BaohaoLiao / RSDLinks
[ICML 2025] Reward-guided Speculative Decoding (RSD) for efficiency and effectiveness.
☆31Updated last month
Alternatives and similar repositories for RSD
Users that are interested in RSD are comparing it to the libraries listed below
Sorting:
- PoC for "SpecReason: Fast and Accurate Inference-Time Compute via Speculative Reasoning" [arXiv '25]☆37Updated 3 weeks ago
- Official repository for paper: O1-Pruner: Length-Harmonizing Fine-Tuning for O1-Like Reasoning Pruning☆79Updated 3 months ago
- Simple extension on vLLM to help you speed up reasoning model without training.☆152Updated this week
- The official implementation of the paper "Towards Efficient Mixture of Experts: A Holistic Study of Compression Techniques (TMLR)".☆70Updated 2 months ago
- Official implementation for Yuan & Liu & Zhong et al., KV Cache Compression, But What Must We Give in Return? A Comprehensive Benchmark o…☆78Updated 3 months ago
- Unofficial implementation for the paper "Mixture-of-Depths: Dynamically allocating compute in transformer-based language models"☆161Updated 11 months ago
- ☆79Updated 4 months ago
- ☆42Updated 6 months ago
- L1: Controlling How Long A Reasoning Model Thinks With Reinforcement Learning☆213Updated 3 weeks ago
- [ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆116Updated 6 months ago
- TokenSkip: Controllable Chain-of-Thought Compression in LLMs☆145Updated 2 months ago
- [NeurIPS 24 Spotlight] MaskLLM: Learnable Semi-structured Sparsity for Large Language Models☆166Updated 5 months ago
- ☆129Updated 3 months ago
- The official implementation of the paper <MoA: Mixture of Sparse Attention for Automatic Large Language Model Compression>☆129Updated last week
- [ACL 2024] Not All Experts are Equal: Efficient Expert Pruning and Skipping for Mixture-of-Experts Large Language Models☆89Updated last year
- ☆107Updated 2 weeks ago
- Code for paper: [ICLR2025 Oral] FlexPrefill: A Context-Aware Sparse Attention Mechanism for Efficient Long-Sequence Inference☆107Updated 2 weeks ago
- ☆50Updated 7 months ago
- Efficient triton implementation of Native Sparse Attention.☆155Updated 2 weeks ago
- The official implementation of paper: SimLayerKV: A Simple Framework for Layer-Level KV Cache Reduction.☆46Updated 7 months ago
- ☆45Updated last year
- [ICLR 2024 Spotlight] Code for the paper "Merge, Then Compress: Demystify Efficient SMoE with Hints from Its Routing Policy"☆84Updated last year
- ☆64Updated last month
- Research Code for preprint "Optimizing Test-Time Compute via Meta Reinforcement Finetuning".☆94Updated 2 months ago
- [ICLR 2025] SuperCorrect: Advancing Small LLM Reasoning with Thought Template Distillation and Self-Correction☆70Updated 2 months ago
- A version of verl to support tool use☆172Updated this week
- [ICLR 2025] SWIFT: On-the-Fly Self-Speculative Decoding for LLM Inference Acceleration☆49Updated 3 months ago
- Ouroboros: Speculative Decoding with Large Model Enhanced Drafting (EMNLP 2024 main)☆106Updated 2 months ago
- ☆93Updated 2 weeks ago
- Homepage for ProLong (Princeton long-context language models) and paper "How to Train Long-Context Language Models (Effectively)"☆186Updated 2 months ago