NeurIPS Paper Maps Where Temporal Context Belongs in Low-Resource Video Adaptation
HildeKuehne · x · 2026-11-03
A NeurIPS 2026 paper by Luc Sträter and Hazel Doughty systematically studies adaptation strategies for video foundation models under limited labeled data: should temporal context live in the backbone, PEFT adapters, or a probing head?
The study spans appearance-focused, motion-focused, and spatially dense tasks, comparing four allocation strategies (static backbone with temporal adaptation, temporal backbone with static adaptation, matched temporal context, and distributed temporal context). Key findings: video PEFT has mostly targeted image-pretrained models even though standard PEFT works on video representations too, and how temporal context is distributed across backbone, PEFT, and probe matters significantly for adaptation quality.
More from Research
- Can't reproduce a published baseline with its own code: how should a paper report it? — InsectActive95 · 2026-11-03
- Why model self-editing matters: zero capability gap between interpreter and target — menhguin · 2026-11-03
- "Don't overupdate": author praises honest failure reporting in interpretability post — 1a3orn · 2026-11-03
- LLMs can edit out flawed reasoning but don't improve it, even on tiny models — 1a3orn · 2026-11-03
- AI Agent Performs Brain Drug Injection on a Live Mouse — ChuckDBrooks · 2026-11-03
- Economics Paper Thanks Claude Opus 5.5, GPT-6 Astra and Gemini Argon 4 — soumitrashukla9 · 2026-11-03