activity
20242026
most citedGeoMix: Towards Geometry-Aware Data Augmentation

5 citations · 5 across the 10 of their papers we have counts for

collaborators
Showing cs.LGShow all

7 papers · 1 filter

cs.LG2026

Tight Sample Complexity of Transformers

Chenxiao Yang, Nathan Srebro, Zhiyuan Li

We tightly characterize the VC dimension of depth- Transformers with a total of parameters, mapping an input sequence of length to a single output, establishing an upper…

cs.LG2026

Hierarchical Domain Generalization

Chenxiao Yang, Zhiyuan Li, Shai Ben-David +1

We study hierarchical domain generalization as a problem of extrapolation from finite observed regions to an entire instance space, replacing i.i.d. sampling with arbitrary domain…

cs.LG2026

Recursive Models for Long-Horizon Reasoning

Chenxiao Yang, Nathan Srebro, Zhiyuan Li

Modern language models reason within bounded context, an inherent constraint that poses a fundamental barrier to long-horizon reasoning. We identify recursion as a core principle f…

cs.LG2025

On Powerful Ways to Generate: Autoregression, Diffusion, and Beyond

Chenxiao Yang, Cai Zhou, David Wipf +1

Diffusion language models have recently emerged as a competitive alternative to autoregressive language models. Beyond next-token generation, they are more efficient and flexible b…

cs.LG2025

PENCIL: Long Thoughts with Short Memory

Chenxiao Yang, Nathan Srebro, David McAllester +1

While state-of-the-art LLMs have demonstrated great promise of using long Chains-of-Thought (CoT) to boost reasoning, scaling it up to more challenging problems at test-time is fun…

cs.LG20245 cited

GeoMix: Towards Geometry-Aware Data Augmentation

Wentao Zhao, Qitian Wu, Chenxiao Yang +1

Mixup has shown considerable success in mitigating the challenges posed by limited labeled data in image classification. By synthesizing samples through the interpolation of featur…