14 papers
Last Translation Benchmark
Vilém Zouhar, Niyati Bafna, Mukund Choudhary +263
For scientific progress, we need benchmarks that test the limits of state-of-the-art models, and evaluation methods that inform us about failure cases. As models get stronger, stan…
When Seeing Is Not Enough: Benchmarking Interactive Visual Grounding in LVLMs
Zhengxiang Wang, Owen Rambow
Visual grounding is typically evaluated as a one-shot mapping from an informative referring expression to a visual target. This formulation misses a central property of real-world…
LVLMs and Humans Ground Differently in Referential Communication
Peter Zeng, Weiling Li, Amie J. Paige +6
For generative AI agents to partner effectively with human users, the ability to accurately predict human intent is critical. But this ability to collaborate remains limited by a c…
Measuring Iterative Temporal Reasoning with Time Puzzles
Zhengxiang Wang, Zeyu Dong
Tool use, such as web search, has become a standard capability even in freely available large language models (LLMs). However, existing benchmarks evaluate temporal reasoning mainl…
LVLMs are Bad at Overhearing Human Referential Communication
Zhengxiang Wang, Weiling Li, Panagiotis Kaliosis +2
During spontaneous conversations, speakers collaborate on novel referring expressions, which they can then re-use in subsequent conversations. Understanding such referring expressi…
Catch Me If You Can? Not Yet: LLMs Still Struggle to Imitate the Implicit Writing Styles of Everyday Authors
Zhengxiang Wang, Nafis Irtiza Tripto, Solha Park +2
As large language models (LLMs) become increasingly integrated into personal writing tools, a critical question arises: can LLMs faithfully imitate an individual's writing style fr…
LLMs can Perform Multi-Dimensional Analytic Writing Assessments: A Case Study of L2 Graduate-Level Academic English Writing
Zhengxiang Wang, Veronika Makarova, Zhi Li +2
The paper explores the performance of LLMs in the context of multi-dimensional analytic writing assessments, i.e. their ability to provide both scores and comments based on multipl…
Evaluating LLMs with Multiple Problems at once
Zhengxiang Wang, Jordan Kodner, Owen Rambow
This paper shows the benefits and fruitfulness of evaluating LLMs with multiple problems at once, a paradigm we call multi-problem evaluation (MPE). Unlike conventional single-prob…
Clustering Document Parts: Detecting and Characterizing Influence Campaigns from Documents
Zhengxiang Wang, Owen Rambow
We propose a novel clustering pipeline to detect and characterize influence campaigns from documents. This approach clusters parts of document, detects clusters that likely reflect…
Probabilistic Linguistic Knowledge and Token-level Text Augmentation
Zhengxiang Wang
This paper investigates the effectiveness of token-level text augmentation and the role of probabilistic linguistic knowledge within a linguistically-motivated evaluation context.…
Learning Transductions and Alignments with RNN Seq2seq Models
Zhengxiang Wang
The paper studies the capabilities of Recurrent-Neural-Network sequence to sequence (RNN seq2seq) models in learning four transduction tasks: identity, reversal, total reduplicatio…
Random Text Perturbations Work, but not Always
Zhengxiang Wang
We present three large-scale experiments on binary text matching classification task both in Chinese and English to evaluate the effectiveness and generalizability of random text p…
Thirty-Two Years of IEEE VIS: Authors, Fields of Study and Citations
Hongtao Hao, Yumian Cui, Zhengxiang Wang +1
The IEEE VIS Conference (VIS) recently rebranded itself as a unified conference and officially positioned itself within the discipline of Data Science. Driven by this movement, we…
Linguistic Knowledge in Data Augmentation for Natural Language Processing: An Example on Chinese Question Matching
Zhengxiang Wang
To investigate the role of linguistic knowledge in data augmentation (DA) for Natural Language Processing (NLP), we designed two adapted DA programs and applied them to LCQMC (a La…