Papers by Ganqu Cui
From Adversarial Arms Race to Model-centric Evaluation: Motivating a Unified Automatic Robustness Evaluation Framework (2023.findings-acl)
Copied to clipboard
Yangyi Chen, Hongcheng Gao, Ganqu Cui, Lifan Yuan, Dehan Kong, Hanlu Wu, Ning Shi, Bo Yuan, Longtao Huang, Hui Xue, Zhiyuan Liu, Maosong Sun, Heng Ji
| Challenge: | Existing models of robustness evaluation are incomprehensive, impractical, and invalid . |
| Approach: | They propose a framework for automatic robustness evaluation that shifts towards model-centric evaluation to further exploit the advantages of adversarial attacks. |
| Outcome: | The proposed framework is based on a model-centric evaluation protocol and a robustness evaluation protocol. |
Why Should Adversarial Perturbations be Imperceptible? Rethink the Research Paradigm in Adversarial NLP (2022.emnlp-main)
Copied to clipboard
| Challenge: | Textual adversarial samples are often misrepresented in research on security, evaluation, explainability, and data augmentation. |
| Approach: | They propose to use adversarial samples to evaluate their methods on security tasks to demonstrate the real-world concerns rather than developing impractical methods. |
| Outcome: | The proposed method has higher practical value than the current benchmark. |
Exploring the Universal Vulnerability of Prompt-based Learning Paradigm (2022.findings-naacl)
Copied to clipboard
| Challenge: | Prompt-based learning inherits the vulnerability from pre-training, where model predictions can be misled by inserting triggers into the text. |
| Approach: | They propose a potential solution to mitigate this vulnerability by injecting triggers into pre-trained language models using only plain text. |
| Outcome: | The proposed learning paradigm inherits the vulnerability from the pre-training stage . it can totally control or severely decrease the performance of prompt-based models . |
Few-shot Classification with Hypersphere Modeling of Prototypes (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for fewshot learning use embeddings in space, but they lack expressivity and are difficult to perform statistically. |
| Approach: | They propose a method where class information is represented by hyperspheres with dynamic sizes with two sets of learnable parameters: the hypersphere’s center and the radius. |
| Outcome: | The proposed method is much more expressive than embeddings and performs better than statistical modeling. |
Controllable Preference Optimization: Toward Controllable Multi-Objective Alignment (2024.emnlp-main)
Copied to clipboard
Yiju Guo, Ganqu Cui, Lifan Yuan, Ning Ding, Zexu Sun, Bowen Sun, Huimin Chen, Ruobing Xie, Jie Zhou, Yankai Lin, Zhiyuan Liu, Maosong Sun
| Challenge: | Existing algorithms for achieving optimal alignment are mostly unidirectional . a recent study suggests that large language models can be ground with evident preferences . |
| Approach: | They propose to ground large language models with evident preferences . they propose to use controllable preference optimization to specify different objectives . |
| Outcome: | The proposed models can provide responses that match various preferences among the ”3H” desiderata. |
Fusing Highly Specialized Language Models for Comprehensive Expertise (2025.acl-long)
Copied to clipboard
Ning Ding, Yulin Chen, Ganqu Cui, Xingtai Lv, Weilin Zhao, Kaiyan Zhang, Ruobing Xie, Bowen Zhou, Zhiyuan Liu, Maosong Sun
| Challenge: | Existing models that focus on language, programming code, and mathematical symbols are not able to achieve mastery of all three domains simultaneously. |
| Approach: | They propose to fuse highly-specialized models that are already sufficiently trained on different domains to achieve a highly-specific model. |
| Outcome: | The proposed model could achieve mastery of the three crucial domains simultaneously. |
Decoder Tuning: Efficient Language Understanding as Decoding (2023.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to adapt pre-trained models with parameters frozen are based on input-side adaptation, which requires thousands of API queries. |
| Approach: | They propose to train a model-as-a-service (MaaS) setting to provide only the inference APIs for users . they argue that input-side adaptation could be arduous due to the lack of gradient signals . |
| Outcome: | The proposed model outperforms state-of-the-art algorithms with a 200x speed-up. |
INTERVENOR: Prompting the Coding Ability of Large Language Models with the Interactive Chain of Repair (2024.findings-acl)
Copied to clipboard
| Challenge: | Experimental results show that INTERVENOR surpasses baseline models, exhibiting improvements of approximately 18% and 4.3% over GPT-3.5 in code generation and code translation tasks. |
| Approach: | They propose a system that prompts Large Language Models to play distinct roles during the code repair process, functioning as both a Code Learner and a code teacher. |
