mlsanityData to 14 Oct 2026

Entropy-Gated Branching for Efficient Test-Time Reasoning

arXiv 2503.21961 ↗First posted 27 Mar 2025cs.CL cs.AIcited by 22

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

Abstract (arXiv)

Test-time compute methods can significantly improve the reasoning capabilities and problem-solving accuracy of large language models (LLMs). However, these approaches require substantially more computational resources, with most compute wasted on exploring low-diversity branches where the model already exhibits high confidence. We observe that a small subset of uncertain reasoning steps has a disproportionately large impact on final prediction accuracy, and branching at these critical junctures tends to yield more diverse and higher-quality candidate reasoning steps. We propose Entropy-Gated Branching (EGB), which branches only at high-uncertainty steps and prunes expansions with a lightweight verifier. On mathematical and financial reasoning benchmarks, EGB improves accuracy by 22.6% over standard inference while operating 31%-75% faster across math benchmarks than test-time beam search with higher performance. Our results show that dynamic resource allocation during inference can substantially improve both efficiency and effectiveness, offering a more scalable pathway to enhanced LLM reasoning capabilities.

Key facts. Cited by 22 papers 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 30 of its 56 references that are in the map. First posted on arXiv 27 Mar 2025.

Builds on

30 of 56 references.

Closest ancestors · 5

Same direction, newest first

  1. 2025Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scaling10 Feb
  2. 2025s1: Simple test-time scaling31 Jan
  3. 2025A Survey on LLM Test-Time Compute via Search: Tasks, LLM Profiling, Search Algorithms, and Relevant Frameworks17 Jan
  4. 2024Forest-of-Thought: Scaling Test-Time Compute for Enhancing LLM Reasoning12 Dec
  5. 2024Improving LLM Reasoning through Scaling Inference Computation with Collaborative Verification5 Oct

Other references · 25

  1. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  2. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  3. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,708
  4. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,371
  5. 2023Let's Verify Step by Stepcited by 3,789
  6. 2019The Curious Case of Neural Text Degenerationcited by 2,916
  7. 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,810
  8. 2022Large Language Models Are Human-Level Prompt Engineerscited by 1,045
  9. 2022Solving math word problems with process- and outcome-based feedbackcited by 901
  10. 2023SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Modelscited by 810
  11. 2023Language Agent Tree Search Unifies Reasoning Acting and Planning in Language Modelscited by 574
  12. 2025SWE-RL: Advancing LLM Reasoning via Reinforcement Learning on Open Software Evolutioncited by 263
13 more
  1. 2025Logic-RL: Unleashing LLM Reasoning with Rule-Based Reinforcement Learningcited by 262
  2. 2023Planning with Large Language Models for Code Generationcited by 182
  3. 2024Fact-Checking the Output of Large Language Models via Token-Level Uncertainty Quantificationcited by 162
  4. 2024Tree Search for Language Model Agentscited by 151
  5. 2023Training Language Models with Language Feedback at Scalecited by 128
  6. 2024Feedback Loops With Language Models Drive In-Context Reward Hackingcited by 77
  7. 2024Can LLMs Learn from Previous Mistakes? Investigating LLMs' Errors to Boost for Reasoningcited by 69
  8. 2023Are ChatGPT and GPT-4 General-Purpose Solvers for Financial Text Analytics? A Study on Several Typical Taskscited by 59
  9. 2018Importance of Search and Evaluation Strategies in Neural Dialogue Modelingcited by 44
  10. 2024Are Large Language Models Good Prompt Optimizers?cited by 37
  11. 2024Black-box Uncertainty Quantification Method for LLM-as-a-Judgecited by 18
  12. 2024What's Wrong? Refining Meeting Summaries with LLM Feedbackcited by 8
  13. 2025Reducing Large Language Model Safety Risks in Women's Health using Semantic Entropycited by 8

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 6; 2026: 16.

22 papers.

Sort:

By area:

  1. 2025Breaking the Exploration Bottleneck: Rubric-Scaffolded Reinforcement Learning for General LLM Reasoningcited by 63
  2. 2025AR-GRPO: Training Autoregressive Image Generation Models via Reinforcement Learningcited by 22
  3. 2025The Sequential Edge: Inverse-Entropy Voting Beats Parallel Self-Consistency at Matched Computecited by 11
  4. 2026Thinking by Subtraction: Confidence-Driven Contrastive Decoding for LLM Reasoningcited by 6
  5. 2025Less is More: Improving LLM Reasoning with Minimal Test-Time Interventioncited by 5
  6. 2025Chain-of-Code Collapse: Reasoning Failures in LLMs via Adversarial Prompting in Code Generationcited by 4
  7. 2026Rethinking Code Complexity Through the Lens of Large Language Modelscited by 4
  8. 2026Code Is More Than Text: Uncertainty Estimation for Code Generationcited by 4
  9. 2025Chain-in-Tree: Back to Sequential Reasoning in LLM Tree Searchcited by 3
  10. 2026Long-Horizon Plan Execution in Large Tool Spaces through Entropy-Guided Branchingcited by 3
  11. 2026Dissecting Failure Dynamics in Large Language Model Reasoningcited by 2
  12. 2026Directional Reasoning Trajectory Change (DRTC): Identifying Critical Trace Segments in Reasoning Modelscited by 1
10 more
  1. 2026From Instance Selection to Fixed-Pool Data Recipe Search for Supervised Fine-Tuningcited by 1
  2. 2026Taming the Thinker: Conditional Entropy Shaping for Adaptive LLM Reasoningcited by 1
  3. 2026ARBITER: Reasoning Trajectory Basins and Majority Vote Failures in Test-Time Samplingcited by 1
  4. 2026What Am I Missing? Question-Answering as Hidden State Probingcited by 1
  5. 2026Embedding Perturbation may Better Reflect Intermediate-Step Uncertainty in LLM Reasoningcited by 0
  6. 2026Entropy-informed Decoding: Adaptive Information-Driven Branchingcited by 0
  7. 2026Read the Trace, Steer the Path: Trajectory-Aware Reinforcement Learning for Diffusion Language Modelscited by 0
  8. 2026Entropy-Gated Latent Recursioncited by 0
  9. 2026On the Diversity of Analogy Making in Large Language Modelscited by 0
  10. 2026Hesitation Has a Geometry: Entropy-Trained Hyperbolic Probes for Sparse Activation Steeringcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Think Twice Before You Write -- an Entropy-based Decoding Strategy to Enhance LLM Reasoning
  2. 2025Think Just Enough: Sequence-Level Entropy as a Confidence Signal for LLM Reasoning
  3. 2025Reasoning on a Budget: A Survey of Adaptive and Controllable Test-Time Compute in LLMs
  4. 2025Revisiting Test-Time Scaling: A Survey and a Diversity-Aware Method for Efficient Reasoning
  5. 2025Wider or Deeper? Scaling LLM Inference-Time Compute with Adaptive Branching Tree Search
  6. 2025DiffAdapt: Difficulty-Adaptive Reasoning for Token-Efficient LLM Inference
  7. 2026Shorthand for Thought: Compressing LLM Reasoning via Entropy-Guided Supertokens
  8. 2026ThinkBooster: A Unified Framework for Seamless Test-Time Scaling of LLM Reasoning