Mitigating Exploration Bias in RL for Multi-Instruction Following

• URL: http://arxiv.org/abs/2608.23830v1
• Date: Mon, 24 Aug 2026 21:19:46 GMT
• Title: Mitigating Exploration Bias in RL for Multi-Instruction Following
• Abstract summary: We find that existing training recipes suffer from exploration bias towards easy instructions when the training data has multiple instructions in a prompt.<n>This bias is caused by two main reasons: 1) the policy model's initial ability to satisfy hard instructions is too low to trigger successful exploration during RL training, so the optimization is biased towards easy instructions.<n>We propose two metrics to measure the exploration bias in instruction following and then introduce a two-stage framework to alleviate it.
• Score: 45.91710862879366
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: RL has emerged as a powerful paradigm for enhancing the instruction following capabilities of LLMs. While existing training recipes achieve substantial gains, we find that they suffer from exploration bias towards easy instructions when the training data has multiple instructions in a prompt. This bias is caused by two main reasons: 1) the policy model's initial ability to satisfy hard instructions is too low to trigger successful exploration during RL training, so the optimization is biased towards easy instructions; and 2) canonical RL training recipes typically employ a cumulative reward (the number of instructions fulfilled), treating all instructions equally, which biases the policy model towards fulfilling easy instructions to obtain the same amount of reward. To address these issues, we first propose two metrics to measure the exploration bias in instruction following and then introduce a two-stage framework to alleviate it: 1) Behavioral Bootstrapping, a lightweight rejection sampling fine-tuning stage before RL to activate hard instructions; and 2) Scarcity-Aware Rewards, a new RL reward function that assigns rewards to instructions based on their empirical scarcity. Experiments show that the proposed metrics are highly correlated with model performance, and our methods unleash the potential of RL training: our best models outperform the baselines by a significant margin across three verifiable instruction following benchmarks. We release codes at https://github.com/mianzhang/MulIF.

Related papers

• Replay Failures as Successes: Sample-Efficient Reinforcement Learning for Instruction Following [42.05102776289243]
  Reinforcement Learning (RL) has shown promise for aligning Large Language Models (LLMs) to follow instructions with various constraints.<n>We propose Hindsight instruction Replay (HiR), a novel sample-efficient RL framework for complex instruction following tasks.
  arXiv  Detail & Related papers  (2025-12-29T13:31:08Z)
• CodeBoost: Boosting Code LLMs by Squeezing Knowledge from Code Snippets with RL [28.43882967593511]
  Code large language models (LLMs) have become indispensable tools for building efficient and automated coding pipelines.<n>Existing models are typically post-trained using reinforcement learning (RL) from general-purpose LLMs using "human instruction-final answer" pairs.<n>We propose CodeBoost, a framework that enhances code LLMs purely from code snippets, without relying on human-annotated instructions.
  arXiv  Detail & Related papers  (2025-08-07T10:31:24Z)
• Beyond Accuracy: Dissecting Mathematical Reasoning for LLMs Under Reinforcement Learning [93.00629872970364]
  Reinforcement learning (RL) has become the dominant paradigm for improving the performance of language models on complex reasoning tasks.<n>We introduce SPARKLE, a fine-grained analytic framework to dissect the effects of RL across three key dimensions.<n>We study whether difficult problems -- those yielding no RL signals and mixed-quality reasoning traces -- can still be effectively used for training.
  arXiv  Detail & Related papers  (2025-06-05T07:53:59Z)
• SeRL: Self-Play Reinforcement Learning for Large Language Models with Limited Data [65.56911325914582]
  We propose Self-play Reinforcement Learning (SeRL) to bootstrap Large Language Models (LLMs) training with limited initial data.<n>The proposed SeRL yields results superior to its counterparts and achieves performance on par with those obtained by high-quality data with verifiable rewards.
  arXiv  Detail & Related papers  (2025-05-25T13:28:04Z)
• AceReason-Nemotron: Advancing Math and Code Reasoning through Reinforcement Learning [50.02117478165099]
  We show that large-scale reinforcement learning can significantly enhance the reasoning capabilities of strong, small- and mid-sized models.<n>We propose a simple yet effective approach: first training on math-only prompts, then on code-only prompts.
  arXiv  Detail & Related papers  (2025-05-22T08:50:47Z)
• On the Emergence of Thinking in LLMs I: Searching for the Right Intuition [34.32871896067864]
  We propose a post-training framework called Reinforcement Learning via Self-Play (RLSP)<n> RLSP involves three steps: supervised fine-tuning with human or synthetic demonstrations of the reasoning process, using an exploration reward signal to encourage diverse and efficient reasoning behaviors, and RL training with an outcome verifier to ensure correctness while preventing reward hacking.<n> Empirical studies in the math domain show that RLSP improves reasoning.
  arXiv  Detail & Related papers  (2025-02-10T18:52:04Z)
• Supervised Advantage Actor-Critic for Recommender Systems [76.7066594130961]
  We propose negative sampling strategy for training the RL component and combine it with supervised sequential learning. Based on sampled (negative) actions (items), we can calculate the "advantage" of a positive action over the average case. We instantiate SNQN and SA2C with four state-of-the-art sequential recommendation models and conduct experiments on two real-world datasets.
  arXiv  Detail & Related papers  (2021-11-05T12:51:15Z)
• URLB: Unsupervised Reinforcement Learning Benchmark [82.36060735454647]
  We introduce the Unsupervised Reinforcement Learning Benchmark (URLB) URLB consists of two phases: reward-free pre-training and downstream task adaptation with extrinsic rewards. We provide twelve continuous control tasks from three domains for evaluation and open-source code for eight leading unsupervised RL methods.
  arXiv  Detail & Related papers  (2021-10-28T15:07:01Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.