4 papers
ConPress: Learning Efficient Reasoning from Multi-Question Contextual Pressure
Jie Deng, Shining Liang, Jun Li +2
Large reasoning models (LRMs) typically solve reasoning-intensive tasks by generating long chain-of-thought (CoT) traces, leading to substantial inference overhead. We identify a r…
PIKA: Expert-Level Synthetic Datasets for Post-Training Alignment from Scratch
Shangjian Yin, Shining Liang, Wenbiao Ding +4
High-quality instruction data is critical for LLM alignment, yet existing open-source datasets often lack efficiency, requiring hundreds of thousands of examples to approach propri…
Lost in Stories: Consistency Bugs in Long Story Generation by LLMs
Junjie Li, Xinrui Guo, Yuhao Wu +3
What happens when a storyteller forgets its own story? Large Language Models (LLMs) can now generate narratives spanning tens of thousands of words, but they often fail to maintain…
Beyond Rejection Sampling: Trajectory Fusion for Scaling Mathematical Reasoning
Jie Deng, Hanshuang Tong, Jun Li +4
Large language models (LLMs) have made impressive strides in mathematical reasoning, often fine-tuned using rejection sampling that retains only correct reasoning trajectories. Whi…