Author

Tonghan Wang

3 papers indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Preprint Aug 2026

Verbalizing Multi-Token Concepts in LLMs

Lens methods inspect model computation by mapping intermediate activations to vocabulary tokens. Yet the concepts humans need to read out often span multiple tokens---entities, phrases, intermediate objects---making token-level readouts incomplete. Reliable multi-token readout with little model-specific preparation rem...

Xiaotao Gong, Zi-Meng Huang, Tonghan Wang · 0 citations

#artificial intelligence May 2026

How LLMs Are Persuaded: A Few Attention Heads, Rerouted

This work studies causal interventions in controlled multiple-choice tasks and identifies that persuasion mainly acts through a small set of mid-layer attention heads, which are decision heads, which mainly acts through a one-dimensional option-routing feature in the option-token representations.

Xiang-Kun Sun, Ling-Kai Kong, Ao Zhang et al. · arXiv.org · 2 citations · ⚡1

#artificial intelligence Preprint Sep 2026

A mechanistic study of language model introspection

Across three model families, two small groups of attention heads are identified with distinct roles in introspective reporting supporting introspective detection and localization of large language models.

Jia-Hong Zou, Xiang-Kun Sun, Ling-Kai Kong et al. · 0 citations