English

Training-Inference Consistent Segmented Execution for Long-Context LLMs

Computation and Language 2026-05-13 v1 Machine Learning

Abstract

Transformer-based large language models face severe scalability challenges in long-context generation due to the computational and memory costs of full-context attention. Under practical computation and memory constraints, many inference-efficient long-context methods improve efficiency by adopting bounded-context or segment-level execution only during inference, while continuing to train models under full-context attention, resulting in a mismatch between training and inference execution and state-transition semantics. Based on this insight, we propose a training-inference consistent segment-level generation framework, in which training and inference follow the same segment-level forward execution semantics. During training, consistency with inference is enforced by restricting gradient propagation to KV states carried over from the immediately preceding segment, while permitting head-specific access to past KV states during the forward pass without involving them in gradient propagation. Across long-context benchmarks, our approach achieves performance comparable to full-context attention, while achieving competitive latency-memory trade-offs against strong inference-efficient baselines, and substantially improving scalability at very long context lengths (e.g., approximately 6x lower peak prefill memory at 128K compared to full-context attention with FlashAttention).

Keywords

Cite

@article{arxiv.2605.11744,
  title  = {Training-Inference Consistent Segmented Execution for Long-Context LLMs},
  author = {Xianpeng Shang and Jiang Li and Zehua Duo and Qianyi Cai and Xiangdong Su},
  journal= {arXiv preprint arXiv:2605.11744},
  year   = {2026}
}

Comments

Accepted by ICML 2026. 19 pages, 6 figures, 3 tables