arXiv subjects

Guoxin Chen

Publications and source records attributed to Guoxin Chen.

34 records · Page 2Linked to original sources

arXiv2025

From Data-Centric to Sample-Centric: Enhancing LLM Reasoning via Progressive Optimization

Reinforcement learning with verifiable rewards (RLVR) has recently advanced the reasoning capabilities of large language models (LLMs). While prior work has emphasized algorithmic design, data curation, and reward shaping, we investigate RLVR from a sample-centric perspective and introduce LPPO (Learning-Progress and Prefix-guided Optimization), a framework of progressive optimization techniques. Our work addresses a critical question: how to best leverage a small set of trusted, high-quality demonstrations, rather than simply scaling up data volume. First, motivated by how hints aid human problem-solving, we propose prefix-guided sampling, an online data augmentation method that incorporates partial solution prefixes from expert demonstrations to guide the policy, particularly for challenging instances. Second, inspired by how humans focus on important questions aligned with their current capabilities, we introduce learning-progress weighting, a dynamic strategy that adjusts each training sample's influence based on model progression. We estimate sample-level learning progress via an exponential moving average of per-sample pass rates, promoting samples that foster learning and de-emphasizing stagnant ones. Experiments on mathematical-reasoning benchmarks demonstrate that our methods outperform strong baselines, yielding faster convergence and a higher performance ceiling.

cs.LG↗

arXiv2025

DiReCT: Diagnostic Reasoning for Clinical Notes via Large Language Models

Large language models (LLMs) have recently showcased remarkable capabilities, spanning a wide range of tasks and applications, including those in the medical domain. Models like GPT-4 excel in medical question answering but may face challenges in the lack of interpretability when handling complex tasks in real clinical settings. We thus introduce the diagnostic reasoning dataset for clinical notes (DiReCT), aiming at evaluating the reasoning ability and interpretability of LLMs compared to human doctors. It contains 511 clinical notes, each meticulously annotated by physicians, detailing the diagnostic reasoning process from observations in a clinical note to the final diagnosis. Additionally, a diagnostic knowledge graph is provided to offer essential knowledge for reasoning, which may not be covered in the training data of existing LLMs. Evaluations of leading LLMs on DiReCT bring out a significant gap between their reasoning ability and that of human doctors, highlighting the critical need for models that can reason effectively in real-world clinical scenarios.

cs.CL↗

arXiv2025

Table-Critic: A Multi-Agent Framework for Collaborative Criticism and Refinement in Table Reasoning

Despite the remarkable capabilities of large language models (LLMs) in various reasoning tasks, they still struggle with table reasoning tasks, particularly in maintaining consistency throughout multi-step reasoning processes. While existing approaches have explored various decomposition strategies, they often lack effective mechanisms to identify and correct errors in intermediate reasoning steps, leading to cascading error propagation. To address these issues, we propose Table-Critic, a novel multi-agent framework that facilitates collaborative criticism and iterative refinement of the reasoning process until convergence to correct solutions. Our framework consists of four specialized agents: a Judge for error identification, a Critic for comprehensive critiques, a Refiner for process improvement, and a Curator for pattern distillation. To effectively deal with diverse and unpredictable error types, we introduce a self-evolving template tree that systematically accumulates critique knowledge through experience-driven learning and guides future reflections. Extensive experiments have demonstrated that Table-Critic achieves substantial improvements over existing methods, achieving superior accuracy and error correction rates while maintaining computational efficiency and lower solution degradation rate.

cs.AI↗

arXiv2025

C-3PO: Compact Plug-and-Play Proxy Optimization to Achieve Human-like Retrieval-Augmented Generation

Retrieval-augmented generation (RAG) systems face a fundamental challenge in aligning independently developed retrievers and large language models (LLMs). Existing approaches typically involve modifying either component or introducing simple intermediate modules, resulting in practical limitations and sub-optimal performance. Inspired by human search behavior -- typically involving a back-and-forth process of proposing search queries and reviewing documents, we propose C-3PO, a proxy-centric framework that facilitates communication between retrievers and LLMs through a lightweight multi-agent system. Our framework implements three specialized agents that collaboratively optimize the entire RAG pipeline without altering the retriever and LLMs. These agents work together to assess the need for retrieval, generate effective queries, and select information suitable for the LLMs. To enable effective multi-agent coordination, we develop a tree-structured rollout approach for reward credit assignment in reinforcement learning. Extensive experiments in both in-domain and out-of-distribution scenarios demonstrate that C-3PO significantly enhances RAG performance while maintaining plug-and-play flexibility and superior generalization capabilities.

