Beyond Ordinal Preferences: Why Alignment Needs Cardinal Human Feedback

📅 2025-08-11
📈 Citations: 0
Influential: 0
📄 PDF

career value

172K/year
🤖 AI Summary
Current large language model alignment relies on ordinal human preferences (e.g., pairwise comparisons), yet theoretical analysis shows such preferences cannot uniquely identify the optimal policy—especially when trade-offs across orthogonal dimensions (e.g., factual accuracy vs. stylistic fluency) are required. Method: This work identifies this as a fundamental limitation and proposes cardinal human feedback, quantifying response quality differences via willingness-to-pay (WTP). We construct a high-fidelity cardinal dataset of 25K samples and integrate it into preference tuning via a novel cardinal loss function. Results: Experiments demonstrate consistent and significant gains over ordinal methods on benchmarks including Arena-Hard, enabling more precise identification of high-impact improvements and enhancing alignment efficiency and robustness. Core contribution: We establish the necessity of cardinal feedback for unambiguous model-level preference identification, thereby advancing beyond the conventional ordinal paradigm in preference learning.

Technology Category

Application Category

📝 Abstract
Alignment techniques for LLMs rely on optimizing preference-based objectives -- where these preferences are typically elicited as ordinal, binary choices between responses. Recent work has focused on improving label quality or mitigating particular biases, but we identify a more fundamental limitation: these methods collect the wrong kind of data. We prove an impossibility result: no algorithm relying solely on ordinal comparisons can systematically recover the most preferred model. Intuitively, ordinal data lacks the information needed to resolve tradeoffs -- e.g., fixing a factual error on one prompt versus improving style on another. We show that selecting the optimal model requires recovering preferences over emph{models} (rather than just responses), which can only be identified given cardinal feedback about response quality. To address this, we collect and publicly release a dataset of 25,000 cardinal judgments using willingness-to-pay elicitations, a well-established tool from experimental economics. Empirically, we find that incorporating cardinal feedback into preference fine-tuning allows models to prioritize high-impact improvements and outperform ordinal-only methods on downstream benchmarks, such as Arena-Hard.
Problem

Research questions and friction points this paper is trying to address.

Ordinal preferences limit optimal model recovery
Cardinal feedback needed for resolving tradeoffs
Current methods collect inadequate preference data
Innovation

Methods, ideas, or system contributions that make the work stand out.

Uses cardinal feedback for model preferences
Collects 25,000 willingness-to-pay judgments
Improves model performance with cardinal data
🔎 Similar Papers
No similar papers found.