arXiv subjects

Zhengxiang Wang

Publications and source records attributed to Zhengxiang Wang.

13 recordsLinked to original sources

arXiv2026

When Seeing Is Not Enough: Benchmarking Interactive Visual Grounding in LVLMs

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 reference: initial referring expressions are often incomplete or ambiguous, requiring participants to establish shared understanding through interaction. We introduce a controlled evaluation framework for interactive visual grounding in large vision-language models (LVLMs), varying how much target information is provided upfront and how much must be acquired through dialogue. Across four human-grounded visual contexts and four interaction protocols, current LVLMs perform significantly below task-level human baselines. Interaction can help when follow-up questions refine or repair an initial target description. Performance is lowest when no initial description is provided and target information must be acquired through questions, indicating that proactive question-driven grounding remains difficult. LVLMs are also poorly calibrated, often reporting confidence that exceeds their empirical accuracy. Follow-up studies confirm these patterns across varied description sources (human versus AI), reasoning efforts, repeated interactions, description providers, and visual contexts. Overall, interactive visual grounding remains challenging, requiring visual matching, information seeking and synthesis.

cs.AI↗

arXiv2026

LVLMs and Humans Ground Differently in Referential Communication

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 critical deficit: an inability to model common ground. We present a referential communication experiment with a factorial design involving director-matcher pairs (human-human, human-AI, AI-human, and AI-AI) that interact with multiple turns in repeated rounds to match pictures of objects not associated with any obvious lexicalized labels. We show that LVLMs cannot interactively generate and resolve referring expressions in a way that enables smooth communication, a crucial skill that underlies human language use. We release our corpus of 356 dialogues (89 pairs over 4 rounds each) along with the online pipeline for data collection and the tools for analyzing accuracy, efficiency, and lexical overlap.

cs.CL↗

arXiv2026

Measuring Iterative Temporal Reasoning with Time Puzzles

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 mainly in static, non-tool-using settings, which poorly reflect how LLMs perform temporal reasoning in practice. We introduce Time Puzzles, a constraint-based date inference task for evaluating iterative temporal reasoning with tools. Each puzzle combines factual temporal anchors with (cross-cultural) calendar relations and may admit one or multiple valid dates. The puzzles are algorithmically generated, enabling controlled and continual evaluation. Across 13 LLMs, even the best model (GPT-5) achieves only 55.3% accuracy without tools, despite using easily searchable facts. While web search improves performance, models perform substantially better when constraints are rewritten with explicit dates, removing the need for factual lookup. These results reveal a gap in reliable tool use for iterative temporal reasoning.

cs.CL↗

arXiv2025

LVLMs are Bad at Overhearing Human Referential Communication

During spontaneous conversations, speakers collaborate on novel referring expressions, which they can then re-use in subsequent conversations. Understanding such referring expressions is an important ability for an embodied agent, so that it can carry out tasks in the real world. This requires integrating and understanding language, vision, and conversational interaction. We study the capabilities of seven state-of-the-art Large Vision Language Models (LVLMs) as overhearers to a corpus of spontaneous conversations between pairs of human discourse participants engaged in a collaborative object-matching task. We find that such a task remains challenging for current LVLMs and they all fail to show a consistent performance improvement as they overhear more conversations from the same discourse participants repeating the same task for multiple rounds. We release our corpus and code for reproducibility and to facilitate future research.

cs.CL↗

arXiv2025

Catch Me If You Can? Not Yet: LLMs Still Struggle to Imitate the Implicit Writing Styles of Everyday Authors

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 from just a few examples? Personal style is often subtle and implicit, making it difficult to specify through prompts yet essential for user-aligned generation. This work presents a comprehensive evaluation of state-of-the-art LLMs' ability to mimic personal writing styles via in-context learning from a small number of user-authored samples. We introduce an ensemble of complementary metrics-including authorship attribution, authorship verification, style matching, and AI detection-to robustly assess style imitation. Our evaluation spans over 40000 generations per model across domains such as news, email, forums, and blogs, covering writing samples from more than 400 real-world authors. Results show that while LLMs can approximate user styles in structured formats like news and email, they struggle with nuanced, informal writing in blogs and forums. Further analysis on various prompting strategies such as number of demonstrations reveal key limitations in effective personalization. Our findings highlight a fundamental gap in personalized LLM adaptation and the need for improved techniques to support implicit, style-consistent generation. To aid future research and for reproducibility, we open-source our data and code.

cs.CL↗

arXiv2025

Evaluating LLMs with Multiple Problems at once

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-problem evaluation, where a prompt presents a single problem and expects one specific answer, MPE places multiple problems together in a single prompt and assesses how well an LLM answers all these problems in a single output. Leveraging 6 classification and 12 reasoning benchmarks that already exist, we introduce a new benchmark called ZeMPE (Zero-shot Multi-Problem Evaluation), comprising 53,100 zero-shot multi-problem prompts. We experiment with a total of 13 LLMs from 5 model families on ZeMPE to present a comprehensive and systematic MPE. Our results show that LLMs are capable of handling multiple problems from a single data source as well as handling them separately, but there are conditions this multiple problem handling capability falls short. In addition, we perform in-depth further analyses and explore model-level factors that may enable multiple problem handling capabilities in LLMs. We release our corpus and code to facilitate future research.

cs.AI↗

arXiv2025

LLMs can Perform Multi-Dimensional Analytic Writing Assessments: A Case Study of L2 Graduate-Level Academic English Writing

