{"ID":2847473,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.02854","arxiv_id":"2511.02854","title":"SELF-REDRAFT: Eliciting Intrinsic Exploration-Exploitation Balance in Test-Time Scaling for Code Generation","abstract":"Test-time scaling without interpreter feedback is essential for real-world code generation scenarios where test cases are not readily available. While existing paradigms often rely on either greedy exploitation (i.e., iterative refinement) or stochastic exploration (i.e., relying on sample-based voting or reranking mechanisms), the balance between these two dimensions remains underexplored. To investigate the LLM's intrinsic ability to balance exploitation and exploration, we introduce SELF-REDRAFT, a framework built upon Self-Refine that encourages the model to propose new drafts for solutions that are fundamentally flawed. Our results show that SELF-REDRAFT consistently achieves better performance than Self-Refine when converged under the same maximum number of iterations. Still, we observe that significant room for improvement remains, largely due to two core aspects of current self-redraft capabilities: constrained capacity for generating instructive feedback and fragile discriminative judgment. We also find that balancing strategies vary notably across different LLMs, reflecting distinct, model-specific behaviors. Overall, our study establishes a baseline for intrinsic exploration-exploitation balancing in test-time scaling and identifies feedback and discrimination as key areas with potential for future advances.","short_abstract":"Test-time scaling without interpreter feedback is essential for real-world code generation scenarios where test cases are not readily available. While existing paradigms often rely on either greedy exploitation (i.e., iterative refinement) or stochastic exploration (i.e., relying on sample-based voting or reranking mec...","url_abs":"https://arxiv.org/abs/2511.02854","url_pdf":"https://arxiv.org/pdf/2511.02854v1","authors":"[\"Yixiang Chen\",\"Tianshi Zheng\",\"Shijue Huang\",\"Zhitao He\",\"Yi R. Fung\"]","published":"2025-10-31T06:30:47Z","proceeding":"cs.SE","tasks":"[\"cs.SE\",\"cs.AI\"]","methods":"[\"Large Language Model\",\"LoRA\"]","has_code":false}