arXiv subjects

Linyang Li

Publications and source records attributed to Linyang Li.

At least 37 records · Page 2Linked to original sources

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↗

arXiv2025

FastMCTS: A Simple Sampling Strategy for Data Synthesis

Synthetic high-quality multi-step reasoning data can significantly enhance the performance of large language models on various tasks. However, most existing methods rely on rejection sampling, which generates trajectories independently and suffers from inefficiency and imbalanced sampling across problems of varying difficulty. In this work, we introduce FastMCTS, an innovative data synthesis strategy inspired by Monte Carlo Tree Search. FastMCTS provides a more efficient sampling method for multi-step reasoning data, offering step-level evaluation signals and promoting balanced sampling across problems of different difficulty levels. Experiments on both English and Chinese reasoning datasets demonstrate that FastMCTS generates over 30\% more correct reasoning paths compared to rejection sampling as the number of generated tokens scales up. Furthermore, under comparable synthetic data budgets, models trained on FastMCTS-generated data outperform those trained on rejection sampling data by 3.9\% across multiple benchmarks. As a lightweight sampling strategy, FastMCTS offers a practical and efficient alternative for synthesizing high-quality reasoning data. Our code will be released soon.

cs.CL↗

arXiv2025

Implicit Reward as the Bridge: A Unified View of SFT and DPO Connections

Post-training processes are essential phases in grounding pre-trained language models to real-world tasks, with learning from demonstrations or preference signals playing a crucial role in this adaptation. We present a unified theoretical framework bridging Supervised Fine-Tuning (SFT) and preference learning in Large Language Model (LLM) post-training. Through rigorous mathematical derivation, we demonstrate that both SFT and preference learning methods like Direct Preference Optimization (DPO) operate within the same optimal policy-reward subspace, with SFT representing a special case of implicit reward learning. Our analysis reveals a critical limitation in conventional SFT: the KL divergence term in distribution matching becomes constant with respect to the policy during optimization, failing to constrain model updates. To address this, we propose a simple yet effective learning rate reduction approach that yields significant performance improvements (up to \textbf{25\%} relative gain and \textbf{6\%} absolute win rate increase in instruction following tasks. Additionally, we derive alternative SFT objectives from various f-divergence functions that preserve the KL term during optimization, further enhancing post-DPO model performance. Finally, we extend the theoretical relationship between LLM logits and Q-functions from preference learning to the SFT context, providing mathematical derivations and experimental validation.

cs.LG↗

arXiv2025

Unearthing Large Scale Domain-Specific Knowledge from Public Corpora

Large language models (LLMs) have demonstrated remarkable potential in various tasks, however, there remains a significant lack of open-source models and data for specific domains. Previous work has primarily focused on manually specifying resources and collecting high-quality data for specific domains, which is extremely time-consuming and labor-intensive. To address this limitation, we introduce large models into the data collection pipeline to guide the generation of domain-specific information and retrieve relevant data from Common Crawl (CC), a large public corpus. We refer to this approach as Retrieve-from-CC. It not only collects data related to domain-specific knowledge but also mines the data containing potential reasoning procedures from the public corpus. By applying this method, we have collected a knowledge domain-related dataset named Retrieve-Pile, which covers four main domains, including the sciences, humanities, and other categories. Through the analysis of , Retrieve-from-CC can effectively retrieve relevant data from the covered knowledge domains and significantly improve the performance in tests of mathematical and knowledge-related reasoning abilities. We have released Retrieve-Pile at https://huggingface.co/datasets/Query-of-CC/Retrieve-Pile.

cs.CL↗

arXiv2025

Super-Resolution on Rotationally Scanned Photoacoustic Microscopy Images Incorporating Scanning Prior

