DecIF: Improving Instruction-Following through Meta-Decomposition
Instruction-following has emerged as a crucial capability for large language models (LLMs). However, existing approaches often rely on pre-existing documents or external resources to synthesize instruction-following data, which limits their flexibility and generalizability. In this paper, we introduce DecIF, a fully autonomous, meta-decomposition guided framework that generates diverse and high-quality instruction-following data using only LLMs. DecIF is grounded in the principle of decomposition. For instruction generation, we guide LLMs to iteratively produce various types of meta-information, which are then combined with response constraints to form well-structured and semantically rich instructions. We further utilize LLMs to detect and resolve potential inconsistencies within the generated instructions. Regarding response generation, we decompose each instruction into atomic-level evaluation criteria, enabling rigorous validation and the elimination of inaccurate instruction-response pairs. Extensive experiments across a wide range of scenarios and settings demonstrate DecIF's superior performance on instruction-following tasks. Further analysis highlights its strong flexibility, scalability, and generalizability in automatically synthesizing high-quality instruction data.
Code (0)
등록된 구현이 없습니다.
Tasks
Instruction FollowingResponse GenerationSimilar Papers 제목 키워드 기반
deCIFer: Crystal Structure Prediction from Powder Diffraction Data using Autoregressive Language Models
Novel materials drive progress across applications from energy storage to electronics. Automated characterization of material structures with machine learning methods offers a promising strategy for accelerating this key…
RubricEval: A Rubric-Level Meta-Evaluation Benchmark for LLM Judges in Instruction Following
Rubric-based evaluation has become a prevailing paradigm for evaluating instruction following in large language models (LLMs). Despite its widespread use, the reliability of these rubric-level evaluations remains unclear…
Instruction FollowingLACMA: Language-Aligning Contrastive Learning with Meta-Actions for Embodied Instruction Following
End-to-end Transformers have demonstrated an impressive success rate for Embodied Instruction Following when the environment has been seen in training. However, they tend to struggle when deployed in an unseen environmen…
Contrastive LearningInstruction FollowingIF-RewardBench: Benchmarking Judge Models for Instruction-Following Evaluation
Instruction-following is a foundational capability of large language models (LLMs), with its improvement hinging on scalable and accurate feedback from judge models. However, the reliability of current judge models in in…
Evaluating Large Language Models at Evaluating Instruction Following
As research in large language models (LLMs) continues to accelerate, LLM-based evaluation has emerged as a scalable and cost-effective alternative to human evaluations for comparing the ever increasing list of models. Th…
Instruction Following