activity
20242026
collaborators

10 papers

cs.LG2026

A Task-Centric Theory for Iterative Self-Improvement with Easy-to-Hard Curricula

Chenruo Liu, Yijun Dong, Yiqiu Shen +1

Iterative self-improvement fine-tunes an autoregressive large language model (LLM) on reward-verified outputs generated by the LLM itself. In contrast to the empirical success of s…

cs.LG2026

Discrepancies are Virtue: Weak-to-Strong Generalization through Lens of Intrinsic Dimension

Yijun Dong, Yicheng Li, Yunai Li +2

Weak-to-strong (W2S) generalization is a type of finetuning (FT) where a strong (large) student model is trained on pseudo-labels generated by a weak teacher. Surprisingly, W2S FT…

math.NA2026

Attention Mechanisms Through the Lens of Numerical Methods: Approximation Methods and Alternative Formulations

Michel Fabrice Serret, Alice Cortinovis, Yijun Dong +10

The attention mechanism is the computational core of modern Transformer architectures, but its quadratic complexity in the input sequence length is the bottleneck for large-scale i…

cs.LG2026

Does Weak-to-strong Generalization Happen under Spurious Correlations?

Chenruo Liu, Yijun Dong, Qi Lei

We initiate a unified theoretical and algorithmic study of a key problem in weak-to-strong (W2S) generalization: when fine-tuning a strong pre-trained student with pseudolabels fro…

cs.LG2026

When does Chain-of-Thought Help: A Markovian Perspective

Zihan Wang, Yijun Dong, Qi Lei

Chain-of-Thought (CoT) prompting is a widely used inference-time technique for improving reasoning, yet its gains are uneven across tasks. We analyze when and why CoT helps by mode…

math.NA2025

Randomized time stepping of nonlinearly parametrized solutions of evolution problems

Yijun Dong, Paul Schwerdtner, Benjamin Peherstorfer

The Dirac-Frenkel variational principle is a widely used building block for using nonlinear parametrizations in the context of model reduction and numerically solving partial diffe…