TeLLMe: An Energy-Efficient Ternary LLM Accelerator for Prefilling and Decoding on Edge FPGAs

Documenti analoghi

• TeLLMe v2: An Efficient End-to-End Ternary LLM Prefill and Decode Accelerator with Table-Lookup Matmul on Edge FPGAs
  di: Qiao, Ye, et al.
  Pubblicazione: (2025)
• PD-Swap: Prefill-Decode Logic Swapping for End-to-End LLM Inference on Edge FPGAs via Dynamic Partial Reconfiguration
  di: Zhang, Yifan, et al.
  Pubblicazione: (2025)
• COBRA: Algorithm-Architecture Co-optimized Binary Transformer Accelerator for Edge Inference
  di: Qiao, Ye, et al.
  Pubblicazione: (2025)
• FASQ: Flexible Accelerated Subspace Quantization for Calibration-Free LLM Compression
  di: Qiao, Ye, et al.
  Pubblicazione: (2026)
• EVA: Accelerating LLM Decoding via an Efficient Vector Quantization Architecture
  di: Duan, Bowen, et al.
  Pubblicazione: (2026)