16 papers
SpecPath: Testing Coding Agents Across Contract-Equivalent Specification Histories
Yangfan Wu, Haozhe Wang, Huanyu Yang +2
Modern coding agents increasingly appear capable of following complex software requirements, yet their success leaves a critical ambiguity: do they resolve the active specification…
Search Beyond What Can Be Taught: Evolving the Knowledge Boundary in Agentic Visual Generation
Haozhe Wang, Weijia Feng, Jinpeng Yu +8
Visual generators excel at rendering, but they confidently fabricate what they do not know. User requests are unbounded, evolving, and deeply long-tailed: new characters, trending…
Bad Seeing or Bad Thinking? Rewarding Perception for Multimodal Reasoning
Haozhe Wang, Qixin Xu, Changpeng Wang +4
Achieving robust perception-reasoning synergy is a central goal for advanced Vision-Language Models (VLMs). Recent advancements have pursued this goal via architectural designs or…
RationalRewards: Reasoning Rewards Scale Visual Generation Both Training and Test Time
Haozhe Wang, Cong Wei, Weiming Ren +3
Most reward models for visual generation reduce rich human judgments to a single unexplained score, discarding the reasoning that underlies preference. We show that teaching reward…
The Combined Problem of Online Task Assignment and Lifelong Path Finding in Logistics Warehouses: Rule-Based Systems Matter
Fengming Zhu, Weijia Xu, Yifei Guo +1
We study the combined problem of online task assignment and lifelong path finding, which is crucial for the logistics industries. However, most literature either (1) focuses on lif…
CogDoc: Towards Unified thinking in Documents
Qixin Xu, Haozhe Wang, Che Liu +2
Current document reasoning paradigms are constrained by a fundamental trade-off between scalability (processing long-context documents) and fidelity (capturing fine-grained, multim…