LLM code modernizers produce semantic drift in 39.7% of legacy-Python-2 cases and endorse 31.7% of those drifts in self-review, with rates varying widely across models but not tracking capability.
Environment-in-the-loop: rethinking code migration with LLM-based agents
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 2roles
background 1polarities
background 1representative citing papers
A survey that organizes existing work on LLM-based agents around code as the central harness, structured in three layers of interfaces, mechanisms, and multi-agent scaling, with applications across domains and listed open challenges.
citing papers explorer
-
Articulate but Wrong: Self-Review Failures in LLM-Based Code Modernization
LLM code modernizers produce semantic drift in 39.7% of legacy-Python-2 cases and endorse 31.7% of those drifts in self-review, with rates varying widely across models but not tracking capability.
-
Code as Agent Harness
A survey that organizes existing work on LLM-based agents around code as the central harness, structured in three layers of interfaces, mechanisms, and multi-agent scaling, with applications across domains and listed open challenges.