cs.CL↗

arXiv2025

Learning Evolving Tools for Large Language Models

Tool learning enables large language models (LLMs) to interact with external tools and APIs, greatly expanding the application scope of LLMs. However, due to the dynamic nature of external environments, these tools and APIs may become outdated over time, preventing LLMs from correctly invoking tools. Existing research primarily focuses on static environments and overlooks this issue, limiting the adaptability of LLMs in real-world applications. In this paper, we propose ToolEVO, a novel framework designed to enhance the adaptive and reflective capabilities of LLMs against tool variability. By leveraging Monte Carlo Tree Search, ToolEVO facilitates active exploration and interaction of LLMs within dynamic environments, allowing for autonomous self-reflection and self-updating of tool usage based on environmental feedback. Additionally, we introduce ToolQA-D, a benchmark specifically designed to evaluate the impact of tool variability. Extensive experiments demonstrate the effectiveness and stability of our approach, highlighting the importance of adaptability to tool variability for effective tool learning. Code: https://github.com/Chen-GX/ToolEVO

cs.CL↗

arXiv2024

AlphaMath Almost Zero: Process Supervision without Process

Although recent advancements in large language models (LLMs) have significantly improved their performance on various tasks, they still face challenges with complex and symbolic multi-step reasoning, particularly in mathematical reasoning. To bolster the mathematical reasoning capabilities of LLMs, most existing efforts concentrate on seeking assistance from either domain experts or GPT-4 for high-quality process-supervised data, which is not only expensive but also labor-intensive. In our study, we propose an innovative framework, AlphaMath, that bypasses the need for process annotations (from humans or GPTs) by leveraging Monte Carlo Tree Search (MCTS). This framework focuses on unleashing the potential of a well-pretrained LLM to autonomously enhance its mathematical reasoning. Specifically, we integrate a value model with the LLM, automatically generating both process supervision and step-level evaluation signals in MCTS. Furthermore, we propose an efficient inference strategy, step-level beam search, where the value model is crafted to assist the policy model (i.e., LLM) in navigating more effective reasoning paths, rather than solely relying on prior probabilities. The experimental results on both in-domain and out-of-domain datasets demonstrate that even without GPT-4 or human-annotated process supervision, our AlphaMath framework achieves comparable or superior results to previous state-of-the-art methods.

cs.CL↗

arXiv2024

Step-level Value Preference Optimization for Mathematical Reasoning

Direct Preference Optimization (DPO) using an implicit reward model has proven to be an effective alternative to reinforcement learning from human feedback (RLHF) for fine-tuning preference aligned large language models (LLMs). However, the overall preference annotations of responses do not fully capture the fine-grained quality of model outputs in complex multi-step reasoning tasks, such as mathematical reasoning. To address this limitation, we introduce a novel algorithm called Step-level Value Preference Optimization (SVPO). Our approach employs Monte Carlo Tree Search (MCTS) to automatically annotate step-level preferences for multi-step reasoning. Furthermore, from the perspective of learning-to-rank, we train an explicit value model to replicate the behavior of the implicit reward model, complementing standard preference optimization. This value model enables the LLM to generate higher reward responses with minimal cost during inference. Experimental results demonstrate that our method achieves state-of-the-art performance on both in-domain and out-of-domain mathematical reasoning benchmarks. Our code is available at \url{https://github.com/MARIO-Math-Reasoning/Super_MARIO}.

cs.CL↗

arXiv2024

SEER: Facilitating Structured Reasoning and Explanation via Reinforcement Learning

Elucidating the reasoning process with structured explanations from question to answer is crucial, as it significantly enhances the interpretability, traceability, and trustworthiness of question-answering (QA) systems. However, structured explanations demand models to perform intricately structured reasoning, which poses great challenges. Most existing methods focus on single-step reasoning through supervised learning, ignoring logical dependencies between steps. Moreover, existing reinforcement learning (RL) based methods overlook the structured relationships, underutilizing the potential of RL in structured reasoning. In this paper, we propose SEER, a novel method that maximizes a structure-based return to facilitate structured reasoning and explanation. Our proposed structure-based return precisely describes the hierarchical and branching structure inherent in structured reasoning, effectively capturing the intricate relationships between different reasoning steps. In addition, we introduce a fine-grained reward function to meticulously delineate diverse reasoning steps. Extensive experiments show that SEER significantly outperforms state-of-the-art methods, achieving an absolute improvement of 6.9% over RL-based methods on EntailmentBank, a 4.4% average improvement on STREET benchmark, and exhibiting outstanding efficiency and cross-dataset generalization performance. Our code is available at https://github.com/Chen-GX/SEER.

cs.CL↗

arXiv2024

Accurate background velocity model building method based on iterative deep learning in sparse transform domain

Whether it is oil and gas exploration or geological science research, it is necessary to accurately grasp the structural information of underground media. Full waveform inversion is currently the most popular seismic wave inversion method, but it is highly dependent on a high-quality initial model. Artificial intelligence algorithm deep learning is completely data-driven and can get rid of the dependence on the initial model. However, the prediction accuracy of deep learning algorithms depends on the scale and diversity of training data sets. How to improve the prediction accuracy of deep learning without increasing the size of the training set while also improving computing efficiency is a worthy issue to study. In this paper, an iterative deep learning algorithm in the sparse transform domain is proposed based on the characteristics of deep learning: first, based on the computational efficiency and the effect of sparse transform, the cosine transform is selected as the sparse transform method, and the seismic data and the corresponding velocity model are cosine transformed to obtain their corresponding sparse expressions, which are then used as the input data and corresponding label data for deep learning; then we give an iterative deep learning algorithm in the cosine transform domain, that is, after obtaining the seismic data residuals and velocity model residuals of the previous round of test results, they are used again as new input data and label data, and re-trained in the cosine domain to obtain a new network, and the prediction results of the previous round are corrected, and then the cycle is repeated until the termination condition is reached. The algorithm effect was verified on the SEG/EAGE salt model and the seabed sulfide physical model site data.

physics.geo-ph↗

arXiv2024

FCS-HGNN: Flexible Multi-type Community Search in Heterogeneous Information Networks

Community search is a personalized community discovery problem designed to identify densely connected subgraphs containing the query node. Recently, community search in heterogeneous information networks (HINs) has received considerable attention. Existing methods typically focus on modeling relationships in HINs through predefined meta-paths or user-specified relational constraints. However, metapath-based methods are primarily designed to identify single-type communities with nodes of the same type rather than multi-type communities involving nodes of different types. Constraint-based methods require users to have a good understanding of community patterns to define a suitable set of relational constraints, which increases the burden on users. In this paper, we propose FCS-HGNN, a novel method for flexibly identifying both single-type and multi-type communities in HINs. Specifically, FCS-HGNN extracts complementary information from different views and dynamically considers the contribution of each relation instead of treating them equally, thereby capturing more fine-grained heterogeneous information. Furthermore, to improve efficiency on large-scale graphs, we further propose LS-FCS-HGNN, which incorporates i) the neighbor sampling strategy to improve training efficiency, and ii) the depth-based heuristic search strategy to improve query efficiency. We conducted extensive experiments to demonstrate the superiority of our proposed methods over state-of-the-art methods, achieving average improvements of 14.3% and 11.1% on single-type and multi-type communities, respectively.

cs.SI↗

arXiv2024

Momentum-resolved electronic structures and strong electronic correlations in graphene-like nitride superconductors

Although transition-metal nitrides have been widely applied for several decades, experimental investigations of their high-resolution electronic band structures are rare due to the lack of high-quality single-crystalline samples. Here, we report on the first momentum-resolved electronic band structures of titanium nitride (TiN) films, a remarkable nitride superconductor. The measurements of crystal structures and electrical transport properties confirmed the high quality of these films. More importantly, with a combination of high-resolution angle-resolved photoelectron spectroscopy and the first-principles calculations, the extracted Coulomb interaction strength of TiN films can be as large as 8.5 eV, whereas resonant photoemission spectroscopy yields a value of 6.26 eV. These large values of Coulomb interaction strength indicate that superconducting TiN is a strongly correlated system. Our results uncover the unexpected electronic correlations in transition-metal nitrides, potentially providing a perspective not only to understand their emergent quantum states but also to develop their applications in quantum devices.

cond-mat.mtrl-sci↗

arXiv2024

Cooperatively Modulating Magnetic Anisotropy and Colossal Magnetoresistance via Atomic-Scale Buffer Layers in Highly Strained La0.7Sr0.3MnO3 Films

Simultaneous control of magnetic anisotropy and magnetoresistance, especially with atomic scale precision, remains a pivotal challenge for realizing advanced spintronic functionalities. Here we demonstrate cooperative continuous control over both magnetoresistance and magnetic anisotropy in highly strained La0.7Sr0.3MnO3 (LSMO) thin films. By inserting varying perovskite buffer layers, compressively strained LSMO films transition from a ferromagnetic insulator with out-of-plane magnetic anisotropy to a metallic state with in-plane anisotropy. Atomic-scale buffer layer insertion enables remarkably acute, precise control to sharply modulate this magnetic phase transformation. A gigantic 10,000% modulation of the colossal magnetoresistance (CMR) and an exceptionally sharp transition from out-of-plane to in-plane magnetic anisotropy are attained in just a few contiguous layers. These atomic-scale correlations among electronic, magnetic, and structural order parameters yield flexible multifunctional control promising for next-generation oxide spintronics.

