8 papers
Blame Errant Words, Not the Canvas: Defect-Aware Spatial Credit Assignment for Visual Text Generation
Sujie Hu, Huaisong Zhang, Pujing Yang +7
Recent reinforcement learning approaches have improved visual text generation, yet generated images still suffer from incorrect wording and malformed glyphs. We identify two couple…
Where, What, Why, and Importance: Structured Defect Grounding for Text-to-Image Feedback
Huaisong Zhang, Hao Yu, Yuxuan Zhang +7
Despite generating increasingly photorealistic images, text-to-image (T2I) models still exhibit localized, subtle, and structurally complex failures. Diagnosing these failures requ…
RewardHarness: Self-Evolving Agentic Post-Training
Yuxuan Zhang, Penghui Du, Bo Li +11
Aligning generative models with human preferences remains a central challenge. While RLHF is widely adopted, it relies on large-scale preference data. This limitation is particular…
Watch Before You Answer: Learning from Visually Grounded Post-Training
Yuxuan Zhang, EunJeong Hwang, Huaisong Zhang +8
It is critical for vision-language models (VLMs) to comprehensively understand visual, temporal, and textual cues. However, despite rapid progress in multimodal modeling, video und…
ClawBench: Can AI Agents Complete Everyday Online Tasks?
Yuxuan Zhang, Yubo Wang, Yipeng Zhu +25
AI agents may be able to assist with emails and documents, but can they reliably complete everyday online workflows on real websites? Everyday online tasks offer a realistic yet un…
OmniAlpha: Aligning Transparency-Aware Generation via Multi-Task Unified Reinforcement Learning
Hao Yu, Jinglin Wang, Jiabo Zhan +7
Transparency-aware generation requires modeling not only RGB appearance but also alpha-based opacity and cross-layer composition, which are essential for tasks such as image mattin…
UltraHorizon: Benchmarking Agent Capabilities in Ultra Long-Horizon Scenarios
Haotian Luo, Huaisong Zhang, Xuelin Zhang +15
Autonomous agents have recently achieved remarkable progress across diverse domains, yet most evaluations focus on short-horizon, fully observable tasks. In contrast, many critical…
ComRoPE: Scalable and Robust Rotary Position Embedding Parameterized by Trainable Commuting Angle Matrices
Hao Yu, Tangyu Jiang, Shuning Jia +7
The Transformer architecture has revolutionized various regions since it was proposed, and its effectiveness largely depends on the ability to encode positional information. Tradit…