Dependency conflicts are common in the Python ecosystem, manifesting as incompatible version constraints, missing packages, and undocumented compatibility relations, which cause many code snippets to fail at runtime.
This paper introduces PLLM+, a hybrid dependency‑repair pipeline evaluated on the HG2.9K benchmark containing 2,891 snippets with failing dependencies. The pipeline first runs inexpensive deterministic stages: static AST‑based interpreter inference, replay of historically successful dependency configurations from the competition‑provided solutions database, and live PyPI validation of candidate package versions.
If those stages do not resolve the issue, the system falls back to a structured LLM‑based repair loop that includes typed error classification and Proposer/Critic agents.
On HG2.9K, PLLM+ solves 1,500 snippets, outperforming the PLLM baseline’s 1,169 solutions, while reducing average runtime from 368.7 seconds to 71.8 seconds per snippet. Among the successful fixes, 1,495 are produced by replaying known configurations, and the LLM fallback contributes only five additional fixes.
The results suggest that, in this benchmark setting, deterministic reuse of previously validated dependency configurations is a simple and effective strategy, with LLM‑based repair serving as a secondary fallback for uncovered cases. Review