paper-with-me

Papers

Code Comparison Tuning for Code Large Language Models

2024-03-28 · Yufan Jiang, Qiaozhi He, Xiaomin Zhuang, Zhihua Wu

We present Code Comparison Tuning (CCT), a simple and effective tuning method for code large language models (Code LLMs) to better handle subtle code errors. Specifically, we integrate the concept of comparison into instruction tuning, both at the token and sequence levels, enabling the model to discern even the slightest deviations in code. To compare the original code with an erroneous version containing manually added code errors, we use token-level preference loss for detailed token-level comparisons. Additionally, we combine code segments to create a new instruction tuning sample for sequence-level comparisons, enhancing the model's bug-fixing capability. Experimental results on the HumanEvalFix benchmark show that CCT surpasses instruction tuning in pass@1 scores by up to 4 points across diverse code LLMs, and extensive analysis demonstrates the effectiveness of our method.

📄 PDF Abstract BibTeX arXiv:2403.19121

Code (0)

등록된 구현이 없습니다.

Tasks

Bug fixing

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Position-Wise Feed-Forward Layer 설명 없음
Residual Connection 설명 없음
Convolution A convolution is a type of matrix operation, consisting of a kernel, a small matrix of weights, that slides over input data performing element-wise multiplication with the…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…

Similar Papers 제목 키워드 기반

PyraNet: A Multi-Layered Hierarchical Dataset for Verilog

2024-12-09 · Bardia Nadimi, Ghali Omar Boutaib, Hao Zheng

Recently, there has been a growing interest in leveraging Large Language Models for Verilog code generation. However, the current quality of the generated Verilog code remains suboptimal. This is largely due to the absen…

Code Generation

A Comparative Study between Full-Parameter and LoRA-based Fine-Tuning on Chinese Instruction Data for Instruction Following Large Language Model

2023-04-17 · Xianghui Sun, Yunjie Ji, Baochang Ma, Xiangang Li

Recently, the instruction-tuning of large language models is a crucial area of research in the field of natural language processing. Due to resource and cost limitations, several researchers have employed parameter-effic…

Instruction FollowingLanguage ModelingLanguage ModellingLarge Language Model

BEFT: Bias-Efficient Fine-Tuning of Language Models in Low-Data Regimes

2025-09-19 · Baichuan Huang, Ananth Balashankar, Amir Aminifar arxiv

Fine-tuning the bias terms of large language models (LLMs) has the potential to achieve unprecedented parameter efficiency while maintaining competitive performance, particularly in low-data regimes. However, the link be…

Large Language Models Meet NL2Code: A Survey

2022-12-19 · Daoguang Zan, Bei Chen, Fengji Zhang, Dianjie Lu 외

The task of generating code from a natural language description, or NL2Code, is considered a pressing and significant challenge in code intelligence. Thanks to the rapid development of pre-training techniques, surging la…

HumanEvalSurvey

From Prompting to Preference Optimization: A Comparative Study of LLM-based Automated Essay Scoring

2026-03-06 · Minh Hoang Nguyen, Vu Hoang Pham, Xuan Thanh Huynh, Phuc Hong Mai 외 arxiv

Large language models (LLMs) have recently reshaped Automated Essay Scoring (AES), yet prior studies typically examine individual techniques in isolation, limiting understanding of their relative merits for English as a …

Automated Essay Scoring