RECAP: Context-Aware Prediction Recycling with Request-Local and Persistent Memories for Multimodal Speculative Decoding
cs.AI
Submitted: 2026-11-04
Updated: 2026-11-04
Code: https://github.com/Senmo996/RECAPofficial-code
Terminology
Sources
- Hydra: Sequentially-Dependent Draft Heads for Medusa Decoding
- Qwen2.5-VL Technical Report
- Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads
- Accelerating Large Language Model Decoding with Speculative Sampling
- Break the Sequential Dependency of LLM Inference Using Lookahead Decoding
- Improving Neural Language Models with a Continuous Cache
- SpecVLM: Fast Speculative Decoding in Vision-Language Models
- Generalization through Memorization: Nearest Neighbor Language Models
- SEED-Bench: Benchmarking Multimodal LLMs with Generative Comprehension
- EAGLE: Speculative Sampling Requires Rethinking Feature Uncertainty
- Speculative Decoding Reimagined for Multimodal Large Language Models
- MMSpec: Benchmarking Speculative Decoding for Vision-Language Models
- MMT-Bench: A Comprehensive Multimodal Benchmark for Evaluating Large Vision-Language Models Towards Multitask AGI
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents