arXiv subjects

Ying Chen

Publications and source records attributed to Ying Chen.

At least 109 records · Page 6Linked to original sources

arXiv2025

HyperST: Hierarchical Hyperbolic Learning for Spatial Transcriptomics Prediction

Spatial Transcriptomics (ST) merges the benefits of pathology images and gene expression, linking molecular profiles with tissue structure to analyze spot-level function comprehensively. Predicting gene expression from histology images is a cost-effective alternative to expensive ST technologies. However, existing methods mainly focus on spot-level image-to-gene matching but fail to leverage the full hierarchical structure of ST data, especially on the gene expression side, leading to incomplete image-gene alignment. Moreover, a challenge arises from the inherent information asymmetry: gene expression profiles contain more molecular details that may lack salient visual correlates in histological images, demanding a sophisticated representation learning approach to bridge this modality gap. We propose HyperST, a framework for ST prediction that learns multi-level image-gene representations by modeling the data's inherent hierarchy within hyperbolic space, a natural geometric setting for such structures. First, we design a Multi-Level Representation Extractors to capture both spot-level and niche-level representations from each modality, providing context-aware information beyond individual spot-level image-gene pairs. Second, a Hierarchical Hyperbolic Alignment module is introduced to unify these representations, performing spatial alignment while hierarchically structuring image and gene embeddings. This alignment strategy enriches the image representations with molecular semantics, significantly improving cross-modal prediction. HyperST achieves state-of-the-art performance on four public datasets from different tissues, paving the way for more scalable and accurate spatial transcriptomics prediction.

cs.CV↗

arXiv2025

Explore More, Learn Better: Parallel MLLM Embeddings under Mutual Information Minimization

Embedding models are a cornerstone of modern AI. Driven by Multimodal Large Language Models (MLLMs), they have made great progress in architecture and data curation, while the holistic paradigm is still limited to SSC, i.e., single input, singular embedding, contrastive supervision, which collapses rich, multifaceted inputs into monolithic embeddings and fails to fully exploit MLLM capabilities. In this paper, we tailor one Parallel Decoupling Framework (PDF) for multimodal embedding learning, by utilizing the proprietary steerability of MLLMs, i.e., their ability to flexibly generate quite differentiated response under explicit instructions. Concretely, PDF conditions a shared MLLM backbone on distinct, learnable prefixes to roll out multiple parallel paths for one input, then relies on these paths to obtain parallel embeddings. To promote full parallel diversity, we employ Mutual Information Minimization (MIM) as an explicit constraint, coupled with per-path contrastive supervision to maintain semantic alignment. Such dual-objectives force PDF to yield robust semantic coverage and a generalizable embedding space. Ultimately, the remarkable embedding space are accessible at inference via one single forward pass, incurring negligible computational overhead. We instantiate PDF on multiple MLLM backbones and prove its effectiveness on MMEB benchmark. Significant gains are consistently achieved across various resolutions and model sizes, e.g., boosting the VLM2Vec-LLaVA-1.6-LR model by a remarkable +8.9% (7B), while the VLM2Vec-Qwen2VL models by +4.2% (2B) and +3.1% (7B). In terms of efficiency, our 2B model surpasses its baseline by +2.6% using only half the computational budget.

cs.LG↗

arXiv2025

PISanitizer: Preventing Prompt Injection to Long-Context LLMs via Prompt Sanitization

Long context LLMs are vulnerable to prompt injection, where an attacker can inject an instruction in a long context to induce an LLM to generate an attacker-desired output. Existing prompt injection defenses are designed for short contexts. When extended to long-context scenarios, they have limited effectiveness. The reason is that an injected instruction constitutes only a very small portion of a long context, making the defense very challenging. In this work, we propose PISanitizer, which first pinpoints and sanitizes potential injected tokens (if any) in a context before letting a backend LLM generate a response, thereby eliminating the influence of the injected instruction. To sanitize injected tokens, PISanitizer builds on two observations: (1) prompt injection attacks essentially craft an instruction that compels an LLM to follow it, and (2) LLMs intrinsically leverage the attention mechanism to focus on crucial input tokens for output generation. Guided by these two observations, we first intentionally let an LLM follow arbitrary instructions in a context and then sanitize tokens receiving high attention that drive the instruction-following behavior of the LLM. By design, PISanitizer presents a dilemma for an attacker: the more effectively an injected instruction compels an LLM to follow it, the more likely it is to be sanitized by PISanitizer. Our extensive evaluation shows that PISanitizer can successfully prevent prompt injection, maintain utility, outperform existing defenses, is efficient, and is robust to optimization-based and strong adaptive attacks. The code is available at https://github.com/sleeepeer/PISanitizer.

