Papers by Zhenpeng Chen
Personality-Guided Code Generation Using Large Language Models (2025.acl-long)
Copied to clipboard
| Challenge: | Existing studies have shown that personality-guided code generation improves software development outcomes when individuals are assigned tasks that match their personality types. |
| Approach: | They evaluate how emulating personality traits appropriate to the coding tasks affects LLM performance by using seven widely adopted LLMs. |
| Outcome: | The proposed approach improves pass rates in 23 out of 28 LLM-dataset combinations, while emulating personality traits can be easily integrated with other prompting strategies to further boost performance. |
Temporal Scaling Law for Large Language Models (2025.emnlp-main)
Copied to clipboard
Yizhe Xiong, Xiansheng Chen, Xin Ye, Hui Chen, Zijia Lin, Haoran Lian, Zhenpeng Su, Wei Huang, Jianwei Niu, Jungong Han, Guiguang Ding
| Challenge: | Existing studies have found that the test loss of LLMs scales as power-laws with model size, computational budget, and dataset size. |
| Approach: | They propose a concept of Temporal Scaling Law to study test loss of LLMs . they break down test loss into fine-grained token positions and develop a dynamic hyperbolic-law . |
| Outcome: | The proposed model predicts the test loss of LLMs as the training steps scale up. |
CartesianMoE: Boosting Knowledge Sharing among Experts via Cartesian Product Routing in Mixture-of-Experts (2025.naacl-long)
Copied to clipboard
Zhenpeng Su, Xing W, Zijia Lin, Yizhe Xiong, Minxuan Lv, Guangyuan Ma, Hui Chen, Songlin Hu, Guiguang Ding
| Challenge: | Large language models (LLMs) have been attracting much attention due to their impressive performance in all kinds of downstream tasks. |
| Approach: | They propose a mix-of-experts model that allows the model size to grow without raising training costs. |
| Outcome: | The proposed model outperforms existing models in perplexity and robustness tests. |
EET: Experience-Driven Early Termination for Cost-Efficient Software Engineering Agents (2026.findings-acl)
Copied to clipboard
| Challenge: | Large language models are reshaping modern software development, but they often incur substantial monetary cost. |
| Approach: | They propose an experience-driven early termination approach that extracts structured experience from prior issue-resolution executions and leverages it to guide early termination during patch generation and selection. |
| Outcome: | The proposed approach reduces cost by 19%–55% with negligible loss in resolution rate (at most 0.2%) EET extracts structured experience from prior issue-resolution executions and leverages it to guide early termination during patch generation and selection. |
Document Hashing with Multi-Grained Prototype-Induced Hierarchical Generative Model (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing document hashing methods only consider flat semantics of documents, preserving hierarchical semantics. |
| Approach: | They propose a hierarchical generative model that can model and leverage hierarchic semantics . they introduce hierarchically-based prototypes into the model to construct a Hierarchical prior distribution . |
| Outcome: | The proposed model outperforms baseline methods on hierarchical and flat datasets. |
LLM-Powered Test Case Generation for Detecting Bugs in Plausible Programs (2025.acl-long)
Copied to clipboard
Kaibo Liu, Zhenpeng Chen, Yiyang Liu, Jie M. Zhang, Mark Harman, Yudong Han, Yun Ma, Yihong Dong, Ge Li, Gang Huang
| Challenge: | TrickCatcher generates test cases that pass existing tests yet contain bugs . a recent study found that tricky bugs are not detected by test suites . |
| Approach: | They propose an LLM-powered approach to generating test cases for uncovering bugs in plausible programs . they use a PUT and specification to generate program variants, an input generator and an Llm to construct test inputs . |
| Outcome: | The proposed approach achieves recall, precision, and F1 scores that are 1.80, 2.65, and 1.66 . trickCatcher generates program variants based on the program under test and its specification . |
Good Reasoning Makes Good Demonstrations: Implicit Reasoning Quality Supervision via In-Context Reinforcement Learning (2026.findings-acl)
Copied to clipboard
| Challenge: | Reinforcement Learning with Verifiable Rewards (RLVR) improves reasoning in large language models but treats all correct solutions equally, potentially reinforcing flawed traces that arrive at correct answers by chance. |
| Approach: | They propose a method that reweights rewards by a factor approximately proportional to Evidence Gain and assigns higher weights to high-quality traces without requiring costly computation. |
| Outcome: | Experiments on mathematical reasoning benchmarks show that Reinforcement Learning with Verifiable Rewards (RLVR) improves reasoning in large language models but treats all correct solutions equally. |
DSMoE: Matrix-Partitioned Experts with Dynamic Routing for Computation-Efficient Dense LLMs (2025.emnlp-main)
Copied to clipboard
Minxuan Lv, Zhenpeng Su, Leiyu Pan, Yizhe Xiong, Zijia Lin, Hui Chen, Wei Zhou, Jungong Han, Guiguang Ding, Wenwu Ou, Di Zhang, Kun Gai, Songlin Hu
| Challenge: | Existing sparsification methods like pruning can lose model knowledge through parameter removal. |
| Approach: | They propose a novel approach that achieves sparsification by partitioning pre-trained FFN layers into computational blocks. |
| Outcome: | The proposed approach achieves superior performance across language modeling and downstream tasks under equivalent computational constraints. |
MiLe Loss: a New Loss for Mitigating the Bias of Learning Difficulties in Generative Language Models (2024.findings-naacl)
Copied to clipboard
| Challenge: | Existing generative language models neglect an inherent challenge in text corpus during training, i.e., the imbalance between frequent tokens and infrequent ones. |
| Approach: | They propose a function to mitigate the imbalance between frequent and infrequent tokens . authors propose 'MiLe Loss' function to assess learning difficulty of tokens during training . |
| Outcome: | Experiments show that models with proposed model can improve on downstream benchmarks. |