User-friendly implementation of the Mixture-of-Sparse-Attention (MoSA). MoSA selects distinct tokens for each head with expert choice routing providing a content-based sparse attention mechanism.
☆28May 3, 2025Updated 10 months ago
Alternatives and similar repositories for MoSA
Users that are interested in MoSA are comparing it to the libraries listed below
Sorting:
- [ICML 2025] LaCache: Ladder-Shaped KV Caching for Efficient Long-Context Modeling of Large Language Models☆17Nov 4, 2025Updated 4 months ago
- ☆16Sep 17, 2024Updated last year
- ☆11Aug 13, 2024Updated last year
- Mixture-of-Basis-Experts for Compressing MoE-based LLMs☆29Dec 24, 2025Updated 2 months ago
- Codes for our paper "AgentMonitor: A Plug-and-Play Framework for Predictive and Secure Multi-Agent Systems"☆13Dec 13, 2024Updated last year
- The official implementation of HybridNorm: Towards Stable and Efficient Transformer Training via Hybrid Normalization☆18Mar 7, 2025Updated 11 months ago
- [ICLR 2026] BARREL: Boundary-Aware Reasoning for Factual and Reliable LRMs☆17May 21, 2025Updated 9 months ago
- [ICLR 2026] Official repo for "Spotlight on Token Perception for Multimodal Reinforcement Learning"☆49Jan 30, 2026Updated last month
- Don't just regulate gradients like in Muon, regulate the weights too☆31Jul 30, 2025Updated 7 months ago
- ☆12Apr 29, 2024Updated last year
- RAG-RewardBench: Benchmarking Reward Models in Retrieval Augmented Generation for Preference Alignment☆16Dec 19, 2024Updated last year
- CoV: Chain-of-View Prompting for Spatial Reasoning☆51Jan 23, 2026Updated last month
- An End-to-End Model with Adaptive Filtering for Retrieval-Augmented Generation☆16Oct 27, 2024Updated last year
- [ICLR-2026] Official Implementation of our paper "THOR: Tool-Integrated Hierarchical Optimization via RL for Mathematical Reasoning".☆31Feb 26, 2026Updated last week
- ☆23May 21, 2025Updated 9 months ago
- ☆14Oct 3, 2024Updated last year
- MobileLLM-R1☆76Sep 30, 2025Updated 5 months ago
- Code for paper: Unified Text-to-Image Generation and Retrieval☆16Jul 6, 2024Updated last year
- Code and Data for "FaithfulRAG: Fact-Level Conflict Modeling for Context-Faithful Retrieval-Augmented Generation" (ACL25)☆29Oct 26, 2025Updated 4 months ago
- The official implementation of Preference Data Reward-Augmentation.☆18May 1, 2025Updated 10 months ago
- This is a simple torch implementation of the high performance Multi-Query Attention☆16Aug 23, 2023Updated 2 years ago
- [ACL 2025] Squeezed Attention: Accelerating Long Prompt LLM Inference☆57Nov 20, 2024Updated last year
- ☆21Jan 16, 2025Updated last year
- ☆18Dec 2, 2024Updated last year
- Official code repo for paper "Great Memory, Shallow Reasoning: Limits of kNN-LMs"☆23Apr 30, 2025Updated 10 months ago
- [NeurIPS 2024] | An Efficient Recipe for Long Context Extension via Middle-Focused Positional Encoding☆22Oct 10, 2024Updated last year
- ☆22Mar 7, 2025Updated 11 months ago
- Source code of our EMNLP 2024 paper "FactAlign: Long-form Factuality Alignment of Large Language Models"☆19Oct 3, 2024Updated last year
- ☆19Jan 10, 2025Updated last year
- ☆28Aug 13, 2025Updated 6 months ago
- Evaluating the faithfulness of long-context language models☆30Oct 21, 2024Updated last year
- Official Implementation of FastKV: Decoupling of Context Reduction and KV Cache Compression for Prefill-Decoding Acceleration☆29Nov 22, 2025Updated 3 months ago
- MetaAgent: Toward Self-Evolving Agent via Tool Meta-Learning☆42Sep 3, 2025Updated 6 months ago
- Suri: Multi-constraint instruction following for long-form text generation (EMNLP’24)☆27Oct 3, 2025Updated 5 months ago
- The official implementation of paper: SimLayerKV: A Simple Framework for Layer-Level KV Cache Reduction.☆51Oct 18, 2024Updated last year
- ☆134May 29, 2025Updated 9 months ago
- ☆49Aug 14, 2025Updated 6 months ago
- ☆30Oct 4, 2025Updated 5 months ago
- Modality Gap–Driven Subspace Alignment Training Paradigm For Multimodal Large Language Models☆51Feb 23, 2026Updated last week