Beemo: Benchmark of Expert-edited Machine-generated Outputs

The rapid proliferation of large language models (LLMs) has increased the volume of machine-generated texts (MGTs) and blurred text authorship in various domains. However, most existing MGT benchmarks include single-author texts (human-written and machine-generated). This conventional design fails to capture more practical multi-author scenarios, where the user refines the LLM response for natural flow, coherence, and factual correctness. Our paper introduces the Benchmark of Expert-edited Machine-generated Outputs (Beemo), which includes 6.5k texts written by humans, generated by ten instruction-finetuned LLMs, and edited by experts for various use cases, ranging from creative writing to summarization. Beemo additionally comprises 13.1k machine-generated and LLM-edited texts, allowing for diverse MGT detection evaluation across various edit types. We document Beemo's creation protocol and present the results of benchmarking 33 configurations of MGT detectors in different experimental setups. We find that expert-based editing evades MGT detection, while LLM-edited texts are unlikely to be recognized as human-written. Beemo and all materials are publicly available.

CollabStory: Multi-LLM Collaborative Story Generation and Authorship Analysis

The rise of unifying frameworks that enable seamless interoperability of Large Language Models (LLMs) has made LLM-LLM collaboration for open-ended tasks a possibility. Despite this, there have not been efforts to explore such collaborative writing. We take the next step beyond human-LLM collaboration to explore this multi-LLM scenario by generating the first exclusively LLM-generated collaborative stories dataset called CollabStory. We focus on single-author ($N=1$) to multi-author (up to $N=5$) scenarios, where multiple LLMs co-author stories. We generate over 32k stories using open-source instruction-tuned LLMs. Further, we take inspiration from the PAN tasks that have set the standard for human-human multi-author writing tasks and analysis. We extend their authorship-related tasks for multi-LLM settings and present baselines for LLM-LLM collaboration. We find that current baselines are not able to handle this emerging scenario. Thus, CollabStory is a resource that could help propel an understanding as well as the development of techniques to discern the use of multiple LLMs. This is crucial to study in the context of writing tasks since LLM-LLM collaboration could potentially overwhelm ongoing challenges related to plagiarism detection, credit assignment, maintaining academic integrity in educational settings, and addressing copyright infringement concerns. We make our dataset and code available at \texttt{\url{https://github.com/saranya-venkatraman/multi_llm_story_writing}}.

ALISON: Fast and Effective Stylometric Authorship Obfuscation

Authorship Attribution (AA) and Authorship Obfuscation (AO) are two competing tasks of increasing importance in privacy research. Modern AA leverages an author's consistent writing style to match a text to its author using an AA classifier. AO is the corresponding adversarial task, aiming to modify a text in such a way that its semantics are preserved, yet an AA model cannot correctly infer its authorship. To address privacy concerns raised by state-of-the-art (SOTA) AA methods, new AO methods have been proposed but remain largely impractical to use due to their prohibitively slow training and obfuscation speed, often taking hours. To this challenge, we propose a practical AO method, ALISON, that (1) dramatically reduces training/obfuscation time, demonstrating more than 10x faster obfuscation than SOTA AO methods, (2) achieves better obfuscation success through attacking three transformer-based AA methods on two benchmark datasets, typically performing 15% better than competing methods, (3) does not require direct signals from a target AA classifier during obfuscation, and (4) utilizes unique stylometric features, allowing sound model interpretation for explainable obfuscation. We also demonstrate that ALISON can effectively prevent four SOTA AA methods from accurately determining the authorship of ChatGPT-generated texts, all while minimally changing the original text semantics. To ensure the reproducibility of our findings, our code and data are available at: https://github.com/EricX003/ALISON.

A Ship of Theseus: Curious Cases of Paraphrasing in LLM-Generated Texts

In the realm of text manipulation and linguistic transformation, the question of authorship has always been a subject of fascination and philosophical inquiry. Much like the \textbf{Ship of Theseus paradox}, which ponders whether a ship remains the same when each of its original planks is replaced, our research delves into an intriguing question: \textit{Does a text retain its original authorship when it undergoes numerous paraphrasing iterations?} Specifically, since Large Language Models (LLMs) have demonstrated remarkable proficiency in the generation of both original content and the modification of human-authored texts, a pivotal question emerges concerning the determination of authorship in instances where LLMs or similar paraphrasing tools are employed to rephrase the text. This inquiry revolves around \textit{whether authorship should be attributed to the original human author or the AI-powered tool, given the tool's independent capacity to produce text that closely resembles human-generated content.} Therefore, we embark on a philosophical voyage through the seas of language and authorship to unravel this intricate puzzle.

The Sentiment Problem: A Critical Survey towards Deconstructing Sentiment Analysis

We conduct an inquiry into the sociotechnical aspects of sentiment analysis (SA) by critically examining 189 peer-reviewed papers on their applications, models, and datasets. Our investigation stems from the recognition that SA has become an integral component of diverse sociotechnical systems, exerting influence on both social and technical users. By delving into sociological and technological literature on sentiment, we unveil distinct conceptualizations of this term in domains such as finance, government, and medicine. Our study exposes a lack of explicit definitions and frameworks for characterizing sentiment, resulting in potential challenges and biases. To tackle this issue, we propose an ethics sheet encompassing critical inquiries to guide practitioners in ensuring equitable utilization of SA. Our findings underscore the significance of adopting an interdisciplinary approach to defining sentiment in SA and offer a pragmatic solution for its implementation.

GPT-who: An Information Density-based Machine-Generated Text Detector

The Uniform Information Density principle posits that humans prefer to spread information evenly during language production. In this work, we examine if the UID principle can help capture differences between Large Language Models (LLMs) and human-generated text. We propose GPT-who, the first psycholinguistically-aware multi-class domain-agnostic statistical-based detector. This detector employs UID-based features to model the unique statistical signature of each LLM and human author for accurate authorship attribution. We evaluate our method using 4 large-scale benchmark datasets and find that GPT-who outperforms state-of-the-art detectors (both statistical- & non-statistical-based) such as GLTR, GPTZero, OpenAI detector, and ZeroGPT by over $20$% across domains. In addition to superior performance, it is computationally inexpensive and utilizes an interpretable representation of text articles. We present the largest analysis of the UID-based representations of human and machine-generated texts (over 400k articles) to demonstrate how authors distribute information differently, and in ways that enable their detection using an off-the-shelf LM without any fine-tuning. We find that GPT-who can distinguish texts generated by very sophisticated LLMs, even when the overlying text is indiscernible.

How do decoding algorithms distribute information in dialogue responses?

Humans tend to follow the Uniform Information Density (UID) principle by distributing information evenly in utterances. We study if decoding algorithms implicitly follow this UID principle, and under what conditions adherence to UID might be desirable for dialogue generation. We generate responses using different decoding algorithms with GPT-2 on the Persona-Chat dataset and collect human judgments on their quality using Amazon Mechanical Turk. We find that (i) surprisingly, model-generated responses follow the UID principle to a greater extent than human responses, and (ii) decoding algorithms that promote UID do not generate higher-quality responses. Instead, when we control for surprisal, non-uniformity of information density correlates with the quality of responses with very low/high surprisal. Our findings indicate that encouraging non-uniform responses is a potential solution to the ``likelihood trap'' problem (quality degradation in very high-likelihood text). Our dataset containing multiple candidate responses per dialog history along with human-annotated quality ratings is available at https://huggingface.co/datasets/saranya132/dialog_uid_gpt2.