11 citations · 14 across the 7 of their papers we have counts for
7 papers
MiroEval: Benchmarking Multimodal Deep Research Agents in Process and Outcome
Fangda Ye, Yuxin Hu, Pengxiang Zhu +19
Recent progress in deep research systems has been impressive, but evaluation still lags behind real user needs. Existing benchmarks predominantly assess final reports using fixed r…
MathGen: Revealing the Illusion of Mathematical Competence through Text-to-Image Generation
Ruiyao Liu, Hui Shen, Ping Zhang +16
Modern generative models have demonstrated the ability to solve challenging mathematical problems. In many real-world settings, however, mathematical solutions must be expressed vi…
Beyond Bradley-Terry Models: A General Preference Model for Language Model Alignment
Yifan Zhang, Ge Zhang, Yue Wu +2
Modeling human preferences is crucial for aligning foundation models with human values. Traditional reward modeling methods, such as the Bradley-Terry (BT) reward model, fall short…
Autonomous Data Selection with Zero-shot Generative Classifiers for Mathematical Texts
Yifan Zhang, Yifan Luo, Yang Yuan +1
We present Autonomous Data Selection (AutoDS), a method that leverages base language models themselves as zero-shot "generative classifiers" to automatically curate high-quality ma…
Augmenting Math Word Problems via Iterative Question Composing
Haoxiong Liu, Yifan Zhang, Yifan Luo +1
Despite the advancements in large language models (LLMs) for mathematical reasoning, solving competition-level math problems remains a significant challenge, especially for open-so…
Cumulative Reasoning with Large Language Models
Yifan Zhang, Jingqin Yang, Yang Yuan +1
Recent advancements in large language models (LLMs) have shown remarkable progress, yet their ability to solve complex problems remains limited. In this work, we introduce Cumulati…
Trade-off Between Efficiency and Consistency for Removal-based Explanations
Yifan Zhang, Haowei He, Zhiquan Tan +1
In the current landscape of explanation methodologies, most predominant approaches, such as SHAP and LIME, employ removal-based techniques to evaluate the impact of individual feat…