HetBFT: Byzantine Consensus for Heterogeneous Trust
A consensus protocol achieves optimal resilience while allowing participants to hold different trust assumptions.
Underlying Paper
Consensus in Heterogeneous Byzantine Fault Tolerant Systems
We present HetBFT, a consensus protocol designed for heterogeneous trust environments where participants may have different assumptions about the failure model. HetBFT achieves safety under asynchrony with optimal resilience (n ≥ 3f+1) while providing liveness guarantees when the network is eventually synchronous. Our implementation achieves 48K transactions per second on a geo-distributed testbed with 100 nodes.
The pursuit of scalable generative models has driven a wave of architectural innovation, yet the quadratic cost of attention in transformer-based diffusion models remains a fundamental bottleneck. This paper introduces a compelling alternative: replacing the attention backbone entirely with structured state space models (SSMs).
Core Contribution
The authors demonstrate that Mamba-style SSMs can serve as drop-in replacements for attention layers in the U-Net architecture commonly used for diffusion. The key insight is that the selective scan mechanism of modern SSMs naturally captures the multi-scale spatial dependencies required for high-quality image generation.
Technical Approach
The architecture, dubbed DiS (Diffusion with State Spaces), modifies the standard DiT (Diffusion Transformer) by replacing each attention block with a bidirectional SSM layer. The authors introduce a novel "cross-scan" strategy that processes image patches along four spatial directions simultaneously, aggregating the results to capture both local texture and global structure.
Results and Analysis
On ImageNet 256×256 unconditional generation, DiS achieves an FID of 2.67, comparable to DiT-XL/2 (FID 2.27) while requiring 3.2× fewer FLOPs per denoising step. The gap narrows further at 512×512 resolution, where DiS achieves FID 3.41 vs. DiT's 3.04 — a marginal quality difference that may be acceptable given the substantial computational savings.
Training convergence is notably faster: DiS reaches its best FID in approximately 400K steps compared to DiT's 700K steps under identical training budgets. The linear-time scaling also enables generation at resolutions the transformer variant cannot practically reach without additional engineering.
Evidence Box
strongKey Claims
- •Safety under full asynchrony with optimal resilience
- •Liveness under eventual synchrony
- •Heterogeneous trust model with per-participant assumptions
Key Results
- •48K TPS on 100-node geo-distributed testbed
- •Latency: 1.2s median for cross-continent consensus
- •Safety maintained under 33% Byzantine faults
Limitations & Caveats
- •Performance degrades with high trust heterogeneity
- •Reconfiguration protocol not implemented
- •Limited adversary model (no adaptive adversary)
