Papers by Chengzu Li
Enriching Patent Claim Generation with European Patent Dataset (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing work on large language models to assist inventors in writing patent claims relies on datasets from the United States Patent and Trademark Office. |
| Approach: | They propose a European patent dataset that provides rich textual data and structured metadata to support multiple patent-related tasks. |
| Outcome: | The proposed dataset outperforms existing datasets and GPT-4o in claim quality and cross-domain generalization. |
TopViewRS: Vision-Language Models as Top-View Spatial Reasoners (2024.emnlp-main)
Copied to clipboard
| Challenge: | Top-view perspective is a typical way in which humans read and reason over different types of maps, but spatial reasoning capabilities of modern VLMs in this setup remain unattested and underexplored. |
| Approach: | They introduce a top-view spatial reasoning dataset and use it to evaluate VLMs across 4 perception and reasoning tasks with different levels of complexity. |
| Outcome: | The proposed model can understand and reason over spatial relations from the top view and can be controlled at different granularities of spatial reasoning. |
LoVeC: Reinforcement Learning for Better Verbalized Confidence in Long-Form Generation (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods for hallucination detection are limited to short-form question answering tasks and do not generalize well to open-ended generation. |
| Approach: | They propose a method that trains LLMs to append a numerical confidence score to each generated statement during long-form generation. |
| Outcome: | The proposed method is 20 faster than traditional self-consistency methods while achieving better calibration. |
Generating Data for Symbolic Language with Large Language Models (2023.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) bring performance and complexity, but they incur a large computational cost in practice. |
| Approach: | They propose a task-based model which uses large language models to generate symbolic language data by an informative prompt and agreement-based verifier. |
| Outcome: | The proposed model can generate symbolic language data with a few human demonstrations and saves a considerable amount of inference effort. |
UnifiedSKG: Unifying and Multi-Tasking Structured Knowledge Grounding with Text-to-Text Language Models (2022.emnlp-main)
Copied to clipboard
Tianbao Xie, Chen Henry Wu, Peng Shi, Ruiqi Zhong, Torsten Scholak, Michihiro Yasunaga, Chien-Sheng Wu, Ming Zhong, Pengcheng Yin, Sida I. Wang, Victor Zhong, Bailin Wang, Chengzu Li, Connor Boyle, Ansong Ni, Ziyu Yao, Dragomir Radev, Caiming Xiong, Lingpeng Kong, Rui Zhang, Noah A. Smith, Luke Zettlemoyer, Tao Yu
| Challenge: | Structured knowledge grounding (SKG) uses structured knowledge to complete user requests . since inputs and outputs of SKG tasks are heterogeneous, they have been studied separately . |
| Approach: | They propose a framework that unifies 21 SKG tasks into a text-to-text format . they use unifiedSKG to benchmark T5 with different sizes . |
| Outcome: | The proposed framework unifies 21 SKG tasks into a text-to-text format . it achieves state-of-the-art performance on almost all of the 21 tasks, the authors show . |
Confidence Estimation for LLMs in Multi-turn Interactions (2026.findings-acl)
Copied to clipboard
Caiqi Zhang, Ruihan Yang, Xiaochen Zhu, Chengzu Li, Tiancheng Hu, Yijiang River Dong, Deqing Yang, Nigel Collier
| Challenge: | Despite recent progress, most prior work studies confidence in single-turn question answering. |
| Approach: | They propose a logit-based probe that measures confidence in multi-turn dialogues . they propose 'infoECE' and a "hinter-guesser" paradigm for generating controlled evaluations based on data . |
| Outcome: | The proposed framework is grounded in calibration and monotonicity of confidence as more information becomes available. |
Lost in Embeddings: Information Loss in Vision–Language Models (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Experiments reveal connectors substantially distort the local geometry of visual representations, with k-nearest neighbors diverging by 40–60% post-projection, correlating with degradation in retrieval performance. |
| Approach: | They propose two approaches to examine and quantify information loss by analyzing latent representation space. |
| Outcome: | The proposed model improves retrieval performance by analyzing changes in k-nearest neighbor relationships between image representations before and after projection. |
Semantic Map-based Generation of Navigation Instructions (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing approaches to navigation instruction generation use a sequence of panorama images as visual input. |
| Approach: | They propose a new approach to navigation instruction generation using semantic maps as visual input and frame it as an image captioning task. |
| Outcome: | The proposed model is based on a dataset of a human vision and language navigation task and human subjects are asked to manually assess the quality of the generated instructions. |
Large Language Models are Miscalibrated In-Context Learners (2025.findings-acl)
Copied to clipboard
| Challenge: | In-context Learning and Supervised Fine-Tuning have emerged as pre-dominant methodologies for machine learning and NLP. |
| Approach: | They propose to use self-ensembling to improve both performance and calibration of language models. |
| Outcome: | The proposed learning paradigms can achieve better calibration and better performance than the previous learning paradigm. |