Haining Wang


pdf bib
What Factors Influence LLMs’ Judgments? A Case Study on Question Answering
Lei Chen | Bobo Li | Li Zheng | Haining Wang | Zixiang Meng | Runfeng Shi | Hao Fei | Jun Zhou | Fei Li | Chong Teng | Donghong Ji
Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)

Large Language Models (LLMs) are now being considered as judges of high efficiency to evaluate the quality of answers generated by candidate models. However, their judgments may be influenced by complex scenarios and inherent biases, raising concerns about their reliability. This study aims to bridge this gap by introducing four unexplored factors and examining the performance of LLMs as judges, namely answer quantity, inducing statements, judging strategy, and judging style. Additionally, we introduce a new dimension of question difficulty to provide a more comprehensive understanding of LLMs’ judgments across varying question intricacies. We employ ChatGPT, GPT-4, Gemini, and Claude-2 as judges and conduct experiments on Vicuna Benchmark and MT-bench. Our study reveals that LLMs’ judging abilities are susceptible to the influence of these four factors, and analyzing from the newly proposed dimension of question difficulty is highly necessary. We also provide valuable insights into optimizing LLMs’ performance as judges, enhancing their reliability and adaptability across diverse evaluation scenarios.


pdf bib
CCTAA: A Reproducible Corpus for Chinese Authorship Attribution Research
Haining Wang | Allen Riddell
Proceedings of the Thirteenth Language Resources and Evaluation Conference

Authorship attribution infers the likely author of an unsigned, single-authored document from a pool of candidates. Despite recent advances, a lack of standard, reproducible testbeds for Chinese language documents impedes progress. In this paper, we present the Chinese Cross-Topic Authorship Attribution (CCTAA) corpus. It is the first standard testbed for authorship attribution on contemporary Chinese prose. The cross-topic design and relatively inflexible genre of newswire contribute to an appropriate level of difficulty. It supports reproducible research by using pre-defined data splits. We show that a sequence classifier based on pre-trained Chinese RoBERTa embedding and a support vector machine classifier using function character n-gram frequency features perform below expectations on this task. The code for generating the corpus and reproducing the baselines is freely available at https://codeberg.org/haining/cctaa.


pdf bib
A Call for Clarity in Contemporary Authorship Attribution Evaluation
Allen Riddell | Haining Wang | Patrick Juola
Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP 2021)

Recent research has documented that results reported in frequently-cited authorship attribution papers are difficult to reproduce. Inaccessible code and data are often proposed as factors which block successful reproductions. Even when original materials are available, problems remain which prevent researchers from comparing the effectiveness of different methods. To solve the remaining problems—the lack of fixed test sets and the use of inappropriately homogeneous corpora—our paper contributes materials for five closed-set authorship identification experiments. The five experiments feature texts from 106 distinct authors. Experiments involve a range of contemporary non-fiction American English prose. These experiments provide the foundation for comparable and reproducible authorship attribution research involving contemporary writing.

pdf bib
Mode Effects’ Challenge to Authorship Attribution
Haining Wang | Allen Riddell | Patrick Juola
Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume

The success of authorship attribution relies on the presence of linguistic features specific to individual authors. There is, however, limited research assessing to what extent authorial style remains constant when individuals switch from one writing modality to another. We measure the effect of writing mode on writing style in the context of authorship attribution research using a corpus of documents composed online (in a web browser) and documents composed offline using a traditional word processor. The results confirm the existence of a “mode effect” on authorial style. Online writing differs systematically from offline writing in terms of sentence length, word use, readability, and certain part-of-speech ratios. These findings have implications for research design and feature engineering in authorship attribution studies.