Photoacoustic Microscopy (PAM) images integrating the advantages of optical contrast and acoustic resolution have been widely used in brain studies. However, there exists a trade-off between scanning speed and image resolution. Compared with traditional raster scanning, rotational scanning provides good opportunities for fast PAM imaging by optimizing the scanning mechanism. Recently, there is a trend to incorporate deep learning into the scanning process to further increase the scanning speed.Yet, most such attempts are performed for raster scanning while those for rotational scanning are relatively rare. In this study, we propose a novel and well-performing super-resolution framework for rotational scanning-based PAM imaging. To eliminate adjacent rows' displacements due to subject motion or high-frequency scanning distortion,we introduce a registration module across odd and even rows in the preprocessing and incorporate displacement degradation in the training. Besides, gradient-based patch selection is proposed to increase the probability of blood vessel patches being selected for training. A Transformer-based network with a global receptive field is applied for better performance. Experimental results on both synthetic and real datasets demonstrate the effectiveness and generalizability of our proposed framework for rotationally scanned PAM images'super-resolution, both quantitatively and qualitatively. Code is available at https://github.com/11710615/PAMSR.git.

eess.IV↗

arXiv2025

Multiple magnetic states, valley electronics, and topological phase transitions in two-dimensional Janus XYZH (X = Sc, Y, La, Y = Cl, Br, I, and Z = S, Se, Te): From monolayers to bilayers

Exploring the coupling between layer, magnetism, valley, and topology in two-dimensional (2D) materials is an important approach to deepen our understanding of materials properties. We propose 27 stable ferromagnetic semiconductor monolayers of Janus XYZH (X = Sc, Y, La, Y = Cl, Br, I, and Z = S, Se, Te). All these monolayers exhibit spontaneous valley polarization, forming ferrovalley (FV) monolayers, showing anomalous valley Hall (AVH) effect. By applying external strain, the topological phase transitions including quantum anomalous Hall (QAH) effect can be introduced. In the ScBrSH bilayer system, AA and AB stacking configurations were constructed through interlayer sliding and rotational operation. The bilayer system exhibits interlayer antiferromagnetic (AFM) ordering with spontaneous valley polarization differing from the FV observed in monolayers. The sliding ferroelectricity observed in the AA stacking indicates that the system exhibits significant multiferroic characteristics. Further analysis shows that interlayer sliding can introduce a layer polarization anomalous valley Hall (LPAVH) effect, which can be precisely controlled by tuning the direction of the ferroelectric polarization. Upon applying external strain, the quantum layer spin Hall (QLSH) effect observed during the topological phase transition in the bilayer can be regarded as the superposition of two QAH monolayers. Furthermore, applying a twisting operation to the bilayer induces unexpected altermagnetism. Our study systematically reveals the rich physical properties of 2D XYZH materials, providing important theoretical foundations and guidance for the design and development of next-generation quantum devices.

cond-mat.mtrl-sci↗

arXiv2025

LongSafety: Enhance Safety for Long-Context LLMs

Recent advancements in model architectures and length extrapolation techniques have significantly extended the context length of large language models (LLMs), paving the way for their application in increasingly complex tasks. However, despite the growing capabilities of long-context LLMs, the safety issues in long-context scenarios remain underexplored. While safety alignment in short context has been widely studied, the safety concerns of long-context LLMs have not been adequately addressed. In this work, we introduce \textbf{LongSafety}, a comprehensive safety alignment dataset for long-context LLMs, containing 10 tasks and 17k samples, with an average length of 40.9k tokens. Our experiments demonstrate that training with LongSafety can enhance long-context safety performance while enhancing short-context safety and preserving general capabilities. Furthermore, we demonstrate that long-context safety does not equal long-context alignment with short-context safety data and LongSafety has generalizing capabilities in context length and long-context safety scenarios.

cs.CL↗

arXiv2025

UnitCoder: Scalable Iterative Code Synthesis with Unit Test Guidance

Large Language Models (LLMs) have demonstrated remarkable capabilities in various tasks, yet code generation remains a major challenge. Current approaches for obtaining high-quality code data primarily focus on (i) collecting large-scale pre-training data and (ii) synthesizing instruction data through prompt engineering with powerful models. While pre-training data faces quality consistency issues, instruction-based synthesis suffers from limited instruction diversity and inherent biases of LLMs. To address this gap, we introduce UnitCoder, a systematic pipeline leveraging model-generated unit tests to both guide and validate the code generation process. Combined with large-scale package-based retrieval from pre-training corpus, we generate a dataset of 500K+ verifiable programs containing diverse API calls. Evaluations on multiple Python benchmarks (BigCodeBench, HumanEval, MBPP) demonstrate that models fine-tuned on our synthetic data exhibit consistent performance improvements. Notably, Llama3.1-8B and InternLM2.5-7B improve from 31\% and 28\% to 40\% and 39\% success rates on BigCodeBench, respectively. Our work presents a scalable approach that leverages model-generated unit tests to guide the synthesis of high-quality code data from pre-training corpora, demonstrating the potential for producing diverse and high-quality post-training data at scale. All code and data will be released (https://github.com).

