8 papers
Multi-Turn Reflective Masking Elicits Reasoning in Mask Diffusion Models
Yanming Zhang, Yihan Bian, Jingyuan Qi +3
While reasoning on autoregressive (AR) models is often performed by chain-of-thought reasoning and reflection, their refinement of previous outputs still relies on fully sequential…
GRAPE: Guided Parameter-Space Evolution for Compact Adversarial Robustness
Zhiyuan Ye, Xiangyu Zhou, Ji Qi +2
Adversarial Training (AT) improves neural network robustness, but most methods train a fixed parameter space from the start. This paper asks whether the order in which parameters b…
EigenData: A Self-Evolving Multi-Agent Platform for Function-Calling Data Synthesis, Auditing, and Repair
Jiaao Chen, Jingyuan Qi, Mingye Gao +3
Function-calling agents -- large language models that invoke tools and APIs -- require high-quality, domain-specific training data spanning executable environments, backing databas…
From Solving to Verifying: A Unified Objective for Robust Reasoning in LLMs
Xiaoxuan Wang, Bo Liu, Song Jiang +4
The reasoning capabilities of large language models (LLMs) have been significantly improved through reinforcement learning (RL). Nevertheless, LLMs still struggle to consistently v…
Data Value in the Age of Scaling: Understanding LLM Scaling Dynamics Under Real-Synthetic Data Mixtures
Haohui Wang, Jingyuan Qi, Jianpeng Chen +9
The rapid progress of large language models (LLMs) is fueled by the growing reliance on datasets that blend real and synthetic data. While synthetic data offers scalability and cos…
AR-RAG: Autoregressive Retrieval Augmentation for Image Generation
Jingyuan Qi, Zhiyang Xu, Qifan Wang +1
We introduce Autoregressive Retrieval Augmentation (AR-RAG), a novel paradigm that enhances image generation by autoregressively incorporating knearest neighbor retrievals at the p…