Codes for the paper "BAPO: Stabilizing Off-Policy Reinforcement Learning for LLMs via Balanced Policy Optimization with Adaptive Clipping" by Zhiheng Xi et al.
☆91Jan 29, 2026Updated last month
Alternatives and similar repositories for BAPO
Users that are interested in BAPO are comparing it to the libraries listed below
Sorting:
- The open-source code for the NeurIPS 2025 paper, "Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learn…☆45Jan 5, 2026Updated last month
- ☆105Dec 5, 2025Updated 2 months ago
- Use the tokenizer in parallel to achieve superior acceleration☆20Mar 21, 2024Updated last year
- ☆27Jun 5, 2023Updated 2 years ago
- The training codes of Jasper-Token-Compression-600M☆19Nov 19, 2025Updated 3 months ago
- This is the code repo for the paper <UTC-IE: A Unified Token-pair Classification Architecture for Information Extraction>☆15Aug 10, 2023Updated 2 years ago
- ☆30Sep 15, 2025Updated 5 months ago
- "FusionFactory: Fusing LLM Capabilities with Routing Data", Tao Feng, Haozhen Zhang, Zijie Lei, Pengrui Han, Mostofa Patwary, Mohammad Sh…☆19Dec 30, 2025Updated 2 months ago
- A collection of awesome think with videos papers.☆90Dec 1, 2025Updated 3 months ago
- Multi-GPU supported kmeans clustering for cluser-clip☆15Jun 3, 2024Updated last year
- [CVPR 2026] Official Code for "ARM-Thinker: Reinforcing Multimodal Generative Reward Models with Agentic Tool Use and Visual Reasoning"☆84Feb 13, 2026Updated 2 weeks ago
- ☆36Dec 16, 2025Updated 2 months ago
- ☆138Feb 13, 2026Updated 2 weeks ago
- FREECODEC: A DISENTANGLED NEURAL SPEECH CODEC WITH FEWER TOKENS☆24Sep 9, 2024Updated last year
- "Omni-R1: Towards the Unified Generative Paradigm for Multimodal Reasoning"☆53Jan 28, 2026Updated last month
- Training tiny models to prove hard theorems☆41Feb 15, 2026Updated 2 weeks ago
- [NeurIPS 2024] Can Language Models Learn to Skip Steps?☆22Jan 25, 2025Updated last year
- Codes for Paper: From Hypergraph Energy Functions to Hypergraph Neural Networks☆23Jun 29, 2023Updated 2 years ago
- NexAU (AU for Agent Universe), a general-purpose agent framework for building intelligent agents with tool capabilities.☆49Updated this week
- A platform for building configurable, database-backed generative AI agentic assistants.☆25Feb 11, 2025Updated last year
- ☆20Jul 5, 2024Updated last year
- ☆19Mar 3, 2025Updated last year
- [Findings of EMNLP'2024] Unified Active Retrieval for Retrieval Augmented Generation☆23Sep 30, 2024Updated last year
- ☆23Dec 8, 2022Updated 3 years ago
- Official Implementation of "Maximum Likelihood Reinforcement Learning (MaxRL)"☆134Updated this week
- code for Preprint paper at Arxiv: MoT: Pre-thinking and Recalling Enable ChatGPT to Self-Improve with Memory-of-Thoughts☆24Nov 29, 2023Updated 2 years ago
- ☆162Dec 18, 2025Updated 2 months ago
- [NeurIPS 2025] BOOM, A Planning-driven Model-Based RL algorithm☆58Feb 4, 2026Updated last month
- a survey of long-context LLMs from four perspectives, architecture, infrastructure, training, and evaluation☆61Mar 31, 2025Updated 11 months ago
- The tool facilitates debugging convergence issues and testing new algorithms and recipes for training LLMs using Nvidia libraries such as…☆18Sep 17, 2025Updated 5 months ago
- Official repo for "PAPO: Perception-Aware Policy Optimization for Multimodal Reasoning"☆117Feb 4, 2026Updated last month
- Measuring how well CLI agents like Claude Code or Codex CLI can post-train base LLMs on a single H100 GPU in 10 hours☆166Updated this week
- [CVPR 2026] OpenMMReasoner: Pushing the Frontiers for Multimodal Reasoning with an Open and General Recipe☆145Feb 23, 2026Updated last week
- PeRL: Parameter-Efficient Reinforcement Learning☆71Feb 23, 2026Updated last week
- [EMNLP 2023] Plan, Verify and Switch: Integrated Reasoning with Diverse X-of-Thoughts☆27Nov 4, 2023Updated 2 years ago
- StereoVLA is powered by stereo vision and supports flexible deployment with high tolerance to camera pose variations.☆52Jan 12, 2026Updated last month
- [KernelGYM & Dr. Kernel] A distributed GPU environment and a collection of RL training methods to support RL for Kernel Generations☆90Feb 6, 2026Updated 3 weeks ago
- EO: Open-source Unified Embodied Foundation Model Series☆51Jan 15, 2026Updated last month
- This code accompanies the paper "Leveraging Skills from Unlabeled Prior Data for Efficient Online Exploration."☆37Jul 11, 2025Updated 7 months ago