影响力指数
论文质量、代表作、近期表现、广度与样本量置信度综合计算
-/100
发表论文3 篇
平均评分
年均产出3.0 篇/年
ZHANG Xinshen
研究方向
multimodal reasoning · visual question answering · panoramic image understanding · egocentric vision · object grounding · interaction reasoning · computer vision · spatial reasoning · 360-degree scenes
7
Audio-FLAN: An Instruction-Following Dataset for Unified Understanding and Generation of Speech, Music, and Sound
ICLR 2026Rejected
7
Alignment Does Matter: Enables Pure-Speech-Token Dialogue with Frozen Text LLMs
ICLR 2026Rejected
5
Towards Omnidirectional Reasoning: A Dataset, Benchmark, and GRPO-based Method
ICLR 2026Desk Rejected
一作