Recursive Think-Answer Process for LLMs and VLMs

Researchers propose Recursive Think-Answer Process (R-TAP), a method enabling large language models (LLMs) and vision-language models (VLMs) to iteratively refine their answers, outperforming single-pass approaches. R-TAP uses a confidence generator and two complementary rewards—Recursively Confidence Increase Reward and Final Answer Confidence Reward—to improve accuracy and reduce self-reflective errors like 'Oops!' in models such as DeepSeek-R1.

Think-Answer reasoners such as DeepSeek-R1 have made notable progress by leveraging interpretable internal reasoning. However, despite the frequent presence of self-reflective cues like "Oops ", they remain vulnerable to output errors during single-pass inference. To address this limitation, we propose an efficient Recursive Think-Answer Process R-TAP that enables models to engage in iterative reasoning cycles and generate more accurate answers, going beyond conventional single-pass approaches. Central to this approach is a confidence generator that evaluates the certainty of model responses and guides subsequent improvements. By incorporating two complementary rewards-Recursively Confidence Increase Reward and Final Answer Confidence Reward-we show that R-TAP-enhanced models consistently outperform conventional single-pass methods for both large language models LLMs and vision-language models VLMs . Moreover, by analyzing the frequency of "Oops"-like expressions in model responses, we find that R-TAP-applied models exhibit significantly fewer self-reflective patterns, resulting in more stable and faster inference-time reasoning. We hope R-TAP pave the way evolving into efficient and elaborated methods to refine the reasoning processes of future AI.