Papers by Alexander Shypula
LLM Program Optimization via Retrieval Augmented Search (2026.findings-acl)
Copied to clipboard
| Challenge: | Recent work shows that large language models have difficulty with program optimization out-of-the-box. |
| Approach: | They propose a blackbox adaptation method that performs beam search over candidate optimizations by a training dataset. |
| Outcome: | The proposed method outperforms retrieval based on the source code in a number of ways. |
Explain-then-translate: an analysis on improving program translation with self-generated explanations (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Using self-generated natural language explanations improves zero-shot performance by 12% on average. |
| Approach: | They propose to use self-generated natural language explanations as an intermediate step for code-to-code translation with language models. |
| Outcome: | The proposed approach improves zero-shot performance by 12% on average . the proposed approach is not evaluated on a broader set of languages including low-resource languages. |