Small and Efficient Mathematical Reasoning LLMs
-
Updated
Jan 27, 2024 - Python
#
gsm8k
Here are 77 public repositories matching this topic...
[ACL 2023] Solving Math Word Problems via Cooperative Reasoning induced Language Models (LLMs + MCTS + Self-Improvement)
nlp acl mcts gpt language-model monte-carlo-tree-search bert reasoning deberta gpt-j math-word-problem gpt-j-6b llms chain-of-thought acl2023 gsm8k
-
Updated
Dec 15, 2023 - Python
Reproducible Language Agent Research
react agent reasoning llm chain-of-thought chatgpt gsm8k program-of-thoughts tree-of-thoughts language-agent graph-of-thoughts doubao agent-leaderboard
-
Updated
Jun 25, 2025 - Python
DICE: Detecting In-distribution Data Contamination with LLM's Internal State
-
Updated
Sep 21, 2024 - Python
Fast runtime for Random Optimization at scale.
-
Updated
Jun 6, 2026 - Python
GRPO and SFT Finetune Qwen3 using Unsloth : Reasoning and Non-Reasoning Dataset
-
Updated
Aug 21, 2026 - Jupyter Notebook
Prompt-engineering study on LLM math reasoning (GSM8K) and code generation (HumanEval): zero/few-shot, self-consistency, self-verification, and experiments on prompt quality, complexity, demonstrations, and diversity.
nlp code-generation self-consistency few-shot-learning humaneval llm prompt-engineering chain-of-thought gsm8k math-reasoning
-
Updated
Jun 27, 2026 - Python
CS336 作业 5：基于 Qwen2.5 模型的 LLM 对齐与推理强化学习。完整实现了监督微调（SFT）与组相对策略优化（GRPO）算法，并在 GSM8K 数据集上完成零样本、在策与离策的训练与评估对比。
-
Updated
Apr 1, 2026 - Python
Short-CoT distilled GSM8K dataset generated with OpenAI gpt-oss-120b.
-
Updated
Feb 19, 2026
OpenCompass evaluation harness inspired by Agent0, with reproducible GPT-4o-mini runs on GSM8K and MATH; co-training is not yet implemented.
-
Updated
Aug 20, 2026 - Python
GSM8K-Consistency is a benchmark database for analyzing the consistency of Arithmetic Reasoning on GSM8K.
prompt prompt-toolkit grade reasoning self-consistency mathematical-reasoning factual-consistency foundation-models large-language-models prompt-engineering arithmetic-reasoning gsm8k grade-school-math logical-consistency self-consistency-benchmark semantics-consistency prompt-perturbation arithmetic-consistency semantics-preserving-transformations semantics-similar
-
Updated
Dec 31, 2023
Hard Reasoning Benchmark filtered with disagreement scores
-
Updated
Feb 14, 2026 - Python
Reproduction & re-implementation of the AloLab paper (arXiv:2605.02363): closing the structured-output reliability gap in small LLMs via iterative black-box prompt optimization. GSM8K, 4 models, ablations, McNemar significance.
python nlp json reproducibility structured-output constrained-decoding large-language-models llm prompt-engineering gsm8k prompt-optimization small-language-models
-
Updated
Jul 25, 2026 - Python
A controlled LoRA finetuning study on process supervision for mathematical reasoning with Qwen2.5-Math-7B-Instruct.
-
Updated
Apr 23, 2026 - Python
An evaluation of prompting techniques (Zero-Shot CoT, Few-Shot, Self-Consistency) on the Mistral-7B model for mathematical reasoning. This project systematically benchmarks 7 distinct methods on the GSM8K dataset.
deep-learning university-project course-project zero-shot reasoning self-consistency few-shot majority-voting mathematical-reasoning huggingface-transformers large-language-models llm prompt-engineering chain-of-thought question-decomposition gsm8k chain-of-thought-reasoning llm-evaluation mistral-7b reasoning-language-models
-
Updated
Nov 2, 2025 - Python
Code and results for consistency-gated self-correction in large language model reasoning.
-
Updated
Jun 16, 2026 - Python
大模型评测平台 — 本地/API/HuggingFace/OpenCompass 三路后端，支持数据生产(Self-Instruct/Evol-Instruct)、长尾场景生成、弱项挖掘、回归分析、污染检测、Bad Case归因。可扩展的 Benchmark 系统和 LLM-as-Judge 自动评分。
python flask humaneval ai-evaluation gsm8k mmlu llm-evaluation benchmark-platform rag-evaluation llm-as-judge opencompass llm-benchmark data-contamination-detection
-
Updated
Jun 7, 2026 - Python
Improve this page
Add a description, image, and links to the gsm8k topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the gsm8k topic, visit your repo's landing page and select "manage topics."