An RL framework uses digital twin representations with hierarchical uncertainty estimates and a novel clinical plausibility reward to train LLMs for surgical VideoQA, achieving SOTA on a new 2000-pair benchmark and two existing datasets.
Smile: A composite lexical-semantic metric for question- answering evaluation.arXiv preprint arXiv:2511.17432, 2025
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Training LLMs with Reinforcement Learning over Digital Twin Representations for Reasoning-Intensive Surgical VideoQA
An RL framework uses digital twin representations with hierarchical uncertainty estimates and a novel clinical plausibility reward to train LLMs for surgical VideoQA, achieving SOTA on a new 2000-pair benchmark and two existing datasets.