Researcher calls ITTO point-tracking benchmark misleading for skipping online methods like Track-On
ftm_guney · x · 2026-10-14
The author argues the NeurIPS ITTO benchmark (dynamic point tracking protocol) claims real-world evaluation but never tests online capabilities, and Track-On isn't even in its SOTA table. Results show all models drop sharply vs TAP-Vid, degrading further at high track motion—so he calls the title misleading.
More from Research
- Beyond automation: Stanford economists sketch a white-box, human-AI collaboration research agenda — AI Snake Oil · 2026-10-15
- AI-in-biology startup hiring MTS to build rigorous benchmarks for AI in biology — kenbwork · 2026-10-15
- SlipSense: multimodal tactile slip detection hits 96.7% F1 at 23.1ms with zero-shot robot transfer — chris_j_paxton · 2026-10-15
- 2021 slides reveal Garçon, Anthropic's internal interpretability infra for multi-GPU models — voooooogel · 2026-10-15
- Microduck robot runs onboard SLAM with just 64 pixels of depth, 4.8 cm median error — chris_j_paxton · 2026-10-15
- UIUC paper explains sparse LLM post-training updates through the lens of Fisher information — tokenbender · 2026-10-15