cs.CR↗

arXiv2025

Robust Subgroup Method Using DE Algorithm for Resonance Self-Shielding Calculation

This paper presents an enhanced version of the subgroup method for resonance self-shielding treatment, termed the robust subgroup method, which integrates Robust Estimation (RE) with a Differential Evolution (DE) algorithm. The RE approach is employed to handle model misspecification and data contamination, while the DE algorithm serves as an optimization tool within the RE framework to obtain constrained solutions. Numerical validation against experimental benchmarks shows that the proposed method removes a systematic absorption bias in conventional subgroup fits that would otherwise depress reactivity. This bias appears only in benchmarks sensitive to U-238. Mechanistically, it reflects a threshold-like conditioning failure: strong self-shielding leverage dominates the loss and is magnified by dilution-induced multicollinearity. This adverse conditioning appears to be seeded by a narrow, sparse resonance structure at low energies in fertile even-even nuclides, thereby causing rapid self-shielding response saturation and a weak Doppler broadening. By bounding influence and enforcing feasibility within an RE-DE framework, the inferred subgroup parameters track the underlying physics more faithfully, improving the predictive fidelity of subsequent transport simulations.

physics.comp-ph↗

arXiv2025

Wave-Particle (Continuous-Discrete) Dualistic Visual Tokenization for Unified Understanding and Generation

The unification of understanding and generation within a single multi-modal large model (MLLM) remains one significant challenge, largely due to the dichotomy between continuous and discrete visual tokenizations. Continuous tokenizer (CT) achieves strong performance by bridging multiple independently-trained understanding modules and generation modules, but suffers from complex multi-stage pipelines and substantial engineering overhead. Conversely, discrete tokenizers (DT) offer a conceptually elegant idea by quantizing each image into a primitive, but inevitably leading to information loss and performance degradation. To resolve this tension, we question the binary choice between CT and DT, inspired by the wave-particle duality of light, and propose the Continuous-Discrete Dualistic Visual Tokenizer (CDD-VT). We treat visual data as a flexible composition of image primitives derived from quantized codebooks, with the crucial insight that the primitive number assigned to each visual sample is adaptively determined according to its complexity: simple instances use a few primitives, emulating discrete tokenization, while complex instances use many, approximating continuous tokenization. Two core components are designed: Diverse Quantitative Primitives, which encourage primitives orthogonality to better populate information space, and Dynamic Primitive Allocator, which assesses sample complexity to determine the optimal set of primitives. Extensive experiments on reconstruction, retrieval and classification show that CDD-VT achieves superior performance over to specialized CT and DT, effectively getting strong result within a concise and scalable MLLM.

cs.CV↗

arXiv2025

Rest-frame Optical Spectroscopy of $z \sim 2$ Quasars with Steep Hard X-ray Spectral Shapes: X-ray Selection of Super-Eddington Accretion and Verification

Super-Eddington accretion is a crucial phase in the growth of supermassive black holes. However, identifying super-Eddington accreting quasars observationally is challenging due to uncertain black-hole mass estimates and other complications. The Eddington ratio parameter does not represent accurately the accretion rate in the super-Eddington regime. On the other hand, super-Eddington accreting quasars appear to show large hard X-ray (rest-frame > 2 keV) power-law photon indices, and they also exhibit distinct optical spectral features including weak [O III] $λ5007$ emission and strong Fe II emission. We test steep X-ray photon-index selection of super-Eddington accreting quasars by obtaining Palomar 200-inch Hale Telescope near-infrared spectra for a pilot sample of nine $Γ=2.0-2.6$ quasars at $z\approx1.4-2.5$. We derive H$β$-based single-epoch virial black-hole masses (median value $\rm 4.3 \times 10^{8}~M_{\odot}$) and Eddington ratios (median value 0.6). The Eddington ratio distribution is consistent with that of the comparison sample, which is a flux-limited sample of quasars at $z\approx1.5-3.5$ with near-infrared spectroscopy. But our super-Eddington candidates do show statistically weaker [O III] emission ($P_{\rm null}=0.0075$) and marginally stronger Fe II emission ($P_{\rm null}=0.06$). We also find one candidate with broad (width of 1960 km/s) and blueshifted (690 km/s) [O III] $λ4959$ and [O III] $λ5007$ lines, which probably originate from a strong [O III] outflow driven by super-Eddington accretion. Overall, the steep X-ray photon-index selection of super-Eddington accreting quasars appears promising. But a larger sample is needed to assess further the reliability of the selection.

