Papers by Ke Chao
“I’ve Decided to Leak”: Probing Internals Behind Prompt Leakage Intents (2025.emnlp-main)
Copied to clipboard
Jianshuo Dong, Yutong Zhang, Liu Yan, Zhenyu Zhong, Tao Wei, Ke Xu, Minlie Huang, Chao Zhang, Han Qiu
| Challenge: | Large language models (LLMs) exhibit prompt leakage vulnerabilities, raising intellectual property and confidentiality concerns. |
| Approach: | They use probing techniques to capture LLMs’ intent-related internal representations and show that they internalize prompt leakage intents in their hidden states before generating tokens. |
| Outcome: | The proposed probes achieve 90%+ AUROC across all tested models, even when applied to new system prompts and attacks. |
Lost in Overlap: Exploring Logit-based Watermark Collision in LLMs (2025.findings-naacl)
Copied to clipboard
| Challenge: | Existing watermarking methods embed imperceptible identifiers into text to address copyright concerns. |
| Approach: | They propose a new philosophy for watermark attacks that addresses watermark collision . they demonstrate that collision poses a threat to all logit-based watermark algorithms . |
| Outcome: | The proposed method improves watermark collision performance on top of other methods. |
DP3: Differentially Private Prompt Perturbation for Multi-turn LLM Inference (2026.findings-acl)
Copied to clipboard
| Challenge: | Large language models (LLMs) are widely used for text understanding and generation . existing methods that assume single-turn interactions break down in multi-turn settings . |
| Approach: | They propose a differentially private prompt perturbation framework for multi-turn LLM inference . DP3 constructs a perturbation mapping table to reuse perturbations for recurring tokens . |
| Outcome: | The proposed framework reduces privacy costs and degrades cross-turn semantic coherence . it also provides a context-aware utility function to maintain semantic consistency across turns . |
Training Language Model to Critique for Better Refinement (2025.findings-acl)
Copied to clipboard
Tianshu Yu, Chao Xiang, Mingchuan Yang, Pei Ke, Bosi Wen, Cunxiang Wang, Jiale Cheng, Li Zhang, Xinyu Mu, Chuxiong Sun, Minlie Huang
| Challenge: | Large language models (LLMs) have remarkable evaluation and critique capabilities, providing insightful feedback and identifying flaws in various tasks. |
| Approach: | They propose a framework to train critic models using refinement signals to generate feedback loops where critiques guide the model in refining its responses. |
| Outcome: | The proposed framework outperforms traditional methods and open-source models in terms of critique quality and refinement outcomes. |
Improving BERT with Syntax-aware Local Attention (2021.findings-acl)
Copied to clipboard
| Challenge: | Recent studies show that attention-based models benefit from more focused attention over local regions. |
| Approach: | They propose a syntax-aware local attention which restrains attention over syntactically relevant words. |
| Outcome: | The proposed model performs better on all benchmark datasets, including sentence classification and sequence labeling tasks. |
Bridging Internal Consistency and External Alignment: A Causal and Dynamic Interpretability Framework for LLM Generation (2026.acl-long)
Copied to clipboard
| Challenge: | Existing interpretability methods focus on internal and external aspects of the model . existing explanations often focus on surface correlations or static dependencies . |
| Approach: | They propose a causal and dynamic interpretability framework for Large Language Models . they characterize backdoor-adjusted causal effects of generated prefix and prompt . |
| Outcome: | The proposed framework provides a unified causal view of internal consistency and external alignment in LLM generation dynamics. |
At Your Own PACE: A Causal Framework for Evaluating EQ in LLMs (2026.findings-acl)
Copied to clipboard
| Challenge: | Emotional Quotient (EQ) has emerged as a competency for seamless human-AI integration. |
| Approach: | They propose a framework for a closed-loop EQ evaluation using a PACE taxonomy to define four dimensions of LLM EQ. |
| Outcome: | The proposed framework achieves high alignment of 89.31% with human preferences while maintaining robust consistency of 83.6%. |