Author name cluster

Hanzhu Chen

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

8 papers

2 author rows

ICML Conference 2025 Conference Paper

Boosting Multi-Domain Fine-Tuning of Large Language Models through Evolving Interactions between Samples

Xize Liang
Lin Yang 0009
Jie Wang 0005
Yiyang Lu
Runyu Wu
Hanzhu Chen
Jianye Hao

The multi-domain fine-tuning of large language models (LLMs) confronts a notorious trade-off among abilities across domains. Existing studies attribute this trade-off to the conflicts between samples rooted in inherent semantics. Recent approaches attempt to mitigate these conflicts through the empirical investigation or heuristic strategies. However, without a fundamental understanding of interactions between samples, they yield only marginal improvements, while incurring substantial trial-and-error costs. To address this challenge, we move beyond empirical studies by modeling interactions between samples as their influence on each other’s loss, estimated using gradients. Intriguingly, we find that these interactions evolve throughout training rather than being purely determined by inherent semantics. Building on this insight, we propose EV olving I nteraction-guided C urriculum ( EVIC ), which iteratively selects samples that positively influence the overall dataset for training. By dynamically adapting the training curriculum to prioritize samples that contribute the most to the model training, EVIC effectively mitigates conflicts and improves the sample efficiency. Extensive experiments on a mixed dataset covering coding, math, and general tasks with several model architectures show that EVIC significantly outperforms all baselines across diverse capabilities.

Details

AIIM Journal 2025 Journal Article

CATI: A medical context-enhanced framework for diagnosis code assignment in the UK Biobank study

Yue Shen
Jie Wang
Zhe Wang
Zhihao Shi
Hanzhu Chen
Zheng Wang
Yukang Jiang
Xiaopu Wang

Diagnosis codes are standard code format of diseases or medical conditions. This study is aimed at assigning diagnosis codes to patients in large-scale biobanks, particularly addressing the issue of missing codes for some patients. This is crucial for downstream disease-related tasks. While recent methods primarily rely on structured biobank data for code assignment, they often overlook the valuable medical context provided by textual information in the biobanks and hierarchical structure of the disease coding system. To address this gap, we have developed CATI, a medical context-enhanced framework for diagnosis Code Assignment by integrating Textual details derived from key features and disease hIerarchy. The study is based on the UK Biobank data and considers Phecodes and ICD-10 codes as standard disease formats. We start by representing ten informative codified features using their formal names and then integrate them into CATI as text embeddings, achieved through prompt tuning on the pre-trained language model BioBERT. Recognizing the hierarchical structure of diagnosis codes, we have developed a novel convolution layer in our method that effectively propagates logits between adjacent diagnosis codes. Evaluation results demonstrate that CATI outperforms existing state-of-the-art methods in terms of both Phecodes and ICD-10 codes, boasting at least a 5. 16% improvement in average AUROC for unseen disease codes and an 8. 68% rise in average AUPRC for disease codes with training instances ranging in (1000, 10000]. This framework contributes to the formation of well-defined cohorts for downstream studies and offers a unique perspective for addressing complex healthcare tasks by incorporating vital medical context.

Details DOI

ICLR Conference 2025 Conference Paper

Knowledge Graph Finetuning Enhances Knowledge Manipulation in Large Language Models

Hanzhu Chen
Xu Shen 0001
Jie Wang 0005
Zehao Wang
Qitan Lv
Junjie He
Rong Wu
Feng Wu 0001

