超简单复现Deepseek-R1-Zero和Deepseek-R1,以「24点游戏」为例。通过zero-RL、SFT以及SFT+RL,以激发LLM的自主验证反思能力。 About Clean, minimal, accessible reproduction of DeepSeek R1-Zero, DeepSeek R1
☆35Apr 5, 2025Updated last year
Alternatives and similar repositories for 24-Game-Reasoning
Users that are interested in 24-Game-Reasoning are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Long CoT Fine-Tuning and Reinforcement Learning for LLMs in the Context of the 24-Point Game: A Toy Project☆27Feb 22, 2025Updated last year
- [ EMNLP 2025 Main ] Enhancing Efficiency and Exploration in Reinforcement Learning for LLMs☆18Nov 7, 2025Updated last year
- Official implementation of IEEE S&P 2025 paper "Prompt Inversion Attack against Collaborative Inference of Large Language Models"☆17Apr 27, 2025Updated last year
- The official implementation of "ICDPO: Effectively Borrowing Alignment Capability of Others via In-context Direct Preference Optimization…☆16Feb 15, 2024Updated 2 years ago
- Training Coder Models with Critique Reinforcement Learning [ICLR26]☆13Apr 11, 2026Updated 7 months ago
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- The rule-based evaluation subset and code implementation of Omni-MATH☆29Dec 23, 2024Updated last year
- SFT+RL boosts multimodal reasoning☆50Jun 27, 2025Updated last year
- [NeurIPS 2026 Pre-to-Post] From P(y|x) to P(y): Investigating Reinforcement Learning in Pre-train Space☆20Apr 16, 2026Updated 7 months ago
- [NeurIPS 2025 (Spotlight✨)] TIME: A Multi-level Benchmark for Temporal Reasoning of LLMs in Real-World Scenario☆33Sep 29, 2026Updated last month
- RollPacker: Taming Long-Tail Rollouts for RL Post-Training with Tail Batching☆12Mar 2, 2026Updated 8 months ago
- ☆34Feb 28, 2026Updated 8 months ago
- Official PyTorch implementation of "Q-Palette: Fractional-Bit Quantizers Toward Optimal Bit Allocation for Efficient LLM Deployment" (Neu…☆12Jan 22, 2026Updated 9 months ago
- Code for the paper 'Neural Variational Gradient Descent'. We perform nonparametric variational inference by transporting samples along a …☆12Jul 29, 2021Updated 5 years ago
- 这是一个open-r1的复现项目,对0.5B、1.5B、3B、7B的qwen模型进行GRPO训练,观察到一些有趣的现象。☆62Apr 13, 2025Updated last year
- End-to-end encrypted email - Proton Mail • AdSpecial offer: 40% Off Yearly / 80% Off First Month. All Proton services are open source and independently audited for security.
- Official code and dataset for our paper: RefineBench: Evaluating Refinement Capability of Language Models via Checklists☆16Dec 1, 2025Updated 11 months ago
- ☆15Jan 18, 2026Updated 10 months ago
- The official repository of the Omni-MATH benchmark.☆93Dec 22, 2024Updated last year
- 这个仓库里面包含小样本学习的经典代码,包括网上整理,以及论文复现,是您入门的不二选择☆10Aug 9, 2025Updated last year
- Efficient retrieval head analysis with triton flash attention that supports topK probability☆13Jun 15, 2024Updated 2 years ago
- ☆10Oct 21, 2024Updated 2 years ago
- ☆17Oct 15, 2023Updated 3 years ago
- FORTE (Full-cycle Office Real-world Task Evaluation) is a general agent benchmark for evaluating AI agents on daily office productivity a…☆20Jun 30, 2026Updated 4 months ago
- 《安全多方计算——可证明安全视角》发布地址☆14Nov 9, 2026Updated last week
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- 本项目对Deepseek-R1-Distill-Qwen-7B进行心理咨询CoT数据的LoRA微调,以进一步提升Deepseek-R1-Distill-Qwen-7B在心理咨询领域的慢思考能力。☆12Mar 11, 2025Updated last year
- More Tokens, Lower Precision: Towards the Optimal Token-Precision Trade-off in KV Cache Compression(EMNLP 2025)☆12Jan 15, 2025Updated last year
- Code for Findings of EMNLP2023 paper "Coarse-to-Fine Dual Encoders are Better Frame Identification Learners"☆13Oct 15, 2023Updated 3 years ago
- TOD-Flow: Modeling the Structure of Task-Oriented Dialogues☆13Feb 7, 2024Updated 2 years ago
- ☆11Mar 24, 2021Updated 5 years ago
- This repository includes code and materials for the paper "Efficient PRM Training Data Synthesis via Formal Verification" (ACL 2026 Findi…☆21Apr 7, 2026Updated 7 months ago
- kaggle:otto competition☆24Feb 13, 2023Updated 3 years ago
- ☆10Jul 11, 2022Updated 4 years ago
- ☆10May 28, 2023Updated 3 years ago
- Bare Metal GPUs on DigitalOcean Gradient AI • AdPurpose-built for serious AI teams training foundational models, running large-scale inference, and pushing the boundaries of what's possible.
- ☆10Jun 29, 2022Updated 4 years ago
- AdaptiveStep: Automatically Dividing Reasoning Step through Model Confidence☆11Mar 2, 2025Updated last year
- ☆26Apr 9, 2025Updated last year
- chatglm_rlhf_finetuning☆30Oct 10, 2023Updated 3 years ago
- SafeWorld: Geo-Diverse Safety Alignment (NeurIPS 2024)☆10Dec 10, 2024Updated last year
- A collection of some awesome public projects about LLM-based Web Agents and Tools.☆14Apr 25, 2024Updated 2 years ago
- diagnosis_zero, R1 Zero reproduce on disease diagnosis☆33Jul 24, 2025Updated last year
Are these results useful?