# Answer First, Reason Later: Aligning Search Relevance via Mode-Balanced Reinforcement Learning

arXiv: https://arxiv.org/abs/2602.10006 · HTML: https://mlsanity.com/paper/2602.10006/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Knowledge- and tool-augmented reasoning, one of 17 directions in LLM reasoning.
- Builds on 25 of its 41 references that are in the map.
- First posted on arXiv 10 Feb 2026.

## Placed in

- LLM reasoning / [Knowledge- and tool-augmented reasoning](https://mlsanity.com/lr/knowledge-and-tool-augmented-reasoning/)

## Builds on (most cited first, 25 of 25)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning](https://mlsanity.com/paper/2503.09516/) (2025)
- [OpenAI o1 System Card](https://mlsanity.com/paper/2412.16720/) (2024)
- [Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?](https://mlsanity.com/paper/2504.13837/) (2025)
- [Is ChatGPT Good at Search? Investigating Large Language Models as Re-Ranking Agents](https://mlsanity.com/paper/2304.09542/) (2023)
- [Reinforcement Learning for Reasoning in Large Language Models with One Training Example](https://mlsanity.com/paper/2504.20571/) (2025)
- [Fine-Tuning LLaMA for Multi-Stage Text Retrieval](https://mlsanity.com/paper/2310.08319/) (2023)
- [The Landscape of Agentic Reinforcement Learning for LLMs: A Survey](https://mlsanity.com/paper/2509.02547/) (2025)
- [ReSearch: Learning to Reason with Search for LLMs via Reinforcement Learning](https://mlsanity.com/paper/2503.19470/) (2025)
- [Retaining by Doing: The Role of On-Policy Data in Mitigating Forgetting](https://mlsanity.com/paper/2510.18874/) (2025)
- [AgentEvolver: Towards Efficient Self-Evolving Agent System](https://mlsanity.com/paper/2511.10395/) (2025)
- [Reinforcement Learning Enhanced LLMs: A Survey](https://mlsanity.com/paper/2412.10400/) (2024)
- [Rank-R1: Enhancing Reasoning in LLM-based Document Rerankers via Reinforcement Learning](https://mlsanity.com/paper/2503.06034/) (2025)
- [Reinforcement Learning Meets Large Language Models: A Survey of Advancements and Applications Across the LLM Lifecycle](https://mlsanity.com/paper/2509.16679/) (2025)
- [A Survey of Generative Search and Recommendation in the Era of Large Language Models](https://mlsanity.com/paper/2404.16924/) (2024)
- [SEEA-R1: Tree-Structured Reinforcement Fine-Tuning for Self-Evolving Embodied Agents](https://mlsanity.com/paper/2506.21669/) (2025)
- [CLPO: Curriculum Learning meets Policy Optimization for LLM Reasoning](https://mlsanity.com/paper/2509.25004/) (2025)
- [Consolidating Ranking and Relevance Predictions of Large Language Models through Post-Processing](https://mlsanity.com/paper/2404.11791/) (2024)
- [ETR: Outcome-Guided Elastic Trust Regions for Policy Optimization](https://mlsanity.com/paper/2601.03723/) (2026)

## Abstract (arXiv)

Building a search relevance model that achieves both low latency and high performance is a long-standing challenge in the search industry. To satisfy the millisecond-level response requirements of online systems while retaining the interpretable reasoning traces of Large Language Models (LLMs), we propose a novel Answer-First, Reason Later (AFRL) paradigm. This paradigm requires the model to output the definitive relevance score in the very first token, followed by a structured logical explanation. Inspired by the success of reasoning models, we adopt a "Supervised Fine-Tuning (SFT) + Reinforcement Learning (RL)" pipeline to achieve AFRL. However, directly applying existing RL training often leads to mode collapse in the search relevance task, where the model forgets complex long-tail rules in pursuit of high rewards. From an information theory perspective: RL inherently minimizes the Reverse KL divergence, which tends to seek probability peaks (mode-seeking) and is prone to "reward hacking." On the other hand, SFT minimizes the Forward KL divergence, forcing the model to cover the data distribution (mode-covering) and effectively anchoring expert rules. Based on this insight, we propose a Mode-Balanced Optimization strategy, incorporating an SFT auxiliary loss into Stepwise-GRPO training to balance these two properties. Furthermore, we construct an automated instruction evolution system and a multi-stage curriculum to ensure expert-level data quality. Extensive experiments demonstrate that our 32B teacher model achieves state-of-the-art performance. Moreover, the AFRL architecture enables efficient knowledge distillation, successfully transferring expert-level logic to a 0.6B model, thereby reconciling reasoning depth with deployment latency.