cs.CL↗

arXiv2025

Case2Code: Scalable Synthetic Data for Code Generation

Large Language Models (LLMs) have shown outstanding breakthroughs in code generation. Recent work improves code LLMs by training on synthetic data generated by some powerful LLMs, which can be challenging to scale due to the dependence on a teacher model and high generation costs. In this paper, we focus on synthesizing code data at scale and propose a \textbf{Case2Code} task by exploiting the expressiveness and correctness of programs. \textbf{Case2Code} is an inductive inference task that aims to infer underlying code implementations by observing input-output examples or program behaviors, By incorporating LLMs to generate program inputs, and executing the program with these inputs to obtain the program outputs, we can synthesize diverse and high-quality \textbf{Case2Code} data at scale for training and evaluating code LLMs. Experimental results show that case-to-code induction is challenging for current representative LLMs if they are untrained. Models trained with \textbf{Case2Code} improve performance not only on distribution case-to-code induction but also on various coding-generation tasks, demonstrating the great potential of large-scale synthetic data and inductive learning.

cs.CL↗

arXiv2024

GAOKAO-Eval: Does high scores truly reflect strong capabilities in LLMs?

Large Language Models (LLMs) are commonly evaluated using human-crafted benchmarks, under the premise that higher scores implicitly reflect stronger human-like performance. However, there is growing concern that LLMs may ``game" these benchmarks due to data leakage, achieving high scores while struggling with tasks simple for humans. To substantively address the problem, we create GAOKAO-Eval, a comprehensive benchmark based on China's National College Entrance Examination (Gaokao), and conduct ``closed-book" evaluations for representative models released prior to Gaokao. Contrary to prevailing consensus, even after addressing data leakage and comprehensiveness, GAOKAO-Eval reveals that high scores still fail to truly reflect human-aligned capabilities. To better understand this mismatch, We introduce the Rasch model from cognitive psychology to analyze LLM scoring patterns and identify two key discrepancies: 1) anomalous consistent performance across various question difficulties, and 2) high variance in performance on questions of similar difficulty. In addition, We identified inconsistent grading of LLM-generated answers among teachers and recurring mistake patterns. we find that the phenomenons are well-grounded in the motivations behind OpenAI o1, and o1's reasoning-as-difficulties can mitigate the mismatch. These results show that GAOKAO-Eval can reveal limitations in LLM capabilities not captured by current benchmarks and highlight the need for more LLM-aligned difficulty analysis.

cs.CL↗

arXiv2024

Spin-layer coupling in altermagnets multilayer: a design principle for spintronics

The discovery of collinear symmetric-compensated altermagnets (AM) with intrinsic spin splitting provides a route towards energy-efficient and ultrafast device applications. Here, using first-principles calculations and symmetry analysis, we propose a series of AM Cr2SX (X=O, S, Se) monolayer and explore the spin splitting in Cr2SX multilayer. A general design principle for realizing the spin-layer coupling in odd/even-layer is mapped out based on the comprehensive analysis of spin group symmetry. The spin splitting behavior related with the MzUt, Mz and ML symmetries in AM multilayer can be significantly modulated by magnetic orders, crystal symmetry and external perpendicular gate field (Ez). Due to the spin-compensated bands of sublayers linked by overall Mz and interlayers ML symmetries, the Cr2S2 odd-layer exhibits the unique coexistence of spin splitting and spin degeneracy at high symmetric paths and X/Y valley, respectively. Furthermore, owing to the higher priority of overall ML symmetry compared to interlayers ML symmetry in AM even-layer, the spin-layer coupling of AM multilayer shows strong odd/even-layer dependence. Our work not only offer a new direction for manipulating spin splitting, but also greatly enrich the research on AM monolayer and multilayer.

