DuPO: Enabling Reliable LLM Self-Verification via Dual Preference Optimization

• URL: http://arxiv.org/abs/2508.14460v1
• Date: Wed, 20 Aug 2025 06:31:18 GMT
• Title: DuPO: Enabling Reliable LLM Self-Verification via Dual Preference Optimization
• Abstract summary: We present DuPO, a dual learning-based preference optimization framework that generates annotation-free feedback.<n>Specifically, DuPO decomposes a primal task's input into known and unknown components, then constructs its dual task to reconstruct the unknown part.<n>It enhances the average translation quality by 2.13 COMET over 756 directions, boosts the mathematical reasoning accuracy by an average of 6.4 points on three challenge benchmarks, and enhances performance by 9.3 points as an inference-time reranker.
• Score: 47.32314866162273
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: We present DuPO, a dual learning-based preference optimization framework that generates annotation-free feedback via a generalized duality. DuPO addresses two key limitations: Reinforcement Learning with Verifiable Rewards (RLVR)'s reliance on costly labels and applicability restricted to verifiable tasks, and traditional dual learning's restriction to strictly dual task pairs (e.g., translation and back-translation). Specifically, DuPO decomposes a primal task's input into known and unknown components, then constructs its dual task to reconstruct the unknown part using the primal output and known information (e.g., reversing math solutions to recover hidden variables), broadening applicability to non-invertible tasks. The quality of this reconstruction serves as a self-supervised reward to optimize the primal task, synergizing with LLMs' ability to instantiate both tasks via a single model. Empirically, DuPO achieves substantial gains across diverse tasks: it enhances the average translation quality by 2.13 COMET over 756 directions, boosts the mathematical reasoning accuracy by an average of 6.4 points on three challenge benchmarks, and enhances performance by 9.3 points as an inference-time reranker (trading computation for accuracy). These results position DuPO as a scalable, general, and annotation-free paradigm for LLM optimization.