Despite the impressive performance of general large language models(LLMs), many of their applications in specific domains (e.g., low-data and knowledge-intensive) still confront significant challenges. Supervised fine-tuning (SFT)---where a general LLM is further trained on a small labeled dataset to adapt for specific tasks or domains---has shown great power for developing domain-specific LLMs. However, existing SFT data primarily consist of Question and Answer (Q&A) pairs, which poses a significant challenge for LLMs to comprehend the correlation and logic of knowledge underlying the Q&A. To address this challenge, we propose a conceptually flexible and general framework to boost SFT, namely Knowledge Graph-Driven Supervised Fine-Tuning (KG-SFT). The key idea of KG-SFT is to generate high-quality explanations for each Q&A pair via a structured knowledge graph to enhance the knowledge comprehension and manipulation of LLMs. Specifically, KG-SFT consists of three components: Extractor, Generator, and Detector. For a given Q&A pair, (i) Extractor first identifies entities within Q&A pairs and extracts relevant reasoning subgraphs from external KGs, (ii) Generator then produces corresponding fluent explanations utilizing these reasoning subgraphs, and (iii) finally, Detector performs sentence-level knowledge conflicts detection on these explanations to guarantee the reliability. KG-SFT focuses on generating high-quality explanations to improve the quality of Q&A pair, which reveals a promising direction for supplementing existing data augmentation methods. Extensive experiments on fifteen different domains and six different languages demonstrate the effectiveness of KG-SFT, leading to an accuracy improvement of up to 18% and an average of 8.7% in low-data scenarios.

Details

NeurIPS Conference 2025 Conference Paper

LogicTree: Improving Complex Reasoning of LLMs via Instantiated Multi-step Synthetic Logical Data

Zehao Wang
Lin Yang
Jie Wang
Kehan Wang
Hanzhu Chen
Bin Wang
Jianye Hao
Defu Lian

Despite their remarkable performance on various tasks, Large Language Models (LLMs) still struggle with logical reasoning, particularly in complex and multi-step reasoning processes. Among various efforts to enhance LLMs' reasoning capabilities, synthesizing large-scale, high-quality logical reasoning datasets has emerged as a promising direction. However, existing methods often rely on predefined templates for logical reasoning data generation, limiting their adaptability to real-world scenarios. To address the limitation, we propose LogicTree, a novel framework for efficiently synthesizing multi-step logical reasoning dataset that excels in both complexity and instantiation. By iteratively searching for applicable logic rules based on structural pattern matching to perform backward deduction, LogicTree constructs multi-step logic trees that capture complex reasoning patterns. Furthermore, we employ a two-stage LLM-based approach to instantiate various real-world scenarios for each logic tree, generating consistent real-world reasoning processes that carry contextual significance. This helps LLMs develop generalizable logical reasoning abilities across diverse scenarios rather than merely memorizing templates. Experiments on multiple benchmarks demonstrate that our approach achieves an average improvement of 9. 4\% in accuracy on complex logical reasoning tasks.

PDF Details

ICML Conference 2025 Conference Paper

ROPO: Robust Preference Optimization for Large Language Models

Xize Liang
Chao Chen 0026
Shuang Qiu
Jie Wang 0005
Yue Wu
Zhihang Fu
Hanzhu Chen
Feng Wu 0001

The prevalent noise in the preference data unavoidably poses significant challenges to the preference alignment of large language models (LLMs). Existing efforts for this problem either marginally alleviate the impact of noise without noise reduction, or rely on external LLMs that incur substantial computational costs. To address these challenges, we propose RO bust P reference O ptimization ( ROPO ), an iterative alignment approach that integrates noise-tolerance and noise filtering without the aid of external models. Specifically, ROPO first formulates the training process with adaptive noise reduction as an optimization problem, which can be efficiently solved in an iterative paradigm. Then, to equip this solving process with noise-tolerance and noise-identification capabilities, we derive a robust loss that suppresses the gradients from samples with high uncertainty. We demonstrate both empirically and theoretically that the derived loss is key to the noise-tolerance and effective filtering of noisy samples. The derived loss further inspires a robustness-guided rejection sampling technique to compensate for the potential important information in discarded queries. Extensive experiments on several widely-used datasets and model architectures demonstrate that ROPO significantly outperforms all baselines under four practical noise settings and the random symmetric noise, with its advantage increasing as the noise rate increases.