physics.app-ph↗

arXiv2024

Piezoelectric Manipulation and Engineering for Layertronics in Two-Dimensional Materials

The electronic transport characteristics of two-dimensional (2D) systems have widespread application prospects in the fabrication of multifunctional nanodevices. However, the current research for basic transport phenomena, such as anomalous valley Hall effect (AVHE) and piezoelectric response, is limited to discrete discussion. Here, we theoretically propose a valley-piezoelectricity coupling strategy beyond the existing paradigm to realize AVHE and layer Hall effect (LHE) in ferrovalley (FV) systems, and its essential principle can be extended to general valleytronic materials. Through first-principles calculations, we demonstrate that the large polarized electric field of 2.8*106 (1.67*107) V/m can be induced by 0.1% uniaxial strain in FV 2H-LaHF (1T-LaHF) monolayers. In addition, the microscopic mechanism of interlayer antiferromagnetic (AFM) state of 2H-LaHF bilayer is uncovered by the spin Hamiltonian and super-superexchange (SSE) interaction. Our findings pave the way for new explorations of valley Hall-related effect involving piezoelectricity.

cond-mat.mes-hall↗

arXiv2024

Possible way to achieve anomalous valley Hall effect by tunable intrinsic piezoelectric polarization in FeO$_2$SiGeN$_2$ monolayer

Valley-related multiple Hall effect and piezoelectric response are novel transport characteristics in low-dimensional system, however few studies have reported their coexistence in a single system as well as their coupling relationships. By first-principles calculations, we propose a multifunctional Janus semiconductor, i.e. FeO$_2$SiGeN$_2$ monolayer with large valley polarization of about 120 meV and in-plane piezoelectric polarization with d11 of -0.714.03 pm/V. The magnetic anisotropy energy can be significantly regulated by electronic correlation strength and strain, which can be attributed to the change of competition relationship about Fe-3d-resolved magnetic anisotropy energy brought about by external regulatory means. Electronic correlation strength can induce phase transitions in Janus FeO$_2$SiGeN$_2$ monolayer from ferrovalley to quantum anomalous Hall phase, while the half-valley metallic state as the boundary of the phase transition can gererate 100% spin- and valley polarization. The related phase transition mechanism is analyzed based on the two-band strained kp model. The presence of piezoelectric strain coefficients d11 in valleytronic material makes the coupling between charge degrees of freedom and valley degrees of freedom possible, and the intrinsic electric field caused by the in-plane piezoelectric response provide the way to realize piezoelectric anomalous valley Hall effect. This work may pave a way to find a new member of materials with valley-related multiple Hall effect and stimulate further experimental works related to valleytronics and piezotronics.

cond-mat.mes-hall↗

arXiv2024

Unified Active Retrieval for Retrieval Augmented Generation

In Retrieval-Augmented Generation (RAG), retrieval is not always helpful and applying it to every instruction is sub-optimal. Therefore, determining whether to retrieve is crucial for RAG, which is usually referred to as Active Retrieval. However, existing active retrieval methods face two challenges: 1. They usually rely on a single criterion, which struggles with handling various types of instructions. 2. They depend on specialized and highly differentiated procedures, and thus combining them makes the RAG system more complicated and leads to higher response latency. To address these challenges, we propose Unified Active Retrieval (UAR). UAR contains four orthogonal criteria and casts them into plug-and-play classification tasks, which achieves multifaceted retrieval timing judgements with negligible extra inference cost. We further introduce the Unified Active Retrieval Criteria (UAR-Criteria), designed to process diverse active retrieval scenarios through a standardized procedure. Experiments on four representative types of user instructions show that UAR significantly outperforms existing work on the retrieval timing judgement and the performance of downstream tasks, which shows the effectiveness of UAR and its helpfulness to downstream tasks.

cs.CL↗

arXiv2024

Orbital origin of magnetic moment enhancement induced by charge density wave in kagome FeGe

