Reading List

1. LLM Basics:

· Attention Is All You Need. Vaswani, et al. https://arxiv.org/abs/1706.03762

· GPT-4 Technical Report. OpenAI. https://arxiv.org/abs/2303.08774

2. LLM for Code

· Evaluating Large Language Models Trained on Code. Chen, et al. https://arxiv.org/abs/2107.03374.

· Code Llama: Open Foundation Models for Code. Rozière, et al. https://ai.meta.com/research/publications/code-llama-open-foundation-models-for-code/.

3. LLM as Programming Assistant

· A Large-Scale Survey on the Usability of AI Programming Assistants: Successes and Challenges. Liang, et al. https://dl.acm.org/doi/abs/10.1145/3597503.3608128.

· GitHub Copilot AI pair programmer: Asset or Liability? Dakhel, et al. https://arxiv.org/abs/2206.15331.

4. LLM for Collaborative Coding

· MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework. Hong, et al. https://arxiv.org/abs/2308.00352.

· Communicative Agents for Software Development. Qian, et al. https://arxiv.org/abs/2307.07924.

5. Augmented LLM with Tools

· Augmented Language Models: a Survey. Mialon, et al. https://arxiv.org/abs/2302.07842.

· Toolformer: Language Models Can Teach Themselves to Use Tools. Schick, et al. https://arxiv.org/abs/2302.04761.

6. LLM for Unit Testing

· Automated Unit Test Improvement using Large Language Models at Meta. Alshahwan and et al. https://arxiv.org/abs/2402.09171.

· An Empirical Evaluation of Using Large Language Models for Automated Unit Test Generation. Schäfer and et al. https://arxiv.org/abs/2302.06527.

7. LLM for Bug Hunting

· Large Language Models are Few-shot Testers: Exploring LLM-based General Bug Reproduction. Kang and et al. https://arxiv.org/abs/2209.11515.

· PentestGPT: An LLM-empowered Automatic Penetration Testing Tool. Deng, et al. https://arxiv.org/abs/2308.06782.

8. LLM for Debugging

· Teaching Large Language Models to Self-Debug. Chen, et al. https://arxiv.org/abs/2304.05128.

· Reflexion: Language Agents with Verbal Reinforcement Learning. Shinn, et al. https://arxiv.org/abs/2303.11366.

9. Reasoning with LLM

· Chain-of-Thought Prompting Elicits Reasoning in Large Language Models. Wei, et al. https://arxiv.org/abs/2201.11903.

· Program of Thoughts Prompting: Disentangling Computation from Reasoning for Numerical Reasoning Tasks. Chen, et al. https://arxiv.org/abs/2211.12588.

10. LLM for Theorem Proving

· Generative Language Modeling for Automated Theorem Proving. Polu and Sutskever. https://arxiv.org/abs/2009.03393.

· LeanDojo: Theorem Proving with Retrieval-Augmented Language Models. Yand, et al. https://proceedings.neurips.cc/paper_files/paper/2023/hash/4441469427094f8873d0fecb0c4e1cee-Abstract-Datasets_and_Benchmarks.html