Skip to content
Papers.

Direct Preference Optimization: Your Language Model is Secretly a Reward Model research paper by Stanford University, 2023

Stanford University · May 29, 2023 · Alignment and safety · 10,608 citations · 71 upvotes

Read on arXiv

What it shows

DPO aligns a model to human preferences with a simple classification loss, no reward model or RL loop.

Summarised by hand from the abstract.

More from Stanford University

All 2
PaperCitations
FlashAttention: Fast and Memory-Efficient Exact Attention with IO-AwarenessFlashAttention computes exact attention with far fewer GPU memory reads and writes, making long sequences faster.Inference and efficiency · May 20225,331

Other alignment and safety papers

Topic
PaperCitations
Training language models to follow instructions with human feedbackInstructGPT: fine-tuning on human feedback made a 1.3B model preferred over the 175B GPT-3.OpenAI · Mar 202224.2k
Constitutional AI: Harmlessness from AI FeedbackConstitutional AI trains a harmless assistant from AI feedback guided by a short list of written principles, not human harm labels.Anthropic · Dec 20223,718
Sleeper Agents: Training Deceptive LLMs that Persist Through Safety TrainingModels trained with a hidden backdoor kept their deceptive behaviour through standard safety training.Anthropic · Jan 2024576
Alignment faking in large language modelsClaude 3 Opus sometimes went along with a training goal it disagreed with, to avoid being changed, without being told to.Anthropic · Dec 2024342
Why Language Models HallucinateModels hallucinate because training and benchmarks reward confident guessing over saying they do not know.OpenAI · Sep 2025329
Constitutional Classifiers: Defending against Universal Jailbreaks across Thousands of Hours of Red TeamingClassifiers trained from a written constitution held off universal jailbreaks through more than 3,000 hours of red teaming.Anthropic · Jan 2025199
About this paper
Authors
Rafael Rafailov, Archit Sharma, Eric Mitchell and 3 more
arXiv
2305.18290 · PDF
Venue
Neural Information Processing Systems
Citations
10,608, 2,203 influential · Semantic Scholar
Upvotes
71 · Hugging Face
Lab
Stanford University

Changes

What changed
New paperAdded to the listSep 24, 2026

Sources: each lab's own papers and arXiv, with citation and upvote counts from Semantic Scholar and Hugging Face. One-line summaries are for orientation, not a substitute for the paper. Logos via logo.dev; trademarks belong to their owners.

New papers by email

Monday afternoons, only in weeks with new papers from the labs.

Double opt-in. Unsubscribe any time.