Interactions among various electronic states such as CDW, magnetism, and superconductivity are of high significance in strongly correlated systems. While significant progress has been made in understanding the relationship between CDW and superconductivity, the interplay between CDW and magnetic order remains largely elusive. Kagome lattices, which intertwine nontrivial topology, charge order, and magnetism, offer an ideal platform for such studies. The kagome magnet FeGe, hosting the unique coupling between CDW and magnetism, has recently garnered considerable attention in that respect. Here we reveal the significant role of the orbital coupling effect during the CDW phase transition, highlighting the orbital origin of the magnetic moment enhancement in FeGe. Our X ray absorption experiments and first principles calculations illuminate the temperature dependent behavior of Fe3d_Ge4p orbital hybridization and corroborate its pivotal impact on the magnetic properties of FeGe. These findings introduce an orbital dimension to the correlation between charge and magnetic degrees of freedom, advancing our understanding of the intriguing quantum phases resulting from this interplay.

cond-mat.str-el↗

arXiv2024

Balanced Data Sampling for Language Model Training with Clustering

Data plays a fundamental role in the training of Large Language Models (LLMs). While attention has been paid to the collection and composition of datasets, determining the data sampling strategy in training remains an open question. Most LLMs are trained with a simple strategy, random sampling. However, this sampling strategy ignores the unbalanced nature of training data distribution, which can be sub-optimal. In this paper, we propose ClusterClip Sampling to balance the text distribution of training data for better model training. Specifically, ClusterClip Sampling utilizes data clustering to reflect the data distribution of the training set and balances the common samples and rare samples during training based on the cluster results. A repetition clip operation is introduced to mitigate the overfitting issue led by samples from certain clusters. Extensive experiments validate the effectiveness of ClusterClip Sampling, which outperforms random sampling and other cluster-based sampling variants under various training datasets and large language models.

cs.CL↗

arXiv2024

InternLM-Math: Open Math Large Language Models Toward Verifiable Reasoning

The math abilities of large language models can represent their abstract reasoning ability. In this paper, we introduce and open-source our math reasoning LLMs InternLM-Math which is continue pre-trained from InternLM2. We unify chain-of-thought reasoning, reward modeling, formal reasoning, data augmentation, and code interpreter in a unified seq2seq format and supervise our model to be a versatile math reasoner, verifier, prover, and augmenter. These abilities can be used to develop the next math LLMs or self-iteration. InternLM-Math obtains open-sourced state-of-the-art performance under the setting of in-context learning, supervised fine-tuning, and code-assisted reasoning in various informal and formal benchmarks including GSM8K, MATH, Hungary math exam, MathBench-ZH, and MiniF2F. Our pre-trained model achieves 30.3 on the MiniF2F test set without fine-tuning. We further explore how to use LEAN to solve math problems and study its performance under the setting of multi-task learning which shows the possibility of using LEAN as a unified platform for solving and proving in math. Our models, codes, and data are released at \url{https://github.com/InternLM/InternLM-Math}.

cs.CL↗

arXiv2024

LLatrieval: LLM-Verified Retrieval for Verifiable Generation

Verifiable generation aims to let the large language model (LLM) generate text with supporting documents, which enables the user to flexibly verify the answer and makes the LLM's output more reliable. Retrieval plays a crucial role in verifiable generation. Specifically, the retrieved documents not only supplement knowledge to help the LLM generate correct answers, but also serve as supporting evidence for the user to verify the LLM's output. However, the widely used retrievers become the bottleneck of the entire pipeline and limit the overall performance. Their capabilities are usually inferior to LLMs since they often have much fewer parameters than the large language model and have not been demonstrated to scale well to the size of LLMs. If the retriever does not correctly find the supporting documents, the LLM can not generate the correct and verifiable answer, which overshadows the LLM's remarkable abilities. To address these limitations, we propose \LLatrieval (Large Language Model Verified Retrieval), where the LLM updates the retrieval result until it verifies that the retrieved documents can sufficiently support answering the question. Thus, the LLM can iteratively provide feedback to retrieval and facilitate the retrieval result to fully support verifiable generation. Experiments show that LLatrieval significantly outperforms extensive baselines and achieves state-of-the-art results.

cs.CL↗