9 papers
The Depth Ceiling: On the Limits of Large Language Models in Discovering Latent Planning
Yi Xu, Philipp Jettkant, Laura Ruis
The viability of chain-of-thought (CoT) monitoring hinges on models being unable to reason effectively in their latent representations. Yet little is known about the limits of such…
Pretraining with Token-Level Adaptive Latent Chain-of-Thought
Boyi Zeng, Yiqin Hao, He Li +8
Scaling large language models by increasing parameters and training data is increasingly constrained by limited high-quality corpora and rising communication costs. This work explo…
Controlled Self-Evolution for Algorithmic Code Optimization
Tu Hu, Ronghao Chen, Shuo Zhang +9
Self-evolution methods enhance code generation through iterative "generate-verify-refine" cycles, yet existing approaches suffer from low exploration efficiency, failing to discove…
RAVR: Reference-Answer-guided Variational Reasoning for Large Language Models
Tianqianjin Lin, Xi Zhao, Xingyao Zhang +5
Reinforcement learning (RL) can refine the reasoning abilities of large language models (LLMs), but critically depends on a key prerequisite: the LLM can already generate high-util…
Structuring GUI Elements through Vision Language Models: Towards Action Space Generation
Yi Xu, Yesheng Zhang, Jiajia Liu +1
Multimodal large language models (MLLMs) have emerged as pivotal tools in enhancing human-computer interaction. In this paper we focus on the application of MLLMs in the field of g…
Many-Turn Jailbreaking
Xianjun Yang, Liqiang Xiao, Shiyang Li +5
Current jailbreaking work on large language models (LLMs) aims to elicit unsafe outputs from given prompts. However, it only focuses on single-turn jailbreaking targeting one speci…