Training Variable Long Sequences with Data-Centric Parallel

• URL: http://arxiv.org/abs/2608.07524v1
• Date: Tue, 14 Jul 2026 17:54:32 GMT
• Title: Training Variable Long Sequences with Data-Centric Parallel
• Abstract summary: Training deep learning models on variable long sequences poses significant computational challenges.<n>Existing methods force a difficult trade-off between efficiency and ease-of-use.<n>We introduce Data-Centric Parallel (DCP), which lets the data itself drive the runtime.
• Score: 13.581830062964025
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Training deep learning models on variable long sequences poses significant computational challenges. Existing methods force a difficult trade-off between efficiency and ease-of-use. Simple approaches use static configurations that cause workload imbalance low efficiency, while complex methods introduces significant complexity and code change for new models. To break this trade-off, we introduce Data-Centric Parallel (DCP). Its core principle is to let the data itself drive the runtime. It achieves this by dynamically adjusting direct runtime settings (e.g., parallel size, gradient accumulation, recomputation) based on each batch's sequence length. Empirical results demonstrate that our method achieves up to a 2.88$\times$ speedup on 32 H200 GPUs. Designed for generalization, it can be integrated into any model with 10 lines of code. We anticipate this simple yet effective approach will serve as a robust baseline and facilitate future advancements in distributed training for variable long sequences.

Related papers

• Training LLMs for Multi-Step Tool Orchestration with Constrained Data Synthesis and Graduated Rewards [76.49428173793386]
  LLMs must invoke multiple dependent APIs in the correct order while propagating intermediate outputs.<n>Training models to handle such faces two obstacles: existing environments focus on simple per-turn function calls with simulated data, and binary rewards.<n>We present a framework addressing both challenges. First, we construct a reinforcement learning environment backed by a large-scale cache of real API responses, enabling a data synthesis pipeline that samples valid multi-step orchestration traces.<n>Second, we propose a graduated reward design that decomposes correctness into atomic validity and orchestration.
  arXiv  Detail & Related papers  (2026-03-25T18:31:39Z)
• Hierarchical Balance Packing: Towards Efficient Supervised Fine-tuning for Long-Context LLM [49.2709992932292]
  Training Long-Context Large Language Models (LLMs) is challenging, as hybrid training with long-context and short-context data often leads to workload imbalances.<n>Existing works mainly use data packing to alleviate this issue, but fail to consider imbalanced attention computation and wasted communication overhead.<n>This paper proposes Hierarchical Balance Packing (HBP), which designs a novel batch-construction method and training recipe to address those inefficiencies.
  arXiv  Detail & Related papers  (2025-03-10T10:52:50Z)
• Hydraulis: Balancing Large Transformer Model Training via Co-designing Parallel Strategies and Data Assignment [25.38589021788461]
  We develop Hydraulis, which jointly optimize the parallel strategies and data assignment.<n> Empirical results demonstrate that Hydraulis outperforms existing systems by 1.32-2.66 times.
  arXiv  Detail & Related papers  (2024-12-10T20:01:53Z)
• DeepSpeed Ulysses: System Optimizations for Enabling Training of Extreme Long Sequence Transformer Models [34.74093040678323]
  We introduce DeepSpeed-Ulysses, a novel, portable and effective methodology for enabling highly efficient and scalable LLM training. DeepSpeed-Ulysses at its core partitions input data along the sequence dimension and employs an efficient all-to-all collective communication for attention. Experiments show that DeepSpeed-Ulysses trains 2.5x faster with 4x longer sequence length than the existing method SOTA baseline.
  arXiv  Detail & Related papers  (2023-09-25T20:15:57Z)
• Parallel $Q$-Learning: Scaling Off-policy Reinforcement Learning under Massively Parallel Simulation [17.827002299991285]
  Reinforcement learning is time-consuming for complex tasks due to the need for large amounts of training data. Recent advances in GPU-based simulation, such as Isaac Gym, have sped up data collection thousands of times on a commodity GPU. This paper presents a Parallel $Q$-Learning scheme that outperforms PPO in wall-clock time.
  arXiv  Detail & Related papers  (2023-07-24T17:59:37Z)
• Does compressing activations help model parallel training? [64.59298055364336]
  We present the first empirical study on the effectiveness of compression methods for model parallelism. We implement and evaluate three common classes of compression algorithms. We evaluate these methods across more than 160 settings and 8 popular datasets.
  arXiv  Detail & Related papers  (2023-01-06T18:58:09Z)
• Efficient Sub-structured Knowledge Distillation [52.5931565465661]
  We propose an approach that is much simpler in its formulation and far more efficient for training than existing approaches. We transfer the knowledge from a teacher model to its student model by locally matching their predictions on all sub-structures, instead of the whole output space.
  arXiv  Detail & Related papers  (2022-03-09T15:56:49Z)
• Colossal-AI: A Unified Deep Learning System For Large-Scale Parallel Training [23.633810934134065]
  Colossal-AI can achieve up to 2.76 times training speedup on large-scale models. System supports parallel training methods such as data, pipeline, tensor, and sequence parallelism.
  arXiv  Detail & Related papers  (2021-10-28T04:45:55Z)
• Scaling Distributed Deep Learning Workloads beyond the Memory Capacity with KARMA [58.040931661693925]
  We propose a strategy that combines redundant recomputing and out-of-core methods. We achieve an average of 1.52x speedup in six different models over the state-of-the-art out-of-core methods. Our data parallel out-of-core solution can outperform complex hybrid model parallelism in training large models, e.g. Megatron-LM and Turning-NLG.
  arXiv  Detail & Related papers  (2020-08-26T07:24:34Z)
• Dynamic Scale Training for Object Detection [111.33112051962514]
  We propose a Dynamic Scale Training paradigm (abbreviated as DST) to mitigate scale variation challenge in object detection. Experimental results demonstrate the efficacy of our proposed DST towards scale variation handling. It does not introduce inference overhead and could serve as a free lunch for general detection configurations.
  arXiv  Detail & Related papers  (2020-04-26T16:48:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.