Related papers

• ExTra: Exploratory Trajectory Optimization for Language Model Reinforcement Learning [75.26064451955199]
  ExTra is a framework that extracts exploration signals from the model's own rollouts.<n>It improves Qwen3-1.7B over GRPO by about +5 points on pass@1 and +7 points on pass@16.
  arXiv  Detail & Related papers  (2026-06-23T15:51:39Z)
• dTRPO: Trajectory Reduction in Policy Optimization of Diffusion Large Language Models [71.26391195427878]
  Diffusion Large Language Models (dLLMs) introduce a new paradigm for language generation.<n>In this work, we aim to improve the policy optimization for dLLMs by reducing the cost of the trajectory probability calculation.<n>We evaluate dTRPO on 7B dLLMs across instruction-following and reasoning benchmarks.
  arXiv  Detail & Related papers  (2026-03-19T11:55:52Z)
• Generative Reasoning Re-ranker [24.386586034456673]
  Generative Reasoning Reranker (GR2) is an end-to-end framework with a three-stage training pipeline tailored for reranking.<n> GR2 generates high-quality reasoning traces through carefully designed prompting and rejection sampling.<n>Experiments on two real-world datasets demonstrate GR2's effectiveness.
  arXiv  Detail & Related papers  (2026-02-08T02:12:24Z)
• MortgageLLM: Domain-Adaptive Pretraining with Residual Instruction Transfer, Alignment Tuning, and Task-Specific Routing [0.7367330074083941]
  We present MortgageLLM, a novel domain-specific large language model.<n>It is developed using a dual-track specialization framework from a single base model.<n>We contribute: (1) application of this residual technique to the highly specialized mortgage finance domain; (2) a dual-expert architecture combining a conversational Q&A model and a structured task model for classification and summarization; and (3) an intelligent task routing mechanism using few-shot classification performed by one of the expert models itself.
  arXiv  Detail & Related papers  (2025-11-26T06:37:57Z)
• Structured Uncertainty guided Clarification for LLM Agents [126.26213027785813]
  LLM agents extend large language models with tool-calling capabilities, but ambiguous user instructions often lead to incorrect invocations and task failures.<n>We introduce a principled formulation of structured uncertainty over tool-call parameters, modeling joint tool-argument clarification as a POMDP with Expected Value of Perfect Information (EVPI) objective for optimal question selection and aspect-based cost modeling to prevent redundancy.<n>Our SAGE-Agent leverages this structured uncertainty to achieve superior efficiency: increasing coverage on ambiguous tasks by 7-39% while reducing clarification questions by 1.5-2.7$times$ compared to strong prompting and uncertainty-based baselines.
  arXiv  Detail & Related papers  (2025-11-11T21:50:44Z)
• CoT-Saliency: Unified Chain-of-Thought Reasoning for Heterogeneous Saliency Tasks [96.64597365827046]
  We present the first unified framework that jointly handles three operationally heterogeneous saliency tasks.<n>We introduce a Chain-of-Thought (CoT) reasoning process in a Vision-Language Model (VLM) to bridge task heterogeneity.<n>We show our model matches or outperforms specialized SOTA methods and strong closed-source VLMs across all tasks.
  arXiv  Detail & Related papers  (2025-11-01T04:37:01Z)
• Omni-DPO: A Dual-Perspective Paradigm for Dynamic Preference Learning of LLMs [28.41899655478021]
  We propose Omni-DPO, a dual-perspective optimization framework that accounts for the inherent quality of each preference pair and the model's evolving performance on those pairs.<n> Experimental results on various models and benchmarks demonstrate the superiority and generalization capabilities of Omni-DPO.
  arXiv  Detail & Related papers  (2025-06-11T17:58:05Z)
• MDPO: Multi-Granularity Direct Preference Optimization for Mathematical Reasoning [0.0]
  We propose the Multi-Granularity Direct Preference Optimization (MDPO) method, optimizing the mathematical reasoning of Large Language Models (LLMs)<n>We conduct experiments on the open-source models Qwen2 and Llama3, achieving improvements of 1.7% and 1.2% on the GSM8K dataset, and 2.3% and 1.2% on the MATH dataset.<n>We also provide a pipeline for constructing MDPO training data that is simple and does not require manual annotation costs.
  arXiv  Detail & Related papers  (2025-05-30T08:42:14Z)
• Teaching Large Language Models to Maintain Contextual Faithfulness via Synthetic Tasks and Reinforcement Learning [80.27561080938747]
  CANOE is a framework to reduce hallucinations of faithfulness of large language models across different downstream tasks without human annotations.<n>Dual-GRPO is a rule-based reinforcement learning method that includes three tailored rule-based rewards derived from synthesized short-form QA data.<n> Experimental results show that CANOE greatly improves the faithfulness of LLMs across 11 different tasks, even outperforming the most advanced LLMs.
  arXiv  Detail & Related papers  (2025-05-22T10:10:07Z)
• Dspy-based Neural-Symbolic Pipeline to Enhance Spatial Reasoning in LLMs [29.735465300269993]
  Large Language Models (LLMs) have demonstrated remarkable capabilities across various tasks, yet they often struggle with spatial reasoning.<n>This paper presents a novel neural-symbolic framework that enhances LLMs' spatial reasoning abilities through iterative feedback between LLMs and Answer Set Programming (ASP)<n>We evaluate our approach on two benchmark datasets: StepGame and SparQA.
  arXiv  Detail & Related papers  (2024-11-27T18:04:05Z)
• TIS-DPO: Token-level Importance Sampling for Direct Preference Optimization With Estimated Weights [73.9088920210495]
  We propose a token-level importance sampling DPO objective named TIS-DPO that assigns importance weights to each token based on its reward.<n>TIS-DPO significantly outperforms various baseline methods on harmlessness and helpfulness alignment and summarization tasks.
  arXiv  Detail & Related papers  (2024-10-06T04:03:00Z)
• Improving Open Information Extraction with Large Language Models: A Study on Demonstration Uncertainty [52.72790059506241]
  Open Information Extraction (OIE) task aims at extracting structured facts from unstructured text. Despite the potential of large language models (LLMs) like ChatGPT as a general task solver, they lag behind state-of-the-art (supervised) methods in OIE tasks.
  arXiv  Detail & Related papers  (2023-09-07T01:35:24Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.