Details

ICML Conference 2024 Conference Paper

Coarse-to-Fine Highlighting: Reducing Knowledge Hallucination in Large Language Models

Qitan Lv
Jie Wang 0005
Hanzhu Chen
Bin Li 0025
Yongdong Zhang 0001
Feng Wu 0001

Generation of plausible but incorrect factual information, often termed hallucination, has attracted significant research interest. Retrieval-augmented language model (RALM)—which enhances models with up-to-date knowledge—emerges as a promising method to reduce hallucination. However, existing RALMs may instead exacerbate hallucination when retrieving lengthy contexts. To address this challenge, we propose COFT, a novel CO arse-to- F ine highligh T ing method to focus on different granularity-level key texts, thereby avoiding getting lost in lengthy contexts. Specifically, COFT consists of three components: recaller, scorer, and selector. First, recaller applies a knowledge graph to extract potential key entities in a given context. Second, scorer measures the importance of each entity by calculating its contextual weight. Finally, selector selects high contextual weight entities with a dynamic threshold algorithm and highlights the corresponding paragraphs, sentences, or words in a coarse-to-fine manner. Extensive experiments on knowledge hallucination benchmark demonstrate the effectiveness of COFT, leading to a superior performance over 30% in F1 score metric. Moreover, COFT also exhibits remarkable versatility across various long-form tasks, such as reading comprehension and question answering.

Details

NeurIPS Conference 2024 Conference Paper

Neural Krylov Iteration for Accelerating Linear System Solving

Jian Luo
Jie Wang
Hong Wang
huanshuo dong
Zijie Geng
Hanzhu Chen
Yufei Kuang

Solving large-scale sparse linear systems is essential in fields like mathematics, science, and engineering. Traditional numerical solvers, mainly based on the Krylov subspace iteration algorithm, suffer from the low-efficiency problem, which primarily arises from the less-than-ideal iteration. To tackle this problem, we propose a novel method, namely Neur al K rylov It era t ion ( NeurKItt ), for accelerating linear system solving. Specifically, NeurKItt employs a neural operator to predict the invariant subspace of the linear system and then leverages the predicted subspace to accelerate linear system solving. To enhance the subspace prediction accuracy, we utilize QR decomposition for the neural operator outputs and introduce a novel projection loss function for training. NeurKItt benefits the solving by using the predicted subspace to guide the iteration process, significantly reducing the number of iterations. We provide extensive experiments and comprehensive theoretical analyses to demonstrate the feasibility and efficiency of NeurKItt. In our main experiments, NeurKItt accelerates the solving of linear systems across various settings and datasets, achieving up to a 5. 5× speedup in computation time and a 16. 1× speedup in the number of iterations.

PDF Details DOI

NeurIPS Conference 2023 Conference Paper

Learning Rule-Induced Subgraph Representations for Inductive Relation Prediction

Tianyu Liu
Qitan Lv
Jie Wang
Shuling Yang
Hanzhu Chen

Inductive relation prediction (IRP)---where entities can be different during training and inference---has shown great power for completing evolving knowledge graphs. Existing works mainly focus on using graph neural networks (GNNs) to learn the representation of the subgraph induced from the target link, which can be seen as an implicit rule-mining process to measure the plausibility of the target link. However, these methods are not able to differentiate the target link and other links during message passing, hence the final subgraph representation will contain irrelevant rule information to the target link, which reduces the reasoning performance and severely hinders the applications for real-world scenarios. To tackle this problem, we propose a novel $\textit{single-source edge-wise}$ GNN model to learn the $\textbf{R}$ule-induc$\textbf{E}$d $\textbf{S}$ubgraph represen$\textbf{T}$ations $(\textbf{REST}$), which encodes relevant rules and eliminates irrelevant rules within the subgraph. Specifically, we propose a $\textit{single-source}$ initialization approach to initialize edge features only for the target link, which guarantees the relevance of mined rules and target link. Then we propose several RNN-based functions for $\textit{edge-wise}$ message passing to model the sequential property of mined rules. REST is a simple and effective approach with theoretical support to learn the $\textit{rule-induced subgraph representation}$. Moreover, REST does not need node labeling, which significantly accelerates the subgraph preprocessing time by up to $\textbf{11. 66}\times$. Experiments on inductive relation prediction benchmarks demonstrate the effectiveness of our REST.

PDF Details