Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

VerIF: Verification Engineering for Reinforcement Learning in Instruction Following

About

Reinforcement learning with verifiable rewards (RLVR) has become a key technique for enhancing large language models (LLMs), with verification engineering playing a central role. However, best practices for RL in instruction following remain underexplored. In this work, we explore the verification challenge in RL for instruction following and propose VerIF, a verification method that combines rule-based code verification with LLM-based verification from a large reasoning model (e.g., QwQ-32B). To support this approach, we construct a high-quality instruction-following dataset, VerInstruct, containing approximately 22,000 instances with associated verification signals. We apply RL training with VerIF to two models, achieving significant improvements across several representative instruction-following benchmarks. The trained models reach state-of-the-art performance among models of comparable size and generalize well to unseen constraints. We further observe that their general capabilities remain unaffected, suggesting that RL with VerIF can be integrated into existing RL recipes to enhance overall model performance. We have released our datasets, codes, and models to facilitate future research at https://github.com/THU-KEG/VerIF.

Hao Peng, Yunjia Qi, Xiaozhi Wang, Bin Xu, Lei Hou, Juanzi Li• 2025

Related benchmarks

TaskDatasetResultRank
Instruction FollowingFollowBench
HSR56.9
85
Instruction FollowingCF-Bench
Instruction Success Rate41
68
Instruction FollowingIFEval--
65
Instruction FollowingMulti-IF
Score66.2
41
Instruction FollowingAgentIF
CSR56.6
29
Instruction FollowingWritingBench
Average Score50.8
29
Instruction FollowingIFEval In-Domain
Precision (L)0.871
23
Instruction FollowingAgentIF (Out-of-Domain)
CSR56.6
23
Instruction FollowingFollowBench In-Domain
HSR56.9
23
Instruction FollowingCFBench In-Domain
ISR41
23
Showing 10 of 14 rows

Other info

Follow for update