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 multiple assessment criteria. Using a corpus of literature reviews written by L2 graduate students and assessed by human experts against 9 analytic criteria, we prompt several popular LLMs to perform the same task under various conditions. To evaluate the quality of feedback comments, we apply a novel feedback comment quality evaluation framework. This framework is interpretable, cost-efficient, scalable, and reproducible, compared to existing methods that rely on manual judgments. We find that LLMs can generate reasonably good and generally reliable multi-dimensional analytic assessments. We release our corpus and code for reproducibility.

cs.CL↗

arXiv2024

Clustering Document Parts: Detecting and Characterizing Influence Campaigns from Documents

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 an influence campaign, and then identifies documents linked to an influence campaign via their association with the high-influence clusters. Our approach outperforms both the direct document-level classification and the direct document-level clustering approach in predicting if a document is part of an influence campaign. We propose various novel techniques to enhance our pipeline, including using an existing event factuality prediction system to obtain document parts, and aggregating multiple clustering experiments to improve the performance of both cluster and document classification. Classifying documents after clustering not only accurately extracts the parts of the documents that are relevant to influence campaigns, but also captures influence campaigns as a coordinated and holistic phenomenon. Our approach makes possible more fine-grained and interpretable characterizations of influence campaigns from documents.

cs.CL↗

arXiv2024

Learning Transductions and Alignments with RNN Seq2seq Models

The paper studies the capabilities of Recurrent-Neural-Network sequence to sequence (RNN seq2seq) models in learning four transduction tasks: identity, reversal, total reduplication, and quadratic copying. These transductions are traditionally well studied under finite state transducers and attributed with increasing complexity. We find that RNN seq2seq models are only able to approximate a mapping that fits the training or in-distribution data, instead of learning the underlying functions. Although attention makes learning more efficient and robust, it does not overcome the out-of-distribution generalization limitation. We establish a novel complexity hierarchy for learning the four tasks for attention-less RNN seq2seq models, which may be understood in terms of the complexity hierarchy of formal languages, instead of string transductions. RNN variants also play a role in the results. In particular, we show that Simple RNN seq2seq models cannot count the input length.

cs.CL↗

arXiv2023

Probabilistic Linguistic Knowledge and Token-level Text Augmentation

This paper investigates the effectiveness of token-level text augmentation and the role of probabilistic linguistic knowledge within a linguistically-motivated evaluation context. Two text augmentation programs, REDA and REDA$_{NG}$, were developed, both implementing five token-level text editing operations: Synonym Replacement (SR), Random Swap (RS), Random Insertion (RI), Random Deletion (RD), and Random Mix (RM). REDA$_{NG}$ leverages pretrained $n$-gram language models to select the most likely augmented texts from REDA's output. Comprehensive and fine-grained experiments were conducted on a binary question matching classification task in both Chinese and English. The results strongly refute the general effectiveness of the five token-level text augmentation techniques under investigation, whether applied together or separately, and irrespective of various common classification model types used, including transformers. Furthermore, the role of probabilistic linguistic knowledge is found to be minimal.

cs.CL↗

arXiv2022

Random Text Perturbations Work, but not Always

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 perturbations as a data augmentation approach for NLP. It is found that the augmentation can bring both negative and positive effects to the test set performance of three neural classification models, depending on whether the models train on enough original training examples. This remains true no matter whether five random text editing operations, used to augment text, are applied together or separately. Our study demonstrates with strong implication that the effectiveness of random text perturbations is task specific and not generally positive.

cs.CL↗

arXiv2022

Linguistic Knowledge in Data Augmentation for Natural Language Processing: An Example on Chinese Question Matching

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 Large-scale Chinese Question Matching Corpus) for a binary Chinese question matching classification task. The two DA programs produce augmented texts by five simple text editing operations (or DA techniques), largely irrespective of language generation rules, but one is enhanced with a pre-trained n-gram language model to fuse it with prior linguistic knowledge. We then trained four neural network models (BOW, CNN, LSTM, and GRU) and a pre-trained model (ERNIE-Gram) on the LCQMCs train sets of varying size as well as the related augmented train sets produced by the two DA programs. The results show that there are no significant performance differences between the models trained on the two types of augmented train sets, both when the five DA techniques are applied together or separately. Moreover, due to the inability of the five DA techniques to make strictly paraphrastic augmented texts, the results indicate the need of sufficient amounts of training examples for the classification models trained on them to mediate the negative impact of false matching augmented text pairs and improve performance, a limitation of random text editing perturbations used as a DA approach. Similar results were also obtained for English.

cs.CL↗

arXiv2022

Thirty-Two Years of IEEE VIS: Authors, Fields of Study and Citations

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 investigated (1) who contributed to VIS, and (2) where VIS stands in the scientific world. We examined the authors and fields of study of 3,240 VIS publications in the past 32 years based on data collected from OpenAlex and IEEE Xplore, among other sources. We also examined the citation flows from referenced papers (i.e., those referenced in VIS) to VIS, and from VIS to citing papers (i.e., those citing VIS). We found that VIS has been becoming increasingly popular and collaborative. The number of publications, of unique authors, and of participating countries have been steadily growing. Both cross-country collaborations, and collaborations between educational and non-educational affiliations, namely "cross-type collaborations", are increasing. The dominance of the US is decreasing, and authors from China are now an important part of VIS. In terms of author affiliation types, VIS is increasingly dominated by authors from universities. We found that the topics, inspirations, and influences of VIS research is limited such that (1) VIS, and their referenced and citing papers largely fall into the Computer Science domain, and (2) citations flow mostly between the same set of subfields within Computer Science. Our citation analyses showed that award-winning VIS papers had higher citations. Interactive visualizations, replication data, source code and supplementary material are available at https://32vis.hongtaoh.com and https://osf.io/zkvjm.

cs.DL↗