Shapley-based Data Valuation for LLM Alignment via Sequential Preference Optimization
cs.LG, cs.GT, stat.ML
Submitted: 2025-12-12
Updated: 2026-09-26
Terminology
Sources
- UltraFeedback: Boosting Language Models with Scaled AI Feedback
- Semivalue-based data valuation is arbitrary and gameable
- Learn Your Reference Model for Real Good Alignment
- Scaling Laws for Neural Language Models
- Bounding the Estimation Error of Sampling-based Shapley Value Approximation
- Qwen2.5 Technical Report
- Proximal Policy Optimization Algorithms
- On the Impact of the Utility in Semivalue-based Data Valuation
- Fair Document Valuation in LLM Summaries via Shapley Values
- Fine-Tuning Language Models from Human Preferences
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks