Author name cluster

Yihong Dong

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.

7 papers

2 author rows

AAAI Conference 2026 Conference Paper

Large Language Model Unlearning for Source Code

Xue Jiang
Yihong Dong
Huangzhao Zhang
Tangxinyu Wang
Zheng Fang
Yingwei Ma
Rongyu Cao
Binhua Li

While Large Language Models (LLMs) excel at code generation, their inherent tendency toward verbatim memorization of training data introduces critical risks like copyright infringement, insecurity emission, and deprecated API utilization, etc. A straightforward yet promising defense is unlearning, i.e., erasing or down-weighting the offending snippets through post-training. However, we find its application to source code often tends to spill over, damaging the basic knowledge of programming languages learned by the LLM and degrading the overall capability. To ease this challenge, we propose PROD for precise source code unlearning. PROD surgically zeroes out the prediction probability of the prohibited tokens, and renormalizes the remaining distribution so that the generated code stays correct. By excising only the targeted snippets, PROD achieves precise forgetting without much degradation of the LLM's overall capability. To facilitate in-depth evaluation against PROD, we establish an unlearning benchmark consisting of three downstream tasks (i.e., unlearning of copyrighted code, insecure code, and deprecated APIs), and introduce Pareto Dominance Ratio (PDR) metric, which indicates both the forget quality and the LLM utility. Our comprehensive evaluation demonstrates that PROD achieves superior overall performance between forget quality and model utility compared to existing unlearning approaches across three downstream tasks, while consistently exhibiting improvements when applied to LLMs of varying series. PROD also exhibits superior robustness against adversarial attacks without generating or exposing the data to be forgotten. These results underscore that our approach not only successfully extends the application boundary of unlearning techniques to source code, but also holds significant implications for advancing reliable code generation.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

FAN: Fourier Analysis Networks

Yihong Dong
Ge Li
Yongding Tao
Xue Jiang
Kechi Zhang
Jia Li
Jinliang Deng
Jing Su

Despite the remarkable successes of general-purpose neural networks, such as MLPs and Transformers, we find that they exhibit notable shortcomings in modeling and reasoning about periodic phenomena, achieving only marginal performance within the training domain and failing to generalize effectively to out-of-domain (OOD) scenarios. Periodicity is ubiquitous throughout nature and science. Therefore, neural networks should be equipped with the essential ability to model and handle periodicity. In this work, we propose FAN, a novel neural network that effectively addresses periodicity modeling challenges while offering broad applicability similar to MLP with fewer parameters and FLOPs. Periodicity is naturally integrated into FAN's structure and computational processes by introducing the Fourier Principle. Unlike existing Fourier-based networks, which possess particular periodicity modeling abilities but face challenges in scaling to deeper networks and are typically designed for specific tasks, our approach overcomes this challenge to enable scaling to large-scale models and maintains the capability to be applied to more types of tasks. Through extensive experiments, we demonstrate the superiority of FAN in periodicity modeling tasks and the effectiveness and generalizability of FAN across a range of real-world tasks. Moreover, we reveal that compared to existing Fourier-based networks, FAN accommodates both periodicity modeling and general-purpose modeling well.

PDF Details

NeurIPS Conference 2025 Conference Paper

Reasoning is Periodicity? Improving Large Language Models Through Effective Periodicity Modeling

Yihong Dong
Ge Li
Xue Jiang
Yongding Tao
Kechi Zhang
Lecheng Wang
Hao Zhu
Huanyu Liu

Periodicity, as one of the most important basic characteristics, lays the foundation for facilitating structured knowledge acquisition and systematic cognitive processes within human learning paradigms. However, the potential flaws of periodicity modeling in Transformer affect the learning efficiency and establishment of underlying principles from data for large language models (LLMs) built upon it. In this paper, we demonstrate that integrating effective periodicity modeling can improve the learning efficiency and performance of LLMs. We introduce FANformer, which adapts Fourier Analysis Network (FAN) into attention mechanism to achieve efficient periodicity modeling, by modifying the feature projection process of attention mechanism. Extensive experimental results on language modeling show that FANformer consistently outperforms Transformer when scaling up model size and training tokens, underscoring its superior learning efficiency. Our pretrained FANformer-1B exhibits marked improvements on downstream tasks compared to open-source LLMs with similar model parameters or training tokens. Moreover, we reveal that FANformer exhibits superior ability to learn and apply rules for reasoning compared to Transformer. The results position FANformer as an effective and promising architecture for advancing LLMs.

