GCN author Thomas Kipf: DETR/Slot Attention-style architecture can do it without diffusion
tkipf · x · 2026-09-16
Thomas Kipf, author of the seminal GCN paper, offered a terse technical take in a discussion: in principle you can achieve this with a DETR-/Slot Attention-style architecture and just a softmax cross entropy loss — no diffusion needed. His core argument: queries/slot inits define types, and softmax can simply be masked according to those types. A short but authoritative comment on architecture choice.
More from Research
- Sora's Diffusion Transformer explained by hand in 14 steps: prompts enter only as scale and shift — ProfTomYeh · 2026-09-16
- Researcher: any agent acting over long horizons provably has a self-model and world model — chris_j_paxton · 2026-09-16
- Ben Antieau guest post on Terence Tao's blog: mathematics needs both 'fast math' and 'slow math' in the LLM era — littmath · 2026-09-16
- VisTW: a Traditional Chinese VLM benchmark for reading Taiwan — and an eval framework that caught a 36-point bug — piske_usagi · 2026-09-16
- Mind2Dialogue Simulates User Mental States to Train Human-Aware LLMs — Zixuan Wang · 2026-09-16
- NUS Survey Maps Six Roles for Foundation Models Across the Game Lifecycle — NationalUniversityofSingapore · 2026-09-16