Get started
Author

Penghui Du

ClawBench: Can AI Agents Complete Everyday Online Tasks?
2026 · Yuxuan Zhang, Yubo Wang, et al. · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · 46
PDF
Re-Align: Structured Reasoning-guided Alignment for In-Context Image Generation and Editing
2026 · He, Runze, Cheng, Yiji, et al. · arXiv (Cornell University) · 8
PDF
RewardHarness: Self-Evolving Agentic Post-Training
2026 · Zhang, Yuxuan, Wei, Cong, et al. · arXiv (Cornell University) · 6
PDF
VGI-Bench: Probing Visual Intelligence in Video Generation Models
2026 · Xuan He, Cong Wei, et al. · Conference on Empirical Methods in Natural Language Processing (EMNLP) · 5
PDF
Watch Before You Answer: Learning from Visually Grounded Post-Training
2026 · Zhang, Yuxuan, Hwang, EunJeong, et al. · arXiv (Cornell University) · 4
PDF
MedClaw: Heuristic Agent Harness for Long-Horizon Surgical Video Reasoning
2026 · Fan, Yingying, Du, Penghui, et al. · arXiv (Cornell University) · 2
PDF
Group Relative Attention Guidance for Image Editing
2026 · Xuanpu Zhang, Xuesong Niu, et al. · 1
LongBanana: An Expert-Verified Benchmark for Long-Context Multi-Reference Image Synthesis
2026 · Cao, Haoxiang, Zhang, Yuxuan, et al. · arXiv (Cornell University)
PDF
Calibrating Attention-Based Token Pruning for Efficient Multi-Image Synthesis in UMMs
2026 · Wu, Zimeng, Tian, Qijian, et al. · arXiv (Cornell University)
PDF
Works count: 9
Citation count: 72
H-index: 4
I10-index: 1
Year
Topic
Type