AI safety researcher

Jiantao Jiao

Google Scholar dblp ORCID

Analysis and theory 2Evaluation and benchmarks 1Interpretability 1

Latest AI safety papers

On this site: 11 papers

September 2025

Emergence of Superposition: Unveiling the Training Dynamics of Chain of Continuous Thought
Sep 27Analysis and theoryarXiv ↗
GSM-Agent: Understanding Agentic Reasoning Using Controllable Environments
Sep 26Evaluation and benchmarksarXiv ↗

June 2025

Generalization or Hallucination? Understanding Out-of-Context Reasoning in Transformers
Jun 12Analysis and theoryarXiv ↗

February 2025

How Do LLMs Perform Two-Hop Reasoning in Context?
Feb 19InterpretabilityarXiv ↗

May 2024

Toxicity Detection for Free
May 29arXiv ↗

Authorship follows dblp and the author's own Google Scholar profile; only AI safety papers on this site are listed.