mlsanityData to 14 Oct 2026

Do GFlowNets Transfer? Case Study on the Game of 24/42

arXiv 2503.01819 ↗First posted 3 Mar 2025cs.AI cs.CLcited by 0

On the map: LLM reasoning → Mathematical reasoning

Abstract (arXiv)

Generating diverse solutions is key to human-like reasoning, yet autoregressive language models focus on single accurate responses, limiting creativity. GFlowNets optimize solution generation as a flow network, promising greater diversity. Our case study shows their limited zero-shot transferability by fine-tuning small and medium-sized large language models on the Game of 24 and testing them on the Game of 42 datasets. Results revealed that GFlowNets struggle to maintain solution diversity and accuracy, highlighting key limitations in their cross-task generalization and the need for future research in improved transfer learning capabilities.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Mathematical reasoning, one of 17 directions in LLM reasoning. Builds on 21 of its 26 references that are in the map. First posted on arXiv 3 Mar 2025.

Builds on

21 of 26 references.

Closest ancestors · 1

Same direction, newest first

  1. 2024MIND: Math Informed syNthetic Dialogues for Pretraining LLMs15 Oct

Other references · 20

  1. 2022Training language models to follow instructions with human feedbackcited by 15,971
  2. 2017Proximal Policy Optimization Algorithmscited by 15,215
  3. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  4. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  5. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,780
  6. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 3,211
  7. 2019The Curious Case of Neural Text Degenerationcited by 2,916
  8. 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 2,724
  9. 2018Hierarchical Neural Story Generationcited by 1,193
  10. 2023Graph of Thoughts: Solving Elaborate Problems with Large Language Modelscited by 886
  11. 2022Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP Taskscited by 868
  12. 2023MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Modelscited by 735
8 more
  1. 2021Cross-Task Generalization via Natural Language Crowdsourcing Instructionscited by 699
  2. 2023MAmmoTH: Building Math Generalist Models through Hybrid Instruction Tuningcited by 487
  3. 2021Flow Network based Generative Models for Non-Iterative Diverse Candidate Generationcited by 392
  4. 2021GFlowNet Foundationscited by 290
  5. 2022Trajectory balance: Improved credit assignment in GFlowNetscited by 251
  6. 2024Turning Up the Heat: Min-p Sampling for Creative and Coherent LLM Outputscited by 150
  7. 2022Learning GFlowNets from partial episodes for improved convergence and stabilitycited by 132
  8. 2023Amortizing intractable inference in large language modelscited by 97

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024GFlowNet Fine-tuning for Diverse Correct Solutions in Mathematical Reasoning Tasks
  2. 2026GFlowRL: Scaling Distribution-Matching RL to Large Language Models
  3. 2025Accurate and Diverse LLM Mathematical Reasoning via Automated PRM-Guided GFlowNets
  4. 2024Flow of Reasoning: Training LLMs for Divergent Reasoning with Minimal Examples
  5. 2025GFlowVLM: Enhancing Multi-step Reasoning in Vision-Language Models with Generative Flow Networks
  6. 2024Proof Flow: Preliminary Study on Generative Flow Network Language Model Tuning for Formal Reasoning
  7. 2023GFlowNets with Human Feedback
  8. 2025Measuring General Intelligence with Generated Games