UCL's TAL lets LLM agents learn from tutorial videos with no parameter updates, +113% on VidLearn
UniversityCollegeLondon · hf · 2026-11-03
UCL researchers introduce Tutorial Agentic Learning (TAL), a nonparametric continual-learning paradigm letting LLM agents keep learning from instructional videos without parameter updates. TAL consolidates videos into a structured tutorialbook memory via a two-stage adversarial process that harvests hard video-grounded questions, and retrieves at inference through a Table-of-Contents tree tied to model reasoning rather than embedding similarity. Alongside, they release VidLearn, a multimodal QA benchmark from 72 chess tutorial videos spanning 12 openings. TAL beats all baselines, improves monotonically with more videos, and gains up to 113%/91% relative over vanilla MLLMs on train/test splits, reaching a stronger opening position in 69% of head-to-head games.
More from Research
- They reverse-engineered a $1B-funded megakernel demo from video frames and open-sourced it — erikphoel · 2026-11-03
- NVIDIA, Tsinghua, CMU, Berkeley and Princeton launch StreamEval, first real-time interactive streaming video benchmark — JaynitMakwana · 2026-11-03
- MIT: AI edits 0.016% of another model's weights, lifting MATH-500 from 7.9% to 48.1% — rohanpaul_ai · 2026-11-03
- Schmidhuber: AI weight editing is 'old hat' from his 1990s fast weight programmers — SchmidhuberAI · 2026-11-03
- SIMIT: vision-language models self-improve at test time by imagining their own practice data — dtudk · 2026-11-03
- Face quality scores can backfire: higher quality often signals a morphing attack — FraunhoferIGD · 2026-11-03