Search papers, labs, and topics across Lattice.
This paper introduces HydraHead, a novel architecture that hybridizes Linear Attention (LA) and Full Attention (FA) at the head level to address the quadratic complexity of attention mechanisms in long-context processing. By employing an interpretability-driven selection strategy for retrieval-critical heads and a scale-normalized fusion module, HydraHead achieves substantial performance improvements with minimal training overhead. The results demonstrate that HydraHead outperforms existing hybrid models in long-context tasks, achieving over 69% improvement over baseline models while maintaining strong general reasoning capabilities.
HydraHead achieves a remarkable 69% performance boost in long-context tasks by intelligently hybridizing attention mechanisms at the head level, challenging traditional layer-wise approaches.
The quadratic complexity of attention poses a critical bottleneck for long-context processing, spurring interest in hybrid attention designs. Most open-source hybrid models adopt a layer-wise strategy. Yet, prior work has noted the inherent difficulty of integrating Linear Attention (LA) with Full Attention (FA), suggesting that the design space of attention hybridization remains underexplored. To probe this space, we conduct interpretability analysis and observe that layers exhibit block-wise functional similarity, while individual heads within the same layer display distinct functional specialization despite sharing input features. This head-level heterogeneity suggests that the head dimension provides a natural and principled granularity for fusing heterogeneous attention signals. Building on this insight, we introduce HydraHead, a novel architecture that hybridizes FA and LA along the head axis. HydraHead features two key innovations: (1) an interpretability-driven selection strategy that identifies retrieval-critical heads and preserves FA only for them, and (2) a scale-normalized fusion module that reconciles the distributional gap between FA and LA head outputs. By leveraging a three-stage transfer pipeline with parameter reuse and distillation, we achieve high-performance hybrid models with minimal training overhead. Under a unified training setup, HydraHead outperforms other hybrid designs in long-context tasks while maintaining strong general reasoning. With interpretability-driven head selection, it matches a 3:1 layer-wise hybrid's long-context performance at a 7:1 LA-to-FA ratio. Crucially, trained on only 15B tokens, HydraHead achieves over 69% improvement over the baseline at 512K context length, approaching Qwen3.5, a leading model of comparable size with a native context length of 256K. This highlights the significant scaling potential of head-level hybridization.
