activity
20242026
most citedThought calibration: Efficient and confident test-time scaling

1 citations · 1 across the 2 of their papers we have counts for

collaborators

7 papers

cs.LG2026

Rethinking Diffusion Models with Symmetries through Canonicalization with Applications to Molecular Graph Generation

Cai Zhou, Zijie Chen, Zian Li +7

Many generative tasks in chemistry and science involve distributions invariant to group symmetries (e.g., permutation and rotation). A common strategy enforces invariance and equiv…

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.CL2025

Next Semantic Scale Prediction via Hierarchical Diffusion Language Models

Cai Zhou, Chenyu Wang, Dinghuai Zhang +4

In this paper we introduce Hierarchical Diffusion Language Models (HDLM) -- a novel family of discrete diffusion models for language modeling. HDLM builds on a hierarchical vocabul…

cs.CL2025

SPG: Sandwiched Policy Gradient for Masked Diffusion Language Models

Chenyu Wang, Paria Rashidinejad, DiJia Su +9

Diffusion large language models (dLLMs) are emerging as an efficient alternative to autoregressive models due to their ability to decode multiple tokens in parallel. However, align…

cs.LG2025

Learning Diffusion Models with Flexible Representation Guidance

Chenyu Wang, Cai Zhou, Sharut Gupta +4

Diffusion models can be improved with additional guidance towards more effective representations of input. Indeed, prior empirical work has already shown that aligning internal rep…

cs.LG20251 cited

Thought calibration: Efficient and confident test-time scaling

Menghua Wu, Cai Zhou, Stephen Bates +1

Reasoning large language models achieve impressive test-time scaling by thinking for longer, but this performance gain comes at significant compute cost. Directly limiting test-tim…