A wide variety of NLP applications, such as machine translation, summarization, and dialog, involve text generation. One major challenge for these applications is how to evaluate whether such generated texts are actually fluent, accurate, or effective. In this work, we conceptualize the evaluation of generated text as a text generation problem, modeled using pre-trained sequence-to-sequence models. The general idea is that models trained to convert the generated text to/from a reference output or the source text will achieve higher scores when the generated text is better. We operationalize this idea using BART, an encoder-decoder based pre-trained model, and propose a metric BARTScore with a number of variants that can be flexibly applied in an unsupervised fashion to evaluation of text from different perspectives (e.g.

While continuous diffusion models excel in modeling continuous distributions, their application to categorical data has been less effective. Recent work has shown that ratio-matching through score-entropy within a continuous-time discrete Markov chain (CTMC) framework serves as a competitive alternative to autoregressive models in language modeling. To enhance this framework, we first introduce three new theorems concerning the KL divergence between the data and learned distribution. Our results serve as the discrete counterpart to those established for continuous diffusion models and allow us to derive an improved upper bound of the perplexity. Second, we empirically show that ratio-matching performed by minimizing the denoising cross-entropy between the clean and corrupted data enables models to outperform those utilizing score-entropy with up to 10% lower perplexity/generative-perplexity, and 15% faster training steps. To further support our findings, we introduce and evaluate a novel CTMC transition-rate matrix that allows prediction refinement, and derive the analytic expression for its matrix exponential which facilitates the computation of conditional ratios thus enabling efficient training and generation.

A wide variety of NLP applications, such as machine translation, summarization, and dialog, involve text generation. One major challenge for these applications is how to evaluate whether such generated texts are actually fluent, accurate, or effective. In this work, we conceptualize the evaluation of generated text as a text generation problem, modeled using pre-trained sequence-to-sequence models. The general idea is that models trained to convert the generated text to/from a reference output or the source text will achieve higher scores when the generated text is better. We operationalize this idea using BART, an encoder-decoder based pre-trained model, and propose a metric BARTScore with a number of variants that can be flexibly applied in an unsupervised fashion to evaluation of text from different perspectives (e.g.

As AI advances in text generation, human trust in AI generated content remains constrained by biases that go beyond concerns of accuracy. This study explores how bias shapes the perception of AI versus human generated content. Through three experiments involving text rephrasing, news article summarization, and persuasive writing, we investigated how human raters respond to labeled and unlabeled content. While the raters could not differentiate the two types of texts in the blind test, they overwhelmingly favored content labeled as "Human Generated," over those labeled "AI Generated," by a preference score of over 30%. We observed the same pattern even when the labels were deliberately swapped. This human bias against AI has broader societal and cognitive implications, as it undervalues AI performance. This study highlights the limitations of human judgment in interacting with AI and offers a foundation for improving human-AI collaboration, especially in creative fields.

In-Context Learning (ICL) has been a powerful emergent property of large language models that has attracted increasing attention in recent years. In contrast to regular gradient-based learning, ICL is highly interpretable and does not require parameter updates. In this paper, we show that, for linearized transformer networks, ICL can be made explicit and permanent through the inclusion of bias terms. We mathematically demonstrate the equivalence between a model with ICL demonstration prompts and the same model with the additional bias terms. Our algorithm (ICLCA) allows for exact conversion in an inexpensive manner. Existing methods are not exact and require expensive parameter updates. We demonstrate the efficacy of our approach through experiments that show the exact incorporation of ICL tokens into a linear transformer. We further suggest how our method can be adapted to achieve cheap approximate conversion of ICL tokens, even in regular transformer networks that are not linearized. Our experiments on GPT-2 show that, even though the conversion is only approximate, the model still gains valuable context from the included bias terms.

Coherence is a linguistic term that refers to the relations between small textual units (sentences, propositions), which make the text logically consistent and meaningful to the reader. With the advances of generative foundational models in NLP, there is a pressing need to automatically assess the human-perceived coherence of automatically generated texts. Up until now, little work has been done on explicitly assessing the coherence of generated texts and analyzing the factors contributing to (in)coherence. Previous work on the topic used other tasks, e.g., sentence reordering, as proxies of coherence, rather than approaching coherence detection heads on. In this paper, we introduce {\sc CoheSentia}, a novel benchmark of human-perceived coherence of automatically generated texts. Our annotation protocol reflects two perspectives; one is global, assigning a single coherence score, and the other is incremental, scoring sentence by sentence. The incremental method produces an (in)coherence score for each text fragment and also pinpoints reasons for incoherence at that point. Our benchmark contains 500 automatically-generated and human-annotated paragraphs, each annotated in both methods, by multiple raters. Our analysis shows that the inter-annotator agreement in the incremental mode is higher than in the holistic alternative, and our experiments show that standard LMs fine-tuned for coherence detection show varied performance on the different factors contributing to (in)coherence. All in all, these models yield unsatisfactory performance, emphasizing the need for developing more reliable methods for coherence assessment.

ChatGPT is a conversational artificial intelligence that is a member of the generative pre-trained transformer of the large language model family. This text generative model was fine-tuned by both supervised learning and reinforcement learning so that it can produce text documents that seem to be written by natural intelligence. Although there are numerous advantages of this generative model, it comes with some reasonable concerns as well. This paper presents a machine learning-based solution that can identify the ChatGPT delivered text from the human written text along with the comparative analysis of a total of 11 machine learning and deep learning algorithms in the classification process. We have tested the proposed model on a Kaggle dataset consisting of 10,000 texts out of which 5,204 texts were written by humans and collected from news and social media. On the corpus generated by GPT-3.5, the proposed algorithm presents an accuracy of 77%.

On AI changing education and the threat of AI plagiarism: "We're going to try and do some things in the short term. There may be ways we can help teachers be a little bit more likely to detect output of a GPT-like system, but a determined person will get around them, and I don't think it'll be something society can or should rely on long term. Generated text is something we all need to adapt to, and that's fine. We adapted to calculators and changed what we tested in maths class, I imagine. This is a more extreme version of that, no doubt. But also the benefits of it are more extreme as well."

Evaluation practices in natural language generation (NLG) have many known flaws, but improved evaluation approaches are rarely widely adopted. This issue has become more urgent, since neural NLG models have improved to the point where they can often no longer be distinguished based on the surface-level features that older metrics rely on. This paper surveys the issues with human and automatic model evaluations and with commonly used datasets in NLG that have been pointed out over the past 20 years. We summarize, categorize, and discuss how researchers have been addressing these issues and what their findings mean for the current state of model evaluations. Building on those insights, we lay out a long-term vision for NLG evaluation and propose concrete steps for researchers to improve their evaluation processes. Finally, we analyze 66 NLG papers from recent NLP conferences in how well they already follow these suggestions and identify which areas require more drastic changes to the status quo.

But what would AI be most concerned about when thinking of its future? What technology wants is a perplexing question ever since it has been asked directly by Kevin Kelly. Yet even before him, many have pondered over the answer to this question, that only the most self-absorbed species could ask of their imperfect creation. But is technology truly our creation? What is the essence of technology?