FOLD: Position-Free Far Memory with Local Order for Long-Context Speculative Decoding

arXiv:2610.20392 · cs.CL, cs.LG · Submitted 2026-10-14 · Read on arXiv

cs.CL, cs.LG

Submitted: 2026-10-14

Updated: 2026-10-14

Code: https://github.com/NuyoahNasuS/FOLD

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Speculative Decoding — 1× in this paper · explained in 2 episodes

Sources

• Training-Free Long-Context Scaling of Large Language Models
• Long Code Arena: a Set of Benchmarks for Long-Context Code Models
• Accelerating Large Language Model Decoding with Speculative Sampling
• Extending Context Window of Large Language Models via Positional Interpolation
• DSpark: Confidence-Scheduled Speculative Decoding with Semi-Autoregressive Generation
• The Llama 3 Herd of Models
• Domino: Decoupling Causal Modeling from Autoregressive Drafting in Speculative Decoding
• H-Spec: Parallel Speculative Decoding Without a Drafter-Side KV Cache
• LLM Maybe LongLM: Self-Extend LLM Context Window Without Tuning
• LiLiCorr: Lightweight Likelihood Correlation of Parallel Drafts for Speculative Decoding
• Jet-Long: Efficient Long-Context Extension with Dynamic Bifocal RoPE
• Qwen3 Technical Report
• Vegas: Self-Speculative Decoding with Verification-Guided Sparse Attention
• TraceLab: Characterizing Coding Agent Workloads for LLM Serving

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering