A position paper arguing that PL techniques, especially formal verification and structure-aware representations, should be deeply integrated into LLM code generation.
Reasoning as a Resource: Optimizing Fast and Slow Thinking in Code Generation Models
1 Pith paper cite this work. Polarity classification is still indexing.
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.
citation-role summary
citation-polarity summary
fields
cs.SE 1years
2025 1verdicts
UNVERDICTED 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Position Paper: Programming Language Techniques for Bridging LLM Code Generation Semantic Gaps
A position paper arguing that PL techniques, especially formal verification and structure-aware representations, should be deeply integrated into LLM code generation.