Revisiting Classics: Progress is Cumulative, Not Revolutionary
DJiafei · x · 2026-08-23
While teaching a new class, the author revisited many classical papers and found a recurring lesson: progress is often cumulative, not revolutionary.
Enduring ideas, from optimal grasp planning to Dex-Net, Contact-GraspNet, and GraspGen, are carried forward by improvements in models, data, compute, and tools.
Related event: Rereading Classic Papers: Progress Is Cumulative, Not Revolutionary(3 posts)→
More from Research
- Berkeley paper uses the Midas myth to rethink robotics beyond human imitation — begusgasper · 2026-10-16
- 31 Tech Curves Visualized: Kurzweil's $1,000-Brain-Compute Call Missed by 400x — cpaik · 2026-10-16
- DeepMind Paper: First-Price Auctions in Agentic Markets Reach 1-1/e^2 Efficiency — dair_ai · 2026-10-16
- View: induction heads fully explain why transformers won and why RL is the right algorithm — akbirthko · 2026-10-16
- SpliceAI2 benchmarks show major gains across nine splicing-variant evaluations — anshulkundaje · 2026-10-16
- GRPO post-training collapses model distributions, clashing with long-horizon agents — zsakib_ · 2026-10-16