mlsanityData to 14 Oct 2026

EquivPruner: Boosting Efficiency and Quality in LLM-Based Search via Action Pruning

arXiv 2505.16312 ↗First posted 22 May 2025cs.AI cs.CLcited by 1

On the map: LLM reasoning → Test-time compute and efficient thinking

Abstract (arXiv)

Large Language Models (LLMs) excel at complex reasoning through search algorithms, yet current strategies often suffer from massive token consumption due to redundant exploration of semantically equivalent steps. Existing semantic similarity methods struggle to accurately identify such equivalence in domain-specific contexts like mathematical reasoning. To address this, we propose EquivPruner, a simple yet effective approach that identifies and prunes semantically equivalent actions during LLM reasoning search. We also introduce MathEquiv, the first dataset we created for mathematical statement equivalence, which enables the training of a lightweight equivalence detector. Extensive experiments across various models and tasks demonstrate that EquivPruner significantly reduces token consumption, improving searching efficiency and often bolstering reasoning accuracy. For instance, when applied to Qwen2.5-Math-7B-Instruct on GSM8K, EquivPruner reduced token consumption by 48.1% while also improving accuracy. Our code is available at https://github.com/Lolo1222/EquivPruner.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning. Builds on 15 of its 31 references that are in the map. First posted on arXiv 22 May 2025.

Builds on

15 of 31 references.

Closest ancestors · 4

Same direction, newest first

  1. 2024Enhancing LLM Reasoning with Reward-guided Tree Search18 Nov
  2. 2024Learning How Hard to Think: Input-Adaptive Allocation of LM Computation7 Oct
  3. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters6 Aug
  4. 2023Large Language Model Guided Tree-of-Thought15 May

Other references · 11

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
  3. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,371
  4. 2023Let's Verify Step by Stepcited by 3,789
  5. 2020Longformer: The Long-Document Transformercited by 3,413
  6. 2018CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledgecited by 1,822
  7. 2024ReST-MCTS*: LLM Self-Training via Process Reward Guided Tree Searchcited by 344
  8. 2024Monte Carlo Tree Search Boosts Reasoning via Iterative Preference Learningcited by 241
  9. 2023Planning with Large Language Models for Code Generationcited by 182
  10. 2024AlphaMath Almost Zero: Process Supervision without Processcited by 162
  11. 2021MathBERT: A Pre-Trained Model for Mathematical Formula Understandingcited by 58

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026PAC-CF: Calibrating Irreversible Frontier Pruning in LLM-Guided Searchcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Navigating the Labyrinth: Evaluating LLMs' Ability to Reason About Search Problems
  2. 2025RaaS: Reasoning-Aware Attention Sparsity for Efficient LLM Reasoning
  3. 2026LinTree: Improving LLM Reasoning with Explicitly Structured Search Histories
  4. 2025AdaR: A Framework for Equipping LLMs with Adaptive Reasoning
  5. 2024MindStar: Enhancing Math Reasoning in Pre-trained LLMs at Inference Time
  6. 2026How Well Do LLMs Perform on the Simplest Long-Chain Reasoning Tasks: An Empirical Study on the Equivalence Class Problem
  7. 2025EffiReason-Bench: A Unified Benchmark for Evaluating and Advancing Efficient Reasoning in Large Language Models
  8. 2024Logic Contrastive Reasoning with Lightweight Large Language Model for Math Word Problems