Skip to content

Author

Yiwei Cai

We have 6 of 15 papers

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.

#natural language process... Preprint Nov 2026

When Safety Becomes An Outlier: Understanding the Retention of LLM Safety Behaviors

Safety-aligned large language models often lose their safety behaviors after fine-tuning, even when safety data are included. While this fragility is well documented, its underlying cause remains unclear. In this paper, we propose that formulaic refusal strings can function as outlier modes: narrow, weakly grounded res...

Binchi Zhang, H. Abdullah, Yi-Wei Cai et al. · 0 citations
#machine learning Preprint Sep 2026

LastOPD: Taming Collapse in Latent On-Policy Distillation

LastOPD is proposed, which applies the latent signal only at the last-layer state, the common interface both LM heads read, and only during a 10-step crossfade into token-level OPD, which keeps the useful part of the latent signal and hands the student to token-level supervision before the collapse sets in.

Jie Yang, Zheng-Yu Fang, Ze-Lin Xu et al. · 3 citations

Detecting Data Poisoning in Code Generation LLMs via Black-Box, Vulnerability-Oriented Scanning

CodeScan is presented, the first black-box, vulnerability-specific scanning framework for auditing code generation LLMs, assuming that the defender specifies the target vulnerability classes and provides corresponding task-relevant prompts.

Shenao Yan, Sha-Sha Jin, S. Ahmed et al. · 0 citations
#artificial intelligence Preprint Sep 2026

TimeEvo: Failure-Driven Self-Evolution of a Time Series Agent

TimeEvo is proposed, which clusters an agent's diagnosed failures into capability gaps, plans a measurement for each, synthesizes evidence-only tools that fill them, and admits the candidate library only through a paired admission gate.

Jie Yang, Yan Zheng, Jia-Rui Sun et al. · 0 citations
#machine learning Preprint Aug 2026

Token Signatures of Code: Comparing Coding Behaviors Across Large Language Models

An interactive visual analytics system is developed to navigate the comparison landscape, identify pairs of interest, and drill down into discriminative tokens and their code contexts, as well as defining two new metrics: robustness and concentration.

Jun-Peng Wang, Yuzhong Chen, Menghai Pan et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.