2 papers
cs.CV2026
PixelBench: Can LLMs Understand and Generate Images Pixel by Pixel?
Haoquan Zhang, Feiyang Liu, Juanxi Tian +3
Large language models (LLMs) can generate visual assets through text-based formats such as SVG and CAD, but these symbolic representations describe images using high-level primitiv…
cs.CL2026
ConversationalVoice: Full-Duplex Speech Data from Real Conversations through Source-Faithful Reconstruction and Conversation-Grounded Expansion
Richard Yucheng He, Baodong Cao, Chen Xu +2
Full-duplex speech models require training data that preserves turn-taking, overlap, interruption, and backchannel behavior, yet these signals are entangled across speakers in nois…