paper-with-me

홈 › Papers

SWE-IF: Aligning Code Evaluation with Human Preference

2025-10-08 · Ming Zhong, Xiang Zhou, Ting-Yun Chang, Qingze Wang, Nan Xu, Xiance Si, Dan Garrette, Shyam Upadhyay, Jeremiah Liu, Jiawei Han, Benoit Schillings, Jiao Sun arxiv

Large Language Models (LLMs) have catalyzed vibe coding, where users leverage LLMs to generate and iteratively refine code through natural language interactions until it passes their vibe check. Vibe check reflects human preference and goes beyond functionality: the solution should feel right, read cleanly, preserve intent, and remain correct. However, current code evaluation remains anchored to pass@k and captures only functional correctness, overlooking non-functional instructions that users routinely apply. In this paper, we hypothesize that instruction following is the missing piece underlying vibe check besides functional correctness. To quantify models' code instruction-following capabilities with measurable signals, we present VeriCode, a taxonomy of 30 verifiable code instructions together with deterministic verifiers. We use the taxonomy to augment established evaluation suites, resulting in SWE-IF, a testbed to assess both instruction following and functional correctness. Evaluating 31 LLMs, we show that even the strongest models struggle to comply with multiple instructions and exhibit functional regression. Most importantly, a composite score of functional correctness and instruction following correlates best with human preference, with instruction following emerging as the primary differentiator among LLMs. Our code, data, and taxonomy are available at https://github.com/maszhongming/SWE-IF.

📄 PDF Abstract BibTeX arXiv:2510.07315

Code (0)

등록된 구현이 없습니다.

Tasks

Instruction Following

Similar Papers 제목 키워드 기반

Diffusion-RPO: Aligning Diffusion Models through Relative Preference Optimization

2024-06-10 · Yi Gu, Zhendong Wang, Yueqin Yin, Yujia Xie 외

Aligning large language models with human preferences has emerged as a critical focus in language modeling research. Yet, integrating preference learning into Text-to-Image (T2I) generative models is still relatively unc…

Language ModelingLanguage Modelling

General Preference Modeling with Preference Representations for Aligning Language Models

2024-10-03 · Yifan Zhang, Ge Zhang, Yue Wu, Kangping Xu 외

Modeling human preferences is crucial for aligning foundation models with human values. Traditional reward modeling methods, such as the Bradley-Terry (BT) reward model, fall short in expressiveness, particularly in addr…

Language ModellingRepresentation Learning

Multiplayer Nash Preference Optimization

2025-09-27 · Fang Wu, Xu Huang, Weihao Xuan, Zhiwei Zhang 외 arxiv

Reinforcement learning from human feedback (RLHF) has emerged as the standard paradigm for aligning large language models with human preferences. However, reward-based methods grounded in the Bradley-Terry assumption str…

Reinforcement Learning

Learning Code Preference via Synthetic Evolution

2024-10-04 · Jiawei Liu, Thanh Nguyen, Mingyue Shang, Hantian Ding 외

Large Language Models (LLMs) have recently demonstrated remarkable coding capabilities. However, assessing code generation based on well-formed properties and aligning it with developer preferences remains challenging. I…

Code Generation

WildFeedback: Aligning LLMs With In-situ User Interactions And Feedback

2024-08-28 · Taiwei Shi, Zhuoer Wang, Longqi Yang, Ying-Chun Lin 외

As large language models (LLMs) continue to advance, aligning these models with human preferences has emerged as a critical challenge. Traditional alignment methods, relying on human or LLM annotated datasets, are limite…