cond-mat.mtrl-sci↗

arXiv2023

Causality and Independence Enhancement for Biased Node Classification

Most existing methods that address out-of-distribution (OOD) generalization for node classification on graphs primarily focus on a specific type of data biases, such as label selection bias or structural bias. However, anticipating the type of bias in advance is extremely challenging, and designing models solely for one specific type may not necessarily improve overall generalization performance. Moreover, limited research has focused on the impact of mixed biases, which are more prevalent and demanding in real-world scenarios. To address these limitations, we propose a novel Causality and Independence Enhancement (CIE) framework, applicable to various graph neural networks (GNNs). Our approach estimates causal and spurious features at the node representation level and mitigates the influence of spurious correlations through the backdoor adjustment. Meanwhile, independence constraint is introduced to improve the discriminability and stability of causal and spurious features in complex biased environments. Essentially, CIE eliminates different types of data biases from a unified perspective, without the need to design separate methods for each bias as before. To evaluate the performance under specific types of data biases, mixed biases, and low-resource scenarios, we conducted comprehensive experiments on five publicly available datasets. Experimental results demonstrate that our approach CIE not only significantly enhances the performance of GNNs but outperforms state-of-the-art debiased node classification methods.

cs.LG↗

arXiv2023

MPrompt: Exploring Multi-level Prompt Tuning for Machine Reading Comprehension

The large language models have achieved superior performance on various natural language tasks. One major drawback of such approaches is they are resource-intensive in fine-tuning new datasets. Soft-prompt tuning presents a resource-efficient solution to fine-tune the pre-trained language models (PLMs) while keeping their weight frozen. Existing soft prompt methods mainly focus on designing the input-independent prompts that steer the model to fit the domain of the new dataset. Those methods often ignore the fine-grained information about the task and context of the text. In this paper, we propose a multi-level prompt tuning (MPrompt) method for machine reading comprehension. It utilizes prompts at task-specific, domain-specific, and context-specific levels to enhance the comprehension of input semantics at different granularities. We also propose an independence constraint to steer each domain-specific prompt to focus on information within its domain to avoid redundancy. Moreover, we present a prompt generator that incorporates context-related knowledge in the prompt generation to enhance contextual relevancy. We conducted extensive experiments on 12 benchmarks of various QA formats and achieved an average improvement of 1.94\% over the state-of-the-art methods.

cs.CL↗

arXiv2022

Effect of twin boundaries on the strength of body-centered cubic tungsten nanowires

Twin boundaries (TBs) are assumed to be obstacles to dislocation motion and increase the strength of metals. Here, we report the abnormal phenomenon that TBs reduce the strength of body-centered cubic (BCC) tungsten (W). [1-11]-oriented W nanowires with (121) twin planes and free of dislocations were fabricated by chemical vapor deposition. In situ tensile tests within the transmission electron microscope were performed on single-crystal and twinned W nanowires. The fracture strength of the twinned W nanowire was 13.7 GPa, 16% lower than that of the single-crystal W nanowire (16.3 GPa). The weakening mechanism of the TB was revealed by a combination of atomic-resolution characterizations and atomistic simulations. Twinned W nanowires failed by the early nucleation of a crack at the intersection of the TB with the surface. The standard strengthening mechanism by dislocation/TB interaction was not operative in W because the high Peierls barrier and stacking fault energy in W hinder dislocation nucleation and glide. These findings provide a new insight into the influence of TBs on the mechanical properties of BCC metals.

cond-mat.mtrl-sci↗

arXiv2022

Chemical-scissor-mediated structural editing of layered transition metal carbides

Intercalation of non-van der Waals (vdW) layered materials can produce new 2D and 3D materials with unique properties, but it is difficult to achieve. Here, we describe a structural editing protocol for 3D non-vdW layered ternary carbides and nitrides (MAX phases) and their 2D vdW derivatives (MXenes). Gap-opening and species-intercalating stages were mediated by chemical scissors and guest intercalants, creating a large family of layered materials with unconventional elements and structures in MAX phases, as well as MXenes with versatile termination species. Removal of surface terminations by metal scissors and stitching of carbide layers by metal atoms leads to a reverse transformation from MXenes to MAX phases, and metal-intercalated 2D carbides. This scissor-mediated structural editing may enable structural and chemical tailoring of other layered ceramics.

cond-mat.mtrl-sci↗