mlsanityData to 14 Oct 2026

PREMISE: Scalable and Strategic Prompt Optimization for Efficient Mathematical Reasoning in Large Models

arXiv 2506.10716 ↗First posted 12 Jun 2025cs.CL cs.AI cs.LGcited by 5

On the map: LLM reasoning → Mathematical reasoning

Abstract (arXiv)

Large reasoning models (LRMs) such as Claude 3.7 Sonnet and OpenAI o1 achieve strong performance on mathematical benchmarks using lengthy chain-of-thought (CoT) reasoning, but the resulting traces are often unnecessarily verbose. This inflates token usage and cost, limiting deployment in latency-sensitive or API-constrained settings. We introduce PREMISE (PRompt-based Efficient Mathematical Inference with Strategic Evaluation), a prompt-only framework that reduces reasoning overhead without modifying model weights. PREMISE combines trace-level diagnostics with gradient-inspired prompt optimization to minimize redundant computation while preserving answer accuracy. The approach jointly optimizes brevity and correctness through a multi-objective textual search that balances token length and answer validity. Unlike prior work, PREMISE runs in a single-pass black-box interface, so it can be applied directly to commercial LLMs. On GSM8K, SVAMP, and Math500 we match or exceed baseline accuracy (96%rightarrow96% with Claude, 91%rightarrow92% with Gemini) while reducing reasoning tokens by up to 87.5% and cutting dollar cost by 69--82%. These results show that prompt-level optimization is a practical and scalable path to efficient LRM inference without compromising reasoning quality.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Placed in Mathematical reasoning, one of 17 directions in LLM reasoning. Builds on 29 of its 36 references that are in the map. First posted on arXiv 12 Jun 2025.

Builds on

29 of 36 references.

References · 29

  1. 2024The Llama 3 Herd of Modelscited by 14,715
  2. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  3. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  4. 2021Evaluating Large Language Models Trained on Codecited by 8,320
  5. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  6. 2021Are NLP Models really able to Solve Simple Math Word Problems?cited by 878
  7. 2024Training Large Language Models to Reason in a Continuous Latent Spacecited by 797
  8. 2024Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMscited by 570
  9. 2025Chain of Draft: Thinking Faster by Writing Lesscited by 297
  10. 2025Towards Large Reasoning Models: A Survey of Reinforced Reasoning with Large Language Modelscited by 293
  11. 2024Token-Budget-Aware LLM Reasoningcited by 247
  12. 2025CODI: Compressing Chain-of-Thought into Continuous Space via Self-Distillationcited by 225
17 more
  1. 2024Compressed Chain of Thought: Efficient Reasoning Through Dense Representationscited by 201
  2. 2024C3oT: Generating Shorter Chain-of-Thought without Compromising Effectivenesscited by 119
  3. 2025Towards Thinking-Optimal Scaling of Test-Time Compute for LLM Reasoningcited by 119
  4. 2025Sketch-of-Thought: Efficient LLM Reasoning with Adaptive Cognitive-Inspired Sketchingcited by 117
  5. 2024Fast Best-of-N Decoding via Speculative Rejectioncited by 112
  6. 2025Reward-Guided Speculative Decoding for Efficient LLM Reasoningcited by 106
  7. 2025LightThinker: Thinking Step-by-Step Compressioncited by 104
  8. 2025Self-Training Elicits Concise Reasoning in Large Language Modelscited by 99
  9. 2025How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approachcited by 97
  10. 2025Token Assorted: Mixing Latent and Text Tokens for Improved Language Model Reasoningcited by 91
  11. 2025Sampling-Efficient Test-Time Scaling: Self-Estimating the Best-of-N Sampling in Early Decodingcited by 63
  12. 2024Can Language Models Learn to Skip Steps?cited by 59
  13. 2024Efficiently Scaling LLM Reasoning with Certaindexcited by 58
  14. 2025InftyThink: Breaking the Length Limits of Long-Context Reasoning in Large Language Modelscited by 54
  15. 2025Dynamic Parallel Tree Search for Efficient LLM Reasoningcited by 42
  16. 2024REVOLVE: Optimizing AI Systems by Tracking Response Evolution in Textual Optimizationcited by 20
  17. 2025Prompt Stability Matters: Evaluating and Optimizing Auto-Generated Prompt in General-Purpose Systemscited by 10

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 3.

5 papers.

  1. 2025A Survey of Context Engineering for Large Language Modelscited by 126
  2. 2026Neural Chain-of-Thought Search: Searching the Optimal Reasoning Path to Enhance Large Language Modelscited by 8
  3. 2025Synthetic Data-Driven Prompt Tuning for Financial QA over Tables and Documentscited by 0
  4. 2026C-MOP: Integrating Momentum and Boundary-Aware Clustering for Enhanced Prompt Evolutioncited by 0
  5. 2026MetaCtrl: Your Large Language Models Can Reason Better and More Concisely with a Metacognitive Controllercited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024MindStar: Enhancing Math Reasoning in Pre-trained LLMs at Inference Time
  2. 2025Systematic Optimization of Open Source Large Language Models for Mathematical Reasoning
  3. 2026CROP: Token-Efficient Reasoning in Large Language Models via Regularized Prompt Optimization
  4. 2025Scalable LLM Reasoning Acceleration with Low-rank Distillation
  5. 2023Query-Dependent Prompt Evaluation and Optimization with Offline Inverse RL
  6. 2025A Survey on Mathematical Reasoning and Optimization with Large Language Models
  7. 2025TRPrompt: Bootstrapping Query-Aware Prompt Optimization from Textual Rewards
  8. 2025Prompting Test-Time Scaling Is A Strong LLM Reasoning Data Augmentation