Jonah Bernard

Quantization @ AMD | CS @ Cornell

[Jonah Bernard @jonahbernard_]

Great work! Simple and effective improvement to DFlash

[Liran Ringel @liranringel]

Introducing DDTree: accelerates speculative decoding by drafting a tree with one block diffusion pass, then verifying multiple likely continuations together.

Paper: https://t.co/cgYBw70O5i
Project page: https://t.co/ygFukxrZLB
Code: https://t.co/2z7U00NsuH

[Jonah Bernard @jonahbernard_]

Just finished reading all 4 papers. Great work!

[Enze Xie @xieenze_jr]

Over the past year, we have been pushing diffusion LLMs from an interesting generation paradigm toward practical systems.

1️⃣Fast-dLLM: training-free acceleration with approximate KV cache + confidence-aware parallel decoding.
2️⃣Fast-dLLM v2: block diffusion LLMs, making diffusion more AR-like and cache-friendly.
3️⃣Fast-dVLM: extending block diffusion to vision-language reasoning.
4️⃣Fast-dDrive: bringing structured block diffusion to autonomous-driving VLAs.

From faster inference → block diffusion → multimodal understanding → real-world driving agents.

code: https://t.co/2RM12zKAxd
Fast-dLLM: https://t.co/lZKC3CPJjU
Fast-dLLM v2: https://t.co/yr1cKg9PVj
Fast-dVLM: https://t.co/2aTbG1m4m3
Fast-dDrive: https://t.co/ED56TqrmGq

[Sunday @sundayrobotics]

Sound on 🔔