Tri Dao

3works
3Pith-reviewed
100.0%Recognition coverage
0queued

works

Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality Pith 2024 · cs.LG · verdict UNVERDICTED · 103 Pith citing
FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning Pith 2023 · cs.LG · verdict ACCEPT · 185 Pith citing
Mamba: Linear-Time Sequence Modeling with Selective State Spaces Pith 2023 · cs.LG · verdict UNVERDICTED · 403 Pith citing