ORPO performs preference alignment during supervised fine-tuning via a monolithic odds ratio penalty, allowing 7B models to outperform larger state-of-the-art models on alignment benchmarks.
Title resolution pending
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
representative citing papers
ST-MoE introduces stability techniques for sparse expert models, allowing a 269B-parameter model to achieve state-of-the-art transfer learning results across reasoning, summarization, and QA tasks at the compute cost of a 32B dense model.
CalibAdv calibrates GRPO advantage signals for search agents by downscaling excessive negative advantages using intermediate-step correctness, improving performance and training stability across three models and seven benchmarks.
PaSaMaster, a recursive self-evolving agentic literature retrieval system, reports 16.5× higher F1 than Google Scholar and 37.8% higher F1 than GPT-5.2 on PaSaMaster-Bench across 38 disciplines with zero source hallucination at ~1% cost.
Proposes Modality-Aware Credit Assignment (MoCA) with blindfolded-reasoning proxy to reward perception fidelity separately from reasoning in VLMs.
citing papers explorer
-
ORPO: Monolithic Preference Optimization without Reference Model
ORPO performs preference alignment during supervised fine-tuning via a monolithic odds ratio penalty, allowing 7B models to outperform larger state-of-the-art models on alignment benchmarks.