PDF Details

NeurIPS Conference 2025 Conference Paper

Recursive Transformer: Boosting Reasoning Ability with State Stack

Kechi Zhang
Ge Li
Huangzhao Zhang
Yihong Dong
Jia Li
Jingjing Xu
Zhi Jin

The Transformer architecture has emerged as a landmark advancement within the broad field of artificial intelligence, effectively catalyzing the advent of large language models (LLMs). However, despite its remarkable capabilities and the substantial progress it has facilitated, the Transformer architecture still has some limitations. One such intrinsic limitation is its inability to effectively recognize regular expressions or deterministic context-free grammars. Standard Transformers lack an explicit mechanism for recursion and structured state transitions, which can hinder systematic generalization on nested and hierarchical patterns. Drawing inspiration from pushdown automata, which efficiently resolve deterministic context-free grammars using stacks, we equip layers with a differentiable stack and propose StackTrans with recursion to address the aforementioned issue within LLMs. Unlike previous approaches that modify the attention computation, StackTrans explicitly incorporates hidden state stacks between Transformer layers. This design maintains compatibility with existing frameworks like flash-attention. Specifically, our design features stack operations -- such as pushing and popping hidden states -- that are differentiable and can be learned in an end-to-end manner. Our comprehensive evaluation spans benchmarks for both Chomsky hierarchy and large-scale natural languages. Across these diverse tasks, StackTrans consistently outperforms standard Transformer models and other baselines. We have successfully scaled StackTrans up from 360M to 7B parameters. In particular, our from-scratch pretrained model StackTrans-360M outperforms several larger open-source LLMs with 2–3x more parameters, showcasing its superior efficiency and reasoning capability.

PDF Details

NeurIPS Conference 2025 Conference Paper

SATURN: SAT-based Reinforcement Learning to Unleash LLMs Reasoning

Huanyu Liu
Ge Li
Jia Li
Hao Zhu
Kechi Zhang
Yihong Dong

How to design reinforcement learning (RL) tasks that effectively unleash the reasoning capability of large language models (LLMs) remains an open question. Existing RL tasks (e. g. , math, programming, and constructing reasoning tasks) suffer from three key limitations: (1) Scalability. They rely heavily on human annotation or expensive LLM synthesis to generate sufficient training data. (2) Verifiability. LLMs' outputs are hard to verify automatically and reliably. (3) Controllable Difficulty. Most tasks lack fine-grained difficulty control, making it hard to train LLMs to develop reasoning ability from easy to hard. To address these limitations, we propose Saturn, a SAT-based RL framework that uses Boolean Satisfiability (SAT) problems to train and evaluate LLMs reasoning. Saturn enables scalable task construction, rule-based verification, and precise difficulty control. Saturn designs a curriculum learning pipeline that continuously improves LLMs' reasoning capability by constructing SAT tasks of increasing difficulty and training LLMs from easy to hard. To ensure stable training, we design a principled mechanism to control difficulty transitions. We introduce Saturn-2. 6k, a dataset of 2, 660 SAT problems with varying difficulty. It supports the evaluation of how LLM reasoning changes with problem difficulty. We apply Saturn to DeepSeek-R1-Distill-Qwen and obtain Saturn-1. 5B and Saturn-7B. We achieve several notable results: (1) On SAT problems, Saturn-1. 5B and Saturn-7B achieve average pass@3 improvements of +14. 0 and +28. 1, respectively. (2) On math and programming tasks, Saturn-1. 5B and Saturn-7B improve average scores by +4. 9 and +1. 8 on benchmarks (e. g. , AIME, LiveCodeBench). (3) Compared to the state-of-the-art (SOTA) approach in constructing RL tasks, Saturn achieves further improvements of +8. 8\%. We release the source code, data, and models to support future research.

PDF Details

NeurIPS Conference 2024 Conference Paper

EvoCodeBench: An Evolving Code Generation Benchmark with Domain-Specific Evaluations

Jia Li
Ge Li
Xuanming Zhang
YunFei Zhao
Yihong Dong
Zhi Jin
Binhua Li
Fei Huang

How to evaluate Large Language Models (LLMs) in code generation remains an open question. Many benchmarks have been proposed, but they have two limitations, i. e. , data leakage and lack of domain-specific evaluation. The former hurts the fairness of benchmarks, and the latter hinders practitioners from selecting superior LLMs for specific programming domains. To address these two limitations, we propose a new benchmark - EvoCodeBench, which has the following advances: (1) Evolving data. EvoCodeBench will be dynamically updated every period (e. g. , 6 months) to avoid data leakage. This paper releases the first version - EvoCodeBench-2403, containing 275 samples from 25 repositories. (2) A domain taxonomy and domain labels. Based on the statistics of open-source communities, we design a programming domain taxonomy consisting of 10 popular domains. Based on the taxonomy, we annotate each sample in EvoCodeBench with a domain label. EvoCodeBench provides a broad platform for domain-specific evaluations. (3) Domain-specific evaluations. Besides the Pass@k, we compute the Domain-Specific Improvement (DSI) and define LLMs' comfort and strange domains. These evaluations help practitioners select superior LLMs in specific domains and discover the shortcomings of existing LLMs. Besides, EvoCodeBench is collected by a rigorous pipeline and aligns with real-world repositories in multiple aspects (e. g. , code distributions). We evaluate 8 popular LLMs (e. g. , gpt-4, DeepSeek Coder, StarCoder 2) on EvoCodeBench and summarize some insights. EvoCodeBench reveals the actual abilities of these LLMs in real-world repositories. For example, the highest Pass@1 of gpt-4 on EvoCodeBench-2403 is only 20. 74%. Besides, we evaluate LLMs in different domains and discover their comfort and strange domains. For example, gpt-4 performs best in most domains but falls behind others in the Internet domain. StarCoder 2-15B unexpectedly performs well in the Database domain and even outperforms 33B LLMs. We release EvoCodeBench, all prompts, and LLMs' completions for further community analysis.

PDF Details DOI

ECAI Conference 2023 Conference Paper

Antecedent Predictions Are More Important Than You Think: An Effective Method for Tree-Based Code Generation

Yihong Dong
Ge Li 0001
Xue Jiang
Zhi Jin 0001

Code generation focuses on automatically converting natural language (NL) utterances into code snippets. Sequence-to-tree (Seq2Tree) approaches are proposed for code generation with the aim of ensuring grammatical correctness of the generated code. These approaches generate subsequent Abstract Syntax Tree (AST) nodes based on the preceding predictions of AST nodes. However, existing Seq2Tree approaches tend to treat both antecedent predictions and subsequent predictions equally, which poses a challenge for models to produce accurate subsequent predictions if the antecedent predictions are incorrect under the constraints of the AST. Given this challenge, it is necessary to pay more attention to antecedent predictions compared to subsequent predictions. To this end, this paper proposes a novel and effective method, named Antecedent Prioritized (AP) Loss, which prioritizes antecedent predictions by leveraging the position information of the generated AST nodes. We design an AST-to-Vector (AST2Vec) method that maps AST node positions to two-dimensional vectors, thereby modeling the position information of AST nodes. To evaluate the effectiveness of our proposed loss, we implement and train an Antecedent Prioritized Tree-based code generation model called APT. Experiments on four benchmark datasets demonstrate that with better antecedent predictions and accompanying subsequent predictions, APT achieves significant improvements, indicating the superiority and generality of our proposed method.

Details