10 papers
MAGIC: A Co-Evolving Attacker-Defender Adversarial Game for Robust LLM Safety
Xiaoyu Wen, Zhida He, Han Qi +7
Ensuring robust safety alignment is crucial for Large Language Models (LLMs), yet existing defenses often lag behind evolving adversarial attacks due to their \textbf{reliance on s…
-Reader: Dual Evolving Graphs for Multimodal Document QA
Yaxin Du, Junru Song, Yifan Zhou +8
Retrieval-augmented generation is a practical paradigm for question answering over long documents, but it remains brittle for multimodal reading where text, tables, and figures are…
KaLM: Knowledge-aligned Autoregressive Language Modeling via Dual-view Knowledge Graph Contrastive Learning
Peng Yu, Cheng Deng, Beiya Dai +2
Autoregressive large language models (LLMs) pre-trained by next token prediction are inherently proficient in generative tasks. However, their performance on knowledge-driven tasks…
ML-Master: Towards AI-for-AI via Integration of Exploration and Reasoning
Zexi Liu, Yuzhu Cai, Xinyu Zhu +6
As AI capabilities advance toward and potentially beyond human-level performance, a natural transition emerges where AI-driven development becomes more efficient than human-centric…
Natural Language Reinforcement Learning
Xidong Feng, Bo Liu, Yan Song +7
Artificial intelligence progresses towards the "Era of Experience," where agents are expected to learn from continuous, grounded interaction. We argue that traditional Reinforcemen…
ReMA: Learning to Meta-think for LLMs with Multi-Agent Reinforcement Learning
Ziyu Wan, Yunxiang Li, Xiaoyu Wen +8
Recent research on Reasoning of Large Language Models (LLMs) has sought to further enhance their performance by integrating meta-thinking -- enabling models to monitor, evaluate, a…