mlsanityData to 14 Oct 2026

GenSelect: A Generative Approach to Best-of-N

arXiv 2507.17797 ↗First posted 23 Jul 2025cs.LG cs.CLcited by 14

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

also in 1 more
Abstract (arXiv)

Generative reward models with parallel sampling have enabled effective test-time scaling for reasoning tasks. Current approaches employ pointwise scoring of individual solutions or pairwise comparisons. However, pointwise methods underutilize LLMs' comparative abilities, while pairwise methods scale inefficiently with larger sampling budgets. We introduce GenSelect, where the LLM uses long reasoning to select the best solution among N candidates. This leverages LLMs' comparative strengths while scaling efficiently across parallel sampling budgets. For math reasoning, we demonstrate that reasoning models, such as QwQ and DeepSeek-R1-0528, excel at GenSelect, outperforming existing scoring approaches with simple prompting.

Key facts. Cited by 14 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 12 of its 18 references that are in the map. First posted on arXiv 23 Jul 2025.

Builds on

12 of 18 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025Sample, Scrutinize and Scale: Effective Inference-Time Search by Scaling Verification3 Feb
  2. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters6 Aug

Other references · 10

  1. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  2. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  3. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,708
  4. 2024Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvementcited by 1,090
  5. 2023LLM-Blender: Ensembling Large Language Models with Pairwise Ranking and Generative Fusioncited by 564
  6. 2024Generative Verifiers: Reward Modeling as Next-Token Predictioncited by 480
  7. 2025The Lessons of Developing Process Reward Models in Mathematical Reasoningcited by 415
  8. 2024OpenMathInstruct-2: Accelerating AI for Math with Massive Open-Source Instruction Datacited by 200
  9. 2024Aligning with Human Judgement: The Role of Pairwise Preference in Large Language Model Evaluatorscited by 149
  10. 2024AceMath: Advancing Frontier Math Reasoning with Post-Training and Reward Modelingcited by 77

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 5; 2026: 9.

14 papers.

Sort:

By area:

  1. 2025Nemotron 3 Nano: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoningcited by 104
  2. 2026Nemotron 3 Super: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoningcited by 38
  3. 2025Don't Pass@k: A Bayesian Framework for Large Language Model Evaluationcited by 15
  4. 2026V1: Unifying Generation and Self-Verification for Parallel Reasonerscited by 12
  5. 2025Scaling Test-Time Compute to Achieve IOI Gold Medal with Open-Weight Modelscited by 10
  6. 2025Scaling Generative Verifiers For Natural Language Mathematical Proof Verification And Selectioncited by 9
  7. 2026VLA-ATTC: Adaptive Test-Time Compute for VLA Models with Relative Action Critic Modelcited by 9
  8. 2025Population-Evolve: a Parallel Sampling and Evolutionary Method for LLM Math Reasoningcited by 6
  9. 2026Parallel Test-Time Scaling with Multi-Sequence Verifierscited by 2
  10. 2026Claim-Level Reliability Assessment for Efficient Test-Time Reasoningcited by 1
  11. 2026Mid-Harness: Scaling Actions Between Model and Harness for Terminal Agentscited by 1
  12. 2026Precision in Practice: Knowledge Guided Code Summarizing Grounded in Industrial Expectationscited by 0
2 more
  1. 2026Selection, Recombination, or a Fresh Solve? A Candidate-Free Control for Single-Pass Test-Time Aggregationcited by 0
  2. 2026Mo' Models, Mo' Problems: How to best select model pools when designing Multi-Agent Systemscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Learning Generative Selection for Best-of-N
  2. 2025Adaptive Test-Time Reasoning via Reward-Guided Dual-Phase Search
  3. 2026Reason Only When Needed: Efficient Generative Reward Modeling via Model-Internal Uncertainty
  4. 2025Revisiting Test-Time Scaling: A Survey and a Diversity-Aware Method for Efficient Reasoning
  5. 2025When To Solve, When To Verify: Compute-Optimal Problem Solving and Generative Verification for LLM Reasoning
  6. 2025ReasonGRM: Enhancing Generative Reward Models through Large Reasoning Models
  7. 2026Beyond Repeated Sampling: Learning Search Policies for LLM Reasoning
  8. 2025Solve-Detect-Verify: Inference-Time Scaling with Flexible Generative Verifier