GroupDPO: Memory-Efficient Group-Wise Direct Preference Optimization

arXiv:2604.15602 · cs.CL · Submitted 2026-04-17 · Read on arXiv

cs.CL

Submitted: 2026-04-17

Updated: 2026-09-01

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Benchmark — 13× in this paper · explained in 2 episodes
• Alignment — 9× in this paper · explained in 1 episode
• Direct Preference Optimization — 8× in this paper
• Pairwise Comparisons — 5× in this paper · explained in 2 episodes
• Adam — 4× in this paper · explained in 2 episodes
• Fine-Tuning — 4× in this paper · explained in 5 episodes
• Reinforcement Learning — 4× in this paper · explained in 4 episodes
• Foundation Model — 3× in this paper · explained in 3 episodes
• Foundation Models — 3× in this paper · explained in 7 episodes
• Accuracy — 2× in this paper · explained in 1 episode
• Calibration — 2× in this paper · explained in 7 episodes
• Decoupling — 2× in this paper · explained in 2 episodes
• Large Language Models (LLMs) — 2× in this paper · explained in 6 episodes
• Preference Ranking Optimization — 2× in this paper
• Regularization — 2× in this paper · explained in 3 episodes
• Aggregation — 1× in this paper · explained in 1 episode
• Computational overhead — 1× in this paper · explained in 2 episodes
• Direct Preference Optimization (DPO) — 1× in this paper · explained in 9 episodes
• Elias — 1× in this paper · explained in 2 episodes
• Failure Modes — 1× in this paper · explained in 3 episodes
• Large Language Model (LLM) — 1× in this paper · explained in 3 episodes
• Proximal Policy Optimization (PPO) — 1× in this paper · explained in 5 episodes
• Reward Function — 1× in this paper · explained in 2 episodes
• Robustness — 1× in this paper · explained in 5 episodes
• Tokenization — 1× in this paper · explained in 2 episodes
• Transformer — 1× in this paper · explained in 2 episodes

Sources

• The Llama 3 Herd of Models
• Multi-Preference Optimization: Generalizing DPO via Set-Level Contrasts
• Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs
• LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code
• gpt-oss-120b & gpt-oss-20b Model Card
• Tulu 3: Pushing Frontiers in Open Language Model Post-Training
• Taming Overconfidence in LLMs: Reward Calibration in RLHF
• Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models
• UltraFeedback: Boosting Language Models with Scaled AI Feedback
• RankGR: Rank-Enhanced Generative Retrieval with Listwise Direct Preference Optimization in Recommendation
• ZebraLogic: On the Scaling Limits of LLMs for Logical Reasoning
• Skywork-Reward-V2: Scaling Preference Data Curation via Human-AI Synergy
• Generalizing Verifiable Instruction Following
• Length Desensitization in Direct Preference Optimization
• Proximal Policy Optimization Algorithms
• Decoupled Weight Decay Regularization
• Olmo 3
• Gemma 3 Technical Report
• Smaug: Fixing Failure Modes of Preference Optimisation with DPO-Positive
• Qwen3 Technical Report

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering