Inference with large language models (LLMs) on long sequences is computationally expensive due to the quadratic complexity of self-attention. Distributed blockwise methods such as Star Attention reduce this cost by sharding context across hosts, but rely on prepending a static, content-blind copy of the first block to every host.
We propose Pulsar Attention, which replaces the static anchor with two lightweight, content-aware components: a small attention-sink prefix that stabilizes softmax, and compact cross-block summaries built via a Max-IDF heuristic that selects chunks containing globally rare tokens.
This reduces the Phase 1 per-GPU FLOPs by up to 3.3$\times$ over Star Attention while retaining an identical KV cache footprint. On RULER and BABILong with Llama-3.1-8B, Pulsar Attention outperforms both Star Attention and dense attention at sequence lengths up to 128K tokens, with absolute gains of up to 4.7% over the dense baseline.
Blogger's Review: Pulsar Attention significantly optimizes the efficiency of long sequence inference by introducing content-aware mechanisms, showcasing the potential of processing large-scale data in distributed environments. Its innovative Max-IDF strategy is noteworthy and provides new insights for future model designs.