astro-ph.HE↗

arXiv2025

A Survey of Scientific Large Language Models: From Data Foundations to Agent Frontiers

Scientific Large Language Models (Sci-LLMs) are transforming how knowledge is represented, integrated, and applied in scientific research, yet their progress is shaped by the complex nature of scientific data. This survey presents a comprehensive, data-centric synthesis that reframes the development of Sci-LLMs as a co-evolution between models and their underlying data substrate. We formulate a unified taxonomy of scientific data and a hierarchical model of scientific knowledge, emphasizing the multimodal, cross-scale, and domain-specific challenges that differentiate scientific corpora from general natural language processing datasets. We systematically review recent Sci-LLMs, from general-purpose foundations to specialized models across diverse scientific disciplines, alongside an extensive analysis of over 270 pre-/post-training datasets, showing why Sci-LLMs pose distinct demands -- heterogeneous, multi-scale, uncertainty-laden corpora that require representations preserving domain invariance and enabling cross-modal reasoning. On evaluation, we examine over 190 benchmark datasets and trace a shift from static exams toward process- and discovery-oriented assessments with advanced evaluation protocols. These data-centric analyses highlight persistent issues in scientific data development and discuss emerging solutions involving semi-automated annotation pipelines and expert validation. Finally, we outline a paradigm shift toward closed-loop systems where autonomous agents based on Sci-LLMs actively experiment, validate, and contribute to a living, evolving knowledge base. Collectively, this work provides a roadmap for building trustworthy, continually evolving artificial intelligence (AI) systems that function as a true partner in accelerating scientific discovery.

cs.CL↗

arXiv2025

Spinons, solitons and random singlets in the spin-chain compound copper benzoate

The $S=1/2$ antiferromagnetic Heisenberg chain is a paradigmatic quantum system hosting exotic excitations such as spinons and solitons, and forming random singlet state in the presence of quenched disorder. Realizing and distinguishing these excitations in a single material remains a significant challenge. Using nuclear magnetic resonance (NMR) on a high-quality single crystal of copper benzoate, we identify and characterize all three excitation types by tuning the magnetic field at ultra-low temperatures. At a low field of 0.2 T, a temperature-independent spin-lattice relaxation rate ($1/T_1$) over more than a decade confirms the presence of spinons. Below 0.4 K, an additional relaxation channel emerges, characterized by $1/T_1 \propto T$ and a spectral weight growing as $-\ln(T/T_0)$, signaling a random-singlet ground state induced by weak quenched disorder. At fields above 0.5 T, a field-induced spin gap $Δ\propto H^{2/3}$ observed in both $1/T_1$ and the Knight shift signifies soliton excitations. Our results establish copper benzoate as a unique experimental platform for studying one-dimensional quantum integrability and the interplay of disorder and correlations.

cond-mat.str-el↗

arXiv2025

Implementation and Performance Evaluation of TCP over QUIC Tunnels

QUIC, a UDP-based transport protocol, addresses several limitations of TCP by offering built-in encryption, stream multiplexing, and improved loss recovery. To extend these benefits to legacy TCP-based applications, this paper explores the implementation and evaluation of a TCP over QUIC tunneling approach. A lightweight, stream-based tunnel is constructed using the Rust-based Quinn library, enabling TCP traffic to traverse QUIC connections transparently. Performance is evaluated under varying network conditions, including packet loss, high latency, and out-of-order delivery. Results indicate that TCP over QUIC maintains significantly higher throughput than native TCP in lossy or unstable environments, with up to a high improvement under 20\% packet loss. However, under ideal network conditions, tunneling introduces modest overhead due to encryption and user-space processing. These findings provide insights into the trade-offs of TCP over QUIC tunneling and its suitability for deployment in dynamic or impaired networks.

cs.NI↗

arXiv2025

MedQ-Bench: Evaluating and Exploring Medical Image Quality Assessment Abilities in MLLMs

Medical Image Quality Assessment (IQA) serves as the first-mile safety gate for clinical AI, yet existing approaches remain constrained by scalar, score-based metrics and fail to reflect the descriptive, human-like reasoning process central to expert evaluation. To address this gap, we introduce MedQ-Bench, a comprehensive benchmark that establishes a perception-reasoning paradigm for language-based evaluation of medical image quality with Multi-modal Large Language Models (MLLMs). MedQ-Bench defines two complementary tasks: (1) MedQ-Perception, which probes low-level perceptual capability via human-curated questions on fundamental visual attributes; and (2) MedQ-Reasoning, encompassing both no-reference and comparison reasoning tasks, aligning model evaluation with human-like reasoning on image quality. The benchmark spans five imaging modalities and over forty quality attributes, totaling 2,600 perceptual queries and 708 reasoning assessments, covering diverse image sources including authentic clinical acquisitions, images with simulated degradations via physics-based reconstructions, and AI-generated images. To evaluate reasoning ability, we propose a multi-dimensional judging protocol that assesses model outputs along four complementary axes. We further conduct rigorous human-AI alignment validation by comparing LLM-based judgement with radiologists. Our evaluation of 14 state-of-the-art MLLMs demonstrates that models exhibit preliminary but unstable perceptual and reasoning skills, with insufficient accuracy for reliable clinical use. These findings highlight the need for targeted optimization of MLLMs in medical IQA. We hope that MedQ-Bench will catalyze further exploration and unlock the untapped potential of MLLMs for medical image quality evaluation.

cs.CV↗

arXiv2025

Search for Active and Inactive Ion Insertion Sites in Organic Crystalline Materials

The position of mobile active and inactive ions, specifically ion insertion sites, within organic crystals, significantly affects the properties of organic materials used for energy storage and ionic transport. Identifying the positions of these atom (and ion) sites in an organic crystal is difficult, especially when the element has a low X-ray scattering power, such as lithium (Li) and hydrogen, which are difficult to detect with powder X-ray diffraction (XRD) methods. First-principles calculations, exemplified by density functional theory (DFT), are very effective for confirming the relative stability of ion positions in materials. However, the lack of effective strategies to identify ion sites in these organic crystalline frameworks renders this task extremely challenging. This work presents two algorithms: (i) Efficient Location of Ion Insertion Sites from Extrema in electrostatic local potential and charge density (ELIISE), and (ii) ElectRostatic InsertioN (ERIN), which leverage charge density and electrostatic potential fields accessed from first-principles calculations, combined with the Simultaneous Ion Insertion and Evaluation (SIIE) workflow -- that inserts all ions simultaneously -- to determine ion positions in organic crystals. We demonstrate that these methods accurately reproduce known ion positions in 16 organic materials and also identify previously overlooked low-energy sites in tetralithium 2,6-naphthalenedicarboxylate (Li$_4$NDC), an organic electrode material, highlighting the importance of inserting all ions simultaneously as done in the SIIE workflow.

cond-mat.mtrl-sci↗

arXiv2025

Development and validation of an AI foundation model for endoscopic diagnosis of esophagogastric junction adenocarcinoma: a cohort and deep learning study

The early detection of esophagogastric junction adenocarcinoma (EGJA) is crucial for improving patient prognosis, yet its current diagnosis is highly operator-dependent. This paper aims to make the first attempt to develop an artificial intelligence (AI) foundation model-based method for both screening and staging diagnosis of EGJA using endoscopic images. In this cohort and learning study, we conducted a multicentre study across seven Chinese hospitals between December 28, 2016 and December 30, 2024. It comprises 12,302 images from 1,546 patients; 8,249 of them were employed for model training, while the remaining were divided into the held-out (112 patients, 914 images), external (230 patients, 1,539 images), and prospective (198 patients, 1,600 images) test sets for evaluation. The proposed model employs DINOv2 (a vision foundation model) and ResNet50 (a convolutional neural network) to extract features of global appearance and local details of endoscopic images for EGJA staging diagnosis. Our model demonstrates satisfactory performance for EGJA staging diagnosis across three test sets, achieving an accuracy of 0.9256, 0.8895, and 0.8956, respectively. In contrast, among representative AI models, the best one (ResNet50) achieves an accuracy of 0.9125, 0.8382, and 0.8519 on the three test sets, respectively; the expert endoscopists achieve an accuracy of 0.8147 on the held-out test set. Moreover, with the assistance of our model, the overall accuracy for the trainee, competent, and expert endoscopists improves from 0.7035, 0.7350, and 0.8147 to 0.8497, 0.8521, and 0.8696, respectively. To our knowledge, our model is the first application of foundation models for EGJA staging diagnosis and demonstrates great potential in both diagnostic accuracy and efficiency.

