the transition temperature further.

cond-mat.supr-con↗

arXiv2025

Interlayer coupling enhanced superconductivity near 100 K in La$_{3-x}$Nd$_x$Ni$_2$O$_7$

Systematically controlling the superconducting transition temperature ($T_\text{c}$) in the bilayer Ruddlesden-Popper nickelate La$_3$Ni$_2$O$_7$ remains a significant challenge. Here, we address this by synthesizing high-quality polycrystalline La$_{3-x}$Nd$_x$Ni$_2$O$_7$ ($0 \leq x \leq 2.4$) with record-level rare-earth substitution. Nd doping compresses the lattice, particularly along the $c$ axis, enhances the spin density wave transition temperature, and elevates the pressure required for the orthorhombic-to-tetragonal structural transition. Superconductivity is observed across all doping levels under high pressures, with the onset $T_\text{c}$ rising to $\sim$93~K for $x = 2.1$ and $2.4$ from the electronic transport measurement. Using the radio-frequency transmission technique, newly applied to nickelate superconductors, we detect signatures of superconductivity at $98 \pm 2$~K in the $x=2.4$ compound, pushing the $T_\text{c}$ frontier further. We identify a universal linear relationship where $T_\text{c}$ decreases with the $c$-axis lattice parameter at a rate of approximately $-28$~K/Å, demonstrating that enhanced interlayer magnetic exchange coupling is the dominant mechanism for superconducting pairing. Our work establishes the critical role of magnetism and provides a unified structural descriptor for elevating $T_\text{c}$ in bilayer nickelates.

cond-mat.supr-con↗

arXiv2025

Superconductivity in monolayer-trilayer phase of La$_3$Ni$_2$O$_7$ under high pressure

The discovery of 80 K superconductivity in pressurized bilayer Ruddlesden-Popper (RP) nickelate La$_3$Ni$_2$O$_7$ has established a new high-temperature superconductor family. The quest to understand the governing principles of RP nickelate superconductivity has become a central focus in condensed matter physics. Here, we report a critical advance by synthesizing and investigating a distinct structural polymorph of the same compound: the monolayer-trilayer (1313) hybrid phase of La$_3$Ni$_2$O$_7$. Under high pressure, synchrotron X-ray diffraction and Raman spectroscopy reveal a structural transition from the orthorhombic $Cmmm$ to the tetragonal $P4/mmm$ space group at 13~GPa. Above 19 GPa, the phase exhibits a clear superconducting transition, confirmed by a zero-resistance state, albeit at a significantly reduced temperature of 3.6 K. The stark contrast with the 80 K transition in the bilayer phase provides a uniquely clean experimental comparison. Our results demonstrate that the superconducting transition temperature is directly governed by the nature of the interlayer coupling, and the bilayer NiO$_6$ block as the essential structural motif for achieving high-$T_\text{c}$ superconductivity in the RP nickelates.

cond-mat.supr-con↗

arXiv2025

R&D-Agent: An LLM-Agent Framework Towards Autonomous Data Science

Recent advances in AI and ML have transformed data science, yet increasing complexity and expertise requirements continue to hinder progress. Although crowd-sourcing platforms alleviate some challenges, high-level machine learning engineering (MLE) tasks remain labor-intensive and iterative. We introduce R&D-Agent, a comprehensive, decoupled, and extensible framework that formalizes the MLE process. R&D-Agent defines the MLE workflow into two phases and six components, turning agent design for MLE from ad-hoc craftsmanship into a principled, testable process. Although several existing agents report promising gains on their chosen components, they can mostly be summarized as a partial optimization from our framework's simple baseline. Inspired by human experts, we designed efficient and effective agents within this framework that achieve state-of-the-art performance. Evaluated on MLE-Bench, the agent built on R&D-Agent ranks as the top-performing machine learning engineering agent, achieving 35.1% any medal rate, demonstrating the ability of the framework to speed up innovation and improve accuracy across a wide range of data science applications. We have open-sourced R&D-Agent on GitHub: https://github.com/microsoft/RD-Agent.

cs.AI↗

arXiv2025

Can Optical Denoising Clean Sonar Images? A Benchmark and Fusion Approach

Object detection in sonar images is crucial for underwater robotics applications including autonomous navigation and resource exploration. However, complex noise patterns inherent in sonar imagery, particularly speckle, reverberation, and non-Gaussian noise, significantly degrade detection accuracy. While denoising techniques have achieved remarkable success in optical imaging, their applicability to sonar data remains underexplored. This study presents the first systematic evaluation of nine state-of-the-art deep denoising models with distinct architectures, including Neighbor2Neighbor with varying noise parameters, Blind2Unblind with different noise configurations, and DSPNet, for sonar image preprocessing. We establish a rigorous benchmark using five publicly available sonar datasets and assess their impact on four representative detection algorithms: YOLOX, Faster R-CNN, SSD300, and SSDMobileNetV2. Our evaluation addresses three unresolved questions: first, how effectively optical denoising architectures transfer to sonar data; second, which model families perform best against sonar noise; and third, whether denoising truly improves detection accuracy in practical pipelines. Extensive experiments demonstrate that while denoising generally improves detection performance, effectiveness varies across methods due to their inherent biases toward specific noise types. To leverage complementary denoising effects, we propose a mutually-supervised multi-source denoising fusion framework where outputs from different denoisers mutually supervise each other at the pixel level, creating a synergistic framework that produces cleaner images.

cs.CV↗

arXiv2025

SPINT: Spatial Permutation-Invariant Neural Transformer for Consistent Intracortical Motor Decoding

Intracortical Brain-Computer Interfaces (iBCI) aim to decode behavior from neural population activity, enabling individuals with motor impairments to regain motor functions and communication abilities. A key challenge in long-term iBCI is the nonstationarity of neural recordings, where the composition and tuning profiles of the recorded populations are unstable across recording sessions. Existing methods attempt to address this issue by explicit alignment techniques; however, they rely on fixed neural identities and require test-time labels or parameter updates, limiting their generalization across sessions and imposing additional computational burden during deployment. In this work, we introduce SPINT - a Spatial Permutation-Invariant Neural Transformer framework for behavioral decoding that operates directly on unordered sets of neural units. Central to our approach is a novel context-dependent positional embedding scheme that dynamically infers unit-specific identities, enabling flexible generalization across recording sessions. SPINT supports inference on variable-size populations and allows few-shot, gradient-free adaptation using a small amount of unlabeled data from the test session. To further promote model robustness to population variability, we introduce dynamic channel dropout, a regularization method for iBCI that simulates shifts in population composition during training. We evaluate SPINT on three multi-session datasets from the FALCON Benchmark, covering continuous motor decoding tasks in human and non-human primates. SPINT demonstrates robust cross-session generalization, outperforming existing zero-shot and few-shot unsupervised baselines while eliminating the need for test-time alignment and fine-tuning. Our work contributes an initial step toward a robust and scalable neural decoding framework for long-term iBCI applications.

q-bio.NC↗