Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving

• URL: http://arxiv.org/abs/2509.22480v1
• Date: Fri, 26 Sep 2025 15:27:50 GMT
• Title: Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Authors: Hang Li, Kaiqi Yang, Yucheng Chu, Hui Liu, Jiliang Tang,
• Abstract summary: We show that higher solution divergence is positively related to better problem-solving abilities across various models.<n>We propose solution divergence as a novel metric that can support both SFT and RL strategies.
• Score: 37.94354699202412
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models (LLMs) have been widely used for problem-solving tasks. Most recent work improves their performance through supervised fine-tuning (SFT) with labeled data or reinforcement learning (RL) from task feedback. In this paper, we study a new perspective: the divergence in solutions generated by LLMs for a single problem. We show that higher solution divergence is positively related to better problem-solving abilities across various models. Based on this finding, we propose solution divergence as a novel metric that can support both SFT and RL strategies. We test this idea on three representative problem domains and find that using solution divergence consistently improves success rates. These results suggest that solution divergence is a simple but effective tool for advancing LLM training and evaluation.

Related papers

• HINT: Helping Ineffective Rollouts Navigate Towards Effectiveness [49.72591739116668]
  Reinforcement Learning (RL) has become a key driver for enhancing the long chain-of-thought (CoT) reasoning capabilities of Large Language Models (LLMs)<n>However, prevalent methods like GRPO often fail when task difficulty exceeds the model's capacity, leading to reward sparsity and inefficient training.<n>We propose HINT: Helping Ineffective rollouts Navigate Towards effectiveness, an adaptive hinting framework.
  arXiv  Detail & Related papers  (2025-10-10T13:42:03Z)
• Staying in the Sweet Spot: Responsive Reasoning Evolution via Capability-Adaptive Hint Scaffolding [59.60915947702282]
  Reinforcement learning with verifiable rewards (RLVR) has achieved remarkable success in enhancing the reasoning capabilities of large language models (LLMs)<n>Existing RLVR methods often suffer from exploration inefficiency due to mismatches between the training data's difficulty and the model's capability.<n>We propose SEELE, a novel supervision-aided RLVR framework that dynamically adjusts problem difficulty to stay within the high-efficiency region.
  arXiv  Detail & Related papers  (2025-09-08T17:36:21Z)
• POMO+: Leveraging starting nodes in POMO for solving Capacitated Vehicle Routing Problem [0.0]
  In this work, we improved POMO, creating a method (textbfPOMO+) that leverages the initial nodes to find a solution in a more informed way.<n>We validated our models on the CVLIBRP dataset and noticed improvements in problem instances with up to 100 customers.
  arXiv  Detail & Related papers  (2025-08-11T21:55:16Z)
• Flow of Reasoning: Training LLMs for Divergent Reasoning with Minimal Examples [12.48027669682156]
  Flow of Reasoning (FoR) aims at improving diversity with minimal data.<n>FoR formulates multi-step LLM reasoning as a Markovian flow on a DAG-structured reasoning graph.<n>Experiments show that, with limited training examples, FoR enables the discovery of diverse, creative, high-quality solutions.
  arXiv  Detail & Related papers  (2024-06-09T07:06:58Z)
• Divide-or-Conquer? Which Part Should You Distill Your LLM? [38.62667131299918]
  We devise a similar strategy that breaks down reasoning tasks into a problem decomposition phase and a problem solving phase. We show that the strategy is able to outperform a single stage solution.
  arXiv  Detail & Related papers  (2024-02-22T22:28:46Z)
• DiLA: Enhancing LLM Tool Learning with Differential Logic Layer [11.810200077863172]
  We propose a novel differential logic layer-aided language modeling (DiLA) approach, where logical constraints are integrated into the forward and backward passes of a network layer. We evaluate the performance of DiLA on two classic reasoning problems and empirically demonstrate its consistent outperformance against existing prompt-based and solver-aided approaches.
  arXiv  Detail & Related papers  (2024-02-19T07:38:57Z)
• Thought Propagation: An Analogical Approach to Complex Reasoning with Large Language Models [62.96551299003463]
  We propose textbftextitThought Propagation (TP) to enhance the complex reasoning ability of Large Language Models. TP first prompts LLMs to propose and solve a set of analogous problems that are related to the input one. TP reuses the results of analogous problems to directly yield a new solution or derive a knowledge-intensive plan for execution to amend the initial solution obtained from scratch.
  arXiv  Detail & Related papers  (2023-10-06T01:40:09Z)
• A Mutual Information Maximization Approach for the Spurious Solution Problem in Weakly Supervised Question Answering [60.768146126094955]
  Weakly supervised question answering usually has only the final answers as supervision signals. There may exist many spurious solutions that coincidentally derive the correct answer, but training on such solutions can hurt model performance. We propose to explicitly exploit such semantic correlations by maximizing the mutual information between question-answer pairs and predicted solutions.
  arXiv  Detail & Related papers  (2021-06-14T05:47:41Z)
• Learning by Fixing: Solving Math Word Problems with Weak Supervision [70.62896781438694]
  Previous neural solvers of math word problems (MWPs) are learned with full supervision and fail to generate diverse solutions. We introduce a textitweakly-supervised paradigm for learning MWPs. Our method only requires the annotations of the final answers and can generate various solutions for a single problem.
  arXiv  Detail & Related papers  (2020-12-19T03:10:21Z)

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.