cs.CV↗

arXiv2025

Bulk Ferroelectric Heterostructures: Imprinted Actuators

Domain switching is the cornerstone of ferroelectric materials. Most associated functionalities can be tuned via domain switching, including but not limited to piezoelectricity, thermal conductivity, domain wall conductivity and topological structures. However, achieving the full potential of reversible ferroelectric domain switching is restricted by the incomplete access to the entire ferroelectric texture space, as well as the memory effects and energy dissipation associated with the hysteretic nature of ferroelectrics. The manipulation of domain switching behaviour is moderately attainable in epitaxial heterostructures by exploiting the valence or lattice mismatch at heterointerfaces, which is generally constrained by the necessity for two dimensional architectures. In this study, domain-engineered bulk ferroelectric heterostructures (DE-BFH), constructed via elemental partitioning, are employed to unleash full potential of bulk ferroelectrics, providing comprehensive control of domain switching characteristics and adjustable reversibility within the entire range of ferroelectric texture space. Exemplar DE-BFH ceramics exhibit unprecedented enhancement in reversible electrostrain and stability in both axial and shear modes, including a record high peak to peak shear strain up to 0.9% at intermediate field levels, confirmed by digital image correlation measurements and in-situ synchrotron XRD studies. The advancement of domain switching behaviour in DE-BFH could also promote development of new types of lead-free piezoelectric devices, including actuators, energy harvesters, multiple state memory devices, and domain wall switch. Moreover, design concept of DE-BFH could contribute to the creation of distinctive ferroelastic, ferromagnetic, and multiferroic materials by broadening its scope to the entire ferroic family, encompassing polycrystalline, single-crystal, and thin-film forms.

cond-mat.mtrl-sci↗

arXiv2025

Towards Interpretable Counterfactual Generation via Multimodal Autoregression

Counterfactual medical image generation enables clinicians to explore clinical hypotheses, such as predicting disease progression, facilitating their decision-making. While existing methods can generate visually plausible images from disease progression prompts, they produce silent predictions that lack interpretation to verify how the generation reflects the hypothesized progression -- a critical gap for medical applications that require traceable reasoning. In this paper, we propose Interpretable Counterfactual Generation (ICG), a novel task requiring the joint generation of counterfactual images that reflect the clinical hypothesis and interpretation texts that outline the visual changes induced by the hypothesis. To enable ICG, we present ICG-CXR, the first dataset pairing longitudinal medical images with hypothetical progression prompts and textual interpretations. We further introduce ProgEmu, an autoregressive model that unifies the generation of counterfactual images and textual interpretations. We demonstrate the superiority of ProgEmu in generating progression-aligned counterfactuals and interpretations, showing significant potential in enhancing clinical decision support and medical education. Project page: https://progemu.github.io.

eess.IV↗

arXiv2025

CRMAgent: A Multi-Agent LLM System for E-Commerce CRM Message Template Generation

In e-commerce private-domain channels such as instant messaging and e-mail, merchants engage customers directly as part of their Customer Relationship Management (CRM) programmes to drive retention and conversion. While a few top performers excel at crafting outbound messages, most merchants struggle to write persuasive copy because they lack both expertise and scalable tools. We introduce CRMAgent, a multi-agent system built on large language models (LLMs) that generates high-quality message templates and actionable writing guidance through three complementary modes. First, group-based learning enables the agent to learn from a merchant's own top-performing messages within the same audience segment and rewrite low-performing ones. Second, retrieval-and-adaptation fetches templates that share the same audience segment and exhibit high similarity in voucher type and product category, learns their successful patterns, and adapts them to the current campaign. Third, a rule-based fallback provides a lightweight zero-shot rewrite when no suitable references are available. Extensive experiments show that CRMAgent consistently outperforms merchants' original templates, delivering significant gains in both audience-match and marketing-effectiveness metrics.

