DiscussLLM introduces a two-stage synthetic data pipeline to annotate multi-turn discussions with five intervention types and trains LLMs to time contributions via a silent token or proactive responses.
Videollm-online: Online video large language model for streaming video
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
dataset 1
citation-polarity summary
years
2025 2verdicts
UNVERDICTED 2roles
dataset 1polarities
use dataset 1representative citing papers
VideoLLaMA3 uses a vision-centric training paradigm and token-reduction design to reach competitive results on image and video benchmarks.
citing papers explorer
-
DiscussLLM: Teaching Large Language Models When to Speak
DiscussLLM introduces a two-stage synthetic data pipeline to annotate multi-turn discussions with five intervention types and trains LLMs to time contributions via a silent token or proactive responses.
-
VideoLLaMA 3: Frontier Multimodal Foundation Models for Image and Video Understanding
VideoLLaMA3 uses a vision-centric training paradigm and token-reduction design to reach competitive results on image and video benchmarks.