| Outcome: | The proposed system surpasses baseline models in code generation and code translation tasks and improves on syntax errors and assertion errors. |
Scalable Efficient Training of Large Language Models with Low-dimensional Projected Attention (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies have found that low-rank pre-training often compromises effectiveness. |
| Approach: | They propose to apply low-dimensional module only to the attention layer to improve both effectiveness and efficiency. |
| Outcome: | The proposed model saves 12.4% time while improving test perplexity and on downstream tasks compared with vanilla Transformer. |
A Close Look into the Calibration of Pre-trained Language Models (2023.acl-long)
Copied to clipboard
| Challenge: | Pre-trained language models (PLMs) may fail in giving reliable estimates of their predictive uncertainty. |
| Approach: | They conduct fine-grained control experiments to study the dynamic change in PLMs’ calibration performance in training. |
| Outcome: | The proposed methods significantly reduce PLMs’ confidence in wrong predictions. |
The Right Time Matters: Data Arrangement Affects Zero-Shot Generalization in Instruction Tuning (2025.findings-acl)
Copied to clipboard
Bingxiang He, Ning Ding, Cheng Qian, Jia Deng, Ganqu Cui, Lifan Yuan, Haiwen Hong, Huan-ang Gao, Longtao Huang, Hui Xue, Huimin Chen, Zhiyuan Liu, Maosong Sun
| Challenge: | Existing work on instruction tuning has focused on task level, without considering that tasks are artificially defined and, to LLMs, merely consist of tokens and representations. |
| Approach: | They propose a training data arrangement framework that allows for continual learning and loss reduction. |
| Outcome: | The proposed framework promotes continual learning and loss reduction on unseen tasks. |
V-GameGym: Visual Game Generation for Code Large Language Models (2026.findings-acl)
Copied to clipboard
Wei Zhang, Jian Yang, Renshuai Tao, Linzheng Chai, Shuyue Guo, Jiajun Wu, Xiaoming Chen, Ganqu Cui, Ning Ding, Xander Xu, HU Wei, Bowen Zhou
| Challenge: | Existing code-related benchmarks focus on single modality rather than visual game development. |
| Approach: | They propose a multimodal benchmark for evaluating code large language models in visual game generation that integrates a clustering-based curation methodology and a pipeline for visual code synthesis. |
| Outcome: | The proposed framework assesses code generation and visual game generation using a sandbox environment. |
Prototypical Verbalizer for Prompt-based Few-shot Tuning (2022.acl-long)
Copied to clipboard
| Challenge: | Prompt-based tuning for pre-trained language models has shown its effectiveness in few-shot learning. |
| Approach: | They propose a prototypical verbalizer which learns prototype vectors as verbalizes by contrastive learning. |
| Outcome: | The proposed verbalizer outperforms existing verbalizing methods on topic classification and entity typing tasks. |
UltraIF: Advancing Instruction Following from the Wild (2025.emnlp-main)
Copied to clipboard
| Challenge: | a lack of transparency has resulted in a gap between research community and leading companies . large language models have demonstrated remarkable capabilities in following complex instructions . |
| Approach: | They propose a method to build large language models that can follow complex instructions with open-source data. |
| Outcome: | The proposed approach can synergize complex instructions and filter responses with evaluation questions. |
Beat LLMs at Their Own Game: Zero-Shot LLM-Generated Text Detection via Querying ChatGPT (2023.emnlp-main)
Copied to clipboard
Biru Zhu, Lifan Yuan, Ganqu Cui, Yangyi Chen, Chong Fu, Bingxiang He, Yangdong Deng, Zhiyuan Liu, Maosong Sun, Ming Gu
| Challenge: | Large language models (LLMs) are capable of performing tasks but are likely to be misused. |
| Approach: | They propose a zero-shot black-box method to detect LLM-generated texts . they revise the text to be detected using the ChatGPT model . |
| Outcome: | The proposed method can detect LLM-generated texts with a zero-shot black-box model . it is based on intuition that the model will make fewer revisions to LLMs than to human-written texts . |
The Best of Both Worlds: Combining Parallel and Sequential Inference Scaling via Aggregation Fine-Tuning (2026.findings-acl)
Copied to clipboard
| Challenge: | Empirical results show that AFT-trained models achieve substantial gains with test-time scaling. |
| Approach: | They introduce a supervised fine-tuning paradigm where models synthesize multiple draft responses into a single, refined answer. |
| Outcome: | Empirical results show that AFT-trained models outperform baseline models while eliminating external guidance. |