cs.CL↗

arXiv2025

Quantum Optimization Benchmarking Library - The Intractable Decathlon

Through recent progress in hardware development, quantum computers have advanced to the point where benchmarking of (heuristic) quantum algorithms at scale is within reach. Particularly in combinatorial optimization - where most algorithms are heuristics - it is key to empirically analyze their performance on hardware and track progress towards quantum advantage. To this extent, we present ten optimization problem classes that are difficult for existing classical algorithms and can (mostly) be linked to practically relevant applications, with the goal to enable systematic, fair, and comparable benchmarks for quantum optimization methods. Further, we introduce the Quantum Optimization Benchmarking Library (QOBLIB) where the problem instances and solution track records can be found. The individual properties of the problem classes vary in terms of objective and variable type, coefficient ranges, and density. Crucially, they all become challenging for established classical methods already at system sizes ranging from less than 100 to, at most, an order of 100,000 decision variables, allowing to approach them with today's quantum computers. We reference the results from state-of-the-art solvers for instances from all problem classes and demonstrate exemplary baseline results obtained with quantum solvers for selected problems. The baseline results illustrate a standardized form to present benchmarking solutions, which has been designed to ensure comparability of the used methods, reproducibility of the respective results, and trackability of algorithmic and hardware improvements over time. We encourage the optimization community to explore the performance of available classical or quantum algorithms and hardware platforms with the benchmarking problem instances presented in this work toward demonstrating quantum advantage in optimization.

quant-ph↗

arXiv2025

UniC-RAG: Universal Knowledge Corruption Attacks to Retrieval-Augmented Generation

Retrieval-augmented generation (RAG) systems are widely deployed in real-world applications in diverse domains such as finance, healthcare, and cybersecurity. However, many studies showed that they are vulnerable to knowledge corruption attacks, where an attacker can inject adversarial texts into the knowledge database of a RAG system to induce the LLM to generate attacker-desired outputs. Existing studies mainly focus on attacking specific queries or queries with similar topics (or keywords). In this work, we propose UniC-RAG, a universal knowledge corruption attack against RAG systems. Unlike prior work, UniC-RAG jointly optimizes a small number of adversarial texts that can simultaneously attack a large number of user queries with diverse topics and domains, enabling an attacker to achieve various malicious objectives, such as directing users to malicious websites, triggering harmful command execution, or launching denial-of-service attacks. We formulate UniC-RAG as an optimization problem and further design an effective solution to solve it, including a balanced similarity-based clustering method to enhance the attack's effectiveness. Our extensive evaluations demonstrate that UniC-RAG is highly effective and significantly outperforms baselines. For instance, UniC-RAG could achieve over 90% attack success rate by injecting 100 adversarial texts into a knowledge database with millions of texts to simultaneously attack a large set of user queries (e.g., 2,000). Additionally, we evaluate existing defenses and show that they are insufficient to defend against UniC-RAG, highlighting the need for new defense mechanisms in RAG systems.

cs.CR↗

arXiv2025

Intern-S1: A Scientific Multimodal Foundation Model

In recent years, a plethora of open-source foundation models have emerged, achieving remarkable progress in some widely attended fields, with performance being quite close to that of closed-source models. However, in high-value but more challenging scientific professional fields, either the fields still rely on expert models, or the progress of general foundation models lags significantly compared to those in popular areas, far from sufficient for transforming scientific research and leaving substantial gap between open-source models and closed-source models in these scientific domains. To mitigate this gap and explore a step further toward Artificial General Intelligence (AGI), we introduce Intern-S1, a specialized generalist equipped with general understanding and reasoning capabilities with expertise to analyze multiple science modal data. Intern-S1 is a multimodal Mixture-of-Experts (MoE) model with 28 billion activated parameters and 241 billion total parameters, continually pre-trained on 5T tokens, including over 2.5T tokens from scientific domains. In the post-training stage, Intern-S1 undergoes offline and then online reinforcement learning (RL) in InternBootCamp, where we propose Mixture-of-Rewards (MoR) to synergize the RL training on more than 1000 tasks simultaneously. Through integrated innovations in algorithms, data, and training systems, Intern-S1 achieved top-tier performance in online RL training. On comprehensive evaluation benchmarks, Intern-S1 demonstrates competitive performance on general reasoning tasks among open-source models and significantly outperforms open-source models in scientific domains, surpassing closed-source state-of-the-art models in professional tasks, such as molecular synthesis planning, reaction condition prediction, predicting thermodynamic stabilities for crystals. Our models are available at https://huggingface.co/internlm/Intern-S1.

cs.LG↗