11 papers
Context Distillation as Latent Memory Management
Ziyang Zheng, Zeju Li, Xiangyu Wen +5
Context distillation compresses contextual information into model parameters, yet existing methods often ignore how multiple distilled latent memories should be stored, retrieved,…
FAME: Forecasting Academic Impact via Continuous-Time Manifold Evolution
Jianrong Ding, Jianyuan Zhong, Zhengyan Shi +1
Large Language Models (LLMs) are increasingly used to brainstorm and evaluate research ideas, yet assessing such judgments is fundamentally difficult because the true impact of a n…
Stabilizing Reinforcement Learning for Diffusion Language Models
Jianyuan Zhong, Kaibo Wang, Ding Ding +5
Group Relative Policy Optimization (GRPO) is highly effective for post-training autoregressive (AR) language models, yet its direct application to diffusion large language models (…
Making Slow Thinking Faster: Compressing LLM Chain-of-Thought via Step Entropy
Zeju Li, Jianyuan Zhong, Ziyang Zheng +5
Large Language Models (LLMs) using Chain-of-Thought (CoT) prompting excel at complex reasoning but generate verbose thought processes with considerable redundancy, leading to incre…
From Craft to Constitution: A Governance-First Paradigm for Principled Agent Engineering
Qiang Xu, Xiangyu Wen, Changran Xu +2
The advent of powerful Large Language Models (LLMs) has ushered in an ``Age of the Agent,'' enabling autonomous systems to tackle complex goals. However, the transition from protot…
Reasoning Scaffolding: Distilling the Flow of Thought from LLMs
Xiangyu Wen, Junhua Huang, Zeju Li +6
The prevailing approach to distilling reasoning from Large Language Models (LLMs)-behavioral cloning from textual rationales-is fundamentally limited. It teaches Small Language Mod…