BaohaoLiao / RSDLinks
[ICML 2025] Reward-guided Speculative Decoding (RSD) for efficiency and effectiveness.
☆32Updated 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]☆39Updated last month
- Simple extension on vLLM to help you speed up reasoning model without training.☆161Updated 3 weeks ago
- L1: Controlling How Long A Reasoning Model Thinks With Reinforcement Learning☆222Updated last month
- Research Code for preprint "Optimizing Test-Time Compute via Meta Reinforcement Finetuning".☆94Updated 3 months ago
- [ICLR 2025] SWIFT: On-the-Fly Self-Speculative Decoding for LLM Inference Acceleration☆51Updated 4 months ago
- Unofficial implementation for the paper "Mixture-of-Depths: Dynamically allocating compute in transformer-based language models"☆163Updated last year
- Activation-aware Singular Value Decomposition for Compressing Large Language Models☆71Updated 8 months ago
- [NeurIPS'24] Official code for *🎯DART-Math: Difficulty-Aware Rejection Tuning for Mathematical Problem-Solving*☆108Updated 6 months ago
- Official repository for paper: O1-Pruner: Length-Harmonizing Fine-Tuning for O1-Like Reasoning Pruning☆82Updated 4 months ago
- ☆80Updated 5 months ago
- [NeurIPS 24 Spotlight] MaskLLM: Learnable Semi-structured Sparsity for Large Language Models☆168Updated 5 months ago
- [ICLR 2024 Spotlight] Code for the paper "Merge, Then Compress: Demystify Efficient SMoE with Hints from Its Routing Policy"☆85Updated this week
- The official implementation of the paper "Towards Efficient Mixture of Experts: A Holistic Study of Compression Techniques (TMLR)".☆71Updated 3 months ago
- [ACL 2024] Not All Experts are Equal: Efficient Expert Pruning and Skipping for Mixture-of-Experts Large Language Models☆92Updated last year
- ☆42Updated 7 months ago
- ☆104Updated 2 weeks ago
- ☆130Updated 4 months ago
- Code for "Critique Fine-Tuning: Learning to Critique is More Effective than Learning to Imitate"☆159Updated 3 weeks ago
- [NeurIPS 2024] Fast Best-of-N Decoding via Speculative Rejection☆45Updated 7 months ago
- Repo for "Z1: Efficient Test-time Scaling with Code"☆61Updated 2 months ago
- Layer-Condensed KV cache w/ 10 times larger batch size, fewer params and less computation. Dramatic speed up with better task performance…☆149Updated 2 months ago
- ☆58Updated this week
- ☆45Updated last week
- Official Implementation of SLEB: Streamlining LLMs through Redundancy Verification and Elimination of Transformer Blocks☆37Updated 4 months ago
- [ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆116Updated 6 months ago
- Unofficial implementations of block/layer-wise pruning methods for LLMs.☆70Updated last year
- [ICML 2025] "SepLLM: Accelerate Large Language Models by Compressing One Segment into One Separator"☆75Updated this week
- Code for the EMNLP24 paper "A simple and effective L2 norm based method for KV Cache compression."☆12Updated 6 months ago
- ☆37Updated 9 months ago
- Code for paper: [ICLR2025 Oral] FlexPrefill: A Context-Aware Sparse Attention Mechanism for Efficient Long-Sequence Inference☆113Updated last month