arXiv (All)
open access
Masking Frequent Tokens Sharpens Direct Preference Optimization
Direct Preference Optimization (DPO) aligns language models by optimizing over sequence-level sums of token-wise implicit reward differences. However, we identify a pervasive pathology in this formulation: a disproportionately small subset of high-frequency token types dominates cumulative sequence
Conceptio Open Archive
Analyze & explore related research in Conceptio
Connect this record directly into your research workflow or AI agent. Explore citations across open repositories, query full text, and export structured proof bundles.
This document is indexed with metadata only: full text is not available in the archive for this record. Open the official source ↗
Related documents
- GAW-PO: Preference Optimization with Gradient-Aligned Token Weights#1156783
- Preference Instability in Reward Models: Detection and Mitigation via Sparse Autoencoders#1143764
- Counting and Min-Cost Encoding for Tokenization in Large Language Models#1156062
- VA-DPO: Valence-Arousal Direct Preference Optimization for Controllable Emotion Generation in Language Models#658952
- VA-DPO: Valence-Arousal Direct Preference Optimization for Controllable Emotion Generation in Language Models#656816
Record · ID 1116776
Retrieved via Conceptio: every document is proof-bundled with source, license, and retrieval metadata.
Retrieved via Conceptio: every document is proof-bundled with source, license, and retrieval metadata.
Bring this data into your agent: Connect via MCP or API · 20 free credits to start · no card required
Read the Docs → · View CLI source on GitHub · Conceptio Open Knowledge Archive
Read the Docs → · View CLI source on GitHub · Conceptio Open Knowledge Archive