Reasoning as a Resource: Optimizing Fast and Slow Thinking in Code Generation Models

June 11, 2025 Β· Declared Dead Β· πŸ› Proceedings of the 1st ACM SIGPLAN International Workshop on Language Models and Programming Languages

πŸ‘» CAUSE OF DEATH: Ghosted
No code link whatsoever

"No code URL or promise found in abstract"

Evidence collected by the PWNC Scanner

Authors Zongjie Li, Shuai Wang arXiv ID 2506.09396 Category cs.SE: Software Engineering Cross-listed cs.AI Citations 2 Venue Proceedings of the 1st ACM SIGPLAN International Workshop on Language Models and Programming Languages Last Checked 4 months ago
Abstract
This position paper proposes a fundamental shift in designing code generation models: treating reasoning depth as a controllable resource. Rather than being an incidental byproduct of prompting, we argue that the trade-off between rapid, direct answers ("fast thinking") and elaborate, chain-of-thought deliberation ("slow thinking") must be explicitly managed. We contend that optimizing reasoning budgets across the entire model lifecycle - from synthetic data creation and benchmarking to real-world deploymen - can unlock superior trade-offs among accuracy, latency, and cost. This paper outlines how adaptive control over reasoning can enrich supervision signals, motivate new multi-dimensional benchmarks, and inform cost-aware, security-conscious deployment policies. By viewing fast and slow thinking as complementary modes to be scheduled, we envision coding agents that think deep when necessary and act fast when possible.
Community shame:
Not yet rated
Community Contributions

Found the code? Know the venue? Think something is wrong? Let us know!

πŸ“œ Similar Papers

In the same crypt β€” Software Engineering

Died the same way β€” πŸ‘» Ghosted