Demystifying GPT Self-Repair for Code Generation

By Theo X. Olausson, Jeevana Priya Inala, Chenglong Wang, Jianfeng Gao, Armando Solar-Lezama. 2023

81 citations recorded by the source corpus.

Subfields: Hardware and Architecture, Software, Large Language Models, Code Generation & Program Synthesis

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Hardware and Architecture

1. Armando Solar-Lezama
2. Xinyun Chen
3. Koushik Sen
4. Ion Stoica
5. Hanxiao Liu
6. Quoc V. Le
7. Furu Wei
8. Tao Ge

In Software

1. Armando Solar-Lezama
2. Xinyun Chen
3. Koushik Sen
4. Naman Jain
5. Tianjun Zhang
6. Ion Stoica
7. Quoc V. Le
8. Adams Wei Yu

In Large Language Models

1. Furu Wei
2. Denny Zhou
3. Jie Huang
4. Xinyun Chen
5. Xuezhi Wang
6. Swaroop Mishra
7. Quoc V. Le
8. Heng Ji

In Code Generation & Program Synthesis

1. Armando Solar-Lezama
2. Xinyun Chen
3. Koushik Sen
4. Quoc V. Le
5. Denny Zhou
6. Rishabh Singh

Citations in the corpus

• LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code (2024, methodology): “The self-repair scenario is based on previous works that tested the self-repair capabilities of LLMs (Olausson et al., 2023; Shinn et al., 2023; Chen et al., 2023).”
• Large Language Models as Optimizers (2023, background): “…al., 2022; Ganguli et al., 2023), improving reasoning (Shinn et al., 2023; Madaan et al., 2023) and code generation performance (Chen et al., 2023e; Olausson et al., 2023; Shinn et al., 2023; Chen et al., 2023b), dialogue applications (Nair et al., 2023; Madaan et al., 2023; Yuan et al., 2023),…”
• Unleashing the Emergent Cognitive Synergy in Large Language Models: A Task-Solving Agent through Multi-Persona Self-Collaboration (2023, background): “(2) SPP is fairly robust to the prompt change and show good performance with only the first demo example.”
• Large Language Models Cannot Self-Correct Reasoning Yet (2023, background): “In contrast, external feedback incorporates inputs from humans, other models (Wang et al., 2023; Paul et al., 2023, inter alia ), or external tools and knowledge sources (Gou et al., 2023; Chen et al., 2023b; Olausson et al., 2023; Gao et al., 2023, inter alia ).”