The Virtues of Brevity: Avoid Overthinking in Parallel Test-Time Reasoning
This work addresses computational efficiency for users of large language models in reasoning tasks, though it is incremental as it builds on existing parallel sampling methods.
The paper tackles the problem of high computational cost in parallel test-time reasoning by proposing a simple heuristic of selecting the shortest solution, which is shown to be competitive with complex methods like self-consistency on two benchmarks while reducing overhead.
Reasoning models represent a significant advance in LLM capabilities, particularly for complex reasoning tasks such as mathematics and coding. Previous studies confirm that parallel test-time compute-sampling multiple solutions and selecting the best one-can further enhance the predictive performance of LLMs. However, strategies in this area often require complex scoring, thus increasing computational cost and complexity. In this work, we demonstrate that the simple and counterintuitive heuristic of selecting the shortest solution is highly effective. We posit that the observed effectiveness stems from models operating in two distinct regimes: a concise, confident conventional regime and a verbose overthinking regime characterized by uncertainty, and we show evidence of a critical point where the overthinking regime begins to be significant. By selecting the shortest answer, the heuristic preferentially samples from the conventional regime. We confirm that this approach is competitive with more complex methods such as self-consistency across two challenging benchmarks while significantly reducing computational overhead. The shortest-answer heuristic provides a Pareto improvement over self-consistency and applies even to tasks where output equality is not well defined.