WildFeedback: Aligning LLMs With In-situ User Interactions And Feedback
This addresses the scalability, subjectivity, and bias challenges in aligning LLMs for real-world users, representing a significant but incremental step over traditional methods.
The paper tackles the problem of aligning large language models (LLMs) with human preferences by introducing WildFeedback, a framework that automatically creates preference datasets from in-situ user feedback during conversations, resulting in LLMs that show significantly improved alignment with user preferences as demonstrated by benchmarks and a checklist-guided evaluation.
As large language models (LLMs) continue to advance, aligning these models with human preferences has emerged as a critical challenge. Traditional alignment methods, relying on human or LLM annotated datasets, are limited by their resource-intensive nature, inherent subjectivity, misalignment with real-world user preferences, and the risk of feedback loops that amplify model biases. To overcome these limitations, we introduce WildFeedback, a novel framework that leverages in-situ user feedback during conversations with LLMs to create preference datasets automatically. Given a corpus of multi-turn user-LLM conversation, WildFeedback identifies and classifies user feedback to LLM responses between conversation turns. The user feedback is then used to create examples of preferred and dispreferred responses according to users' preference. Our experiments demonstrate that LLMs fine-tuned on WildFeedback dataset exhibit significantly improved alignment with user preferences, as evidenced by both traditional benchmarks and our proposed checklist-guided evaluation. By incorporating in-situ feedback from actual users, WildFeedback addresses the scalability, subjectivity, and bias challenges that plague existing approaches, marking a significant step toward developing LLMs that are more responsive to the diverse and evolving needs of their users.