7 papers
Rewarding Novel Deductions: Solver-guided Process Supervision for Logical Reasoning
Muhammad Asif Ali, Wenqing Wang, Huan Wang +1
Logical reasoning remains a major challenge for large language models (LLMs), particularly on structured problems that require precise constraint tracking, consistency preservation…
SCOUT: Active Information Foraging for Long-Text Understanding with Decoupled Epistemic States
Zhenliang Zhang, Wenqing Wang, Yong Hu +4
Long-Text Understanding (LTU) at million-token scale requires balancing reasoning fidelity with computational efficiency. Frontier long-context LLMs can process millions of token c…
SagaScale: A Realistic, Scalable, and High-Quality Long-Context Benchmark Built from Full-Length Novels
Guancheng Du, Yong Hu, Wenqing Wang +2
Large Language Models (LLMs) have shown significant progress, but understanding long and complex documents remains challenging. Many long-context benchmarks have been proposed, but…
BeautyGuard: Designing a Multi-Agent Roundtable System for Proactive Beauty Tech Compliance through Stakeholder Collaboration
Junwei Li, Wenqing Wang, Huiliu Mao +2
As generative AI enters enterprise workflows, ensuring compliance with legal, ethical, and reputational standards becomes a pressing challenge. In beauty tech, where biometric and…
A-IPO: Adaptive Intent-driven Preference Optimization
Wenqing Wang, Muhammad Asif Ali, Ali Shoker +4
Human preferences are diverse and dynamic, shaped by regional, cultural, and social factors. Existing alignment methods like Direct Preference Optimization (DPO) and its variants o…
Dissecting Embodied Abilities in Multimodal Language Models through Skill-level Evaluation and Diagnosis
Yu Qi, Haibo Zhao, Ziyu Guo +17
Understanding the capability bottlenecks of embodied multimodal large language models (MLLMs) is crucial for improving embodied agents. However, existing embodied benchmarks mainly…
Defining and Detecting Vulnerability in Human Evaluation Guidelines: A Preliminary Study Towards Reliable NLG Evaluation
Jie Ruan, Wenqing Wang, Xiaojun Wan
Human evaluation serves as the gold standard for assessing the quality of Natural Language Generation (NLG) systems. Nevertheless, the evaluation guideline, as a pivotal element en…