Likelihood-based Mitigation of Evaluation Bias in Large Language Models

Large Language Models (LLMs) are widely used to evaluate natural language generation tasks as automated metrics. However, the likelihood, a measure of LLM's plausibility for a sentence, can vary due to superficial differences in sentences, such as word order and sentence structure. It is therefore possible that there might be a likelihood bias if LLMs are used for evaluation: they might overrate sentences with higher likelihoods while underrating those with lower likelihoods. In this paper, we investigate the presence and impact of likelihood bias in LLM-based evaluators. We also propose a method to mitigate the likelihood bias. Our method utilizes highly biased instances as few-shot examples for in-context learning. Our experiments in evaluating the data-to-text and grammatical error correction tasks reveal that several LLMs we test display a likelihood bias. Furthermore, our proposed method successfully mitigates this bias, also improving evaluation performance (in terms of correlation of models with human scores) significantly.

How You Prompt Matters! Even Task-Oriented Constraints in Instructions Affect LLM-Generated Text Detection

Against the misuse (e.g., plagiarism or spreading misinformation) of Large Language Models (LLMs), many recent works have presented LLM-generated-text detectors with promising detection performance. Spotlighting a situation where users instruct LLMs to generate texts (e.g., essay writing), there are various ways to write the instruction (e.g., what task-oriented constraint to include). In this paper, we discover that even a task-oriented constraint in instruction can cause the inconsistent performance of current detectors to the generated texts. Specifically, we focus on student essay writing as a realistic domain and manually create the task-oriented constraint for each factor on essay quality by Ke and Ng (2019). Our experiment shows that the detection performance variance of the current detector on texts generated by instruction with each task-oriented constraint is up to 20 times larger than the variance caused by generating texts multiple times and paraphrasing the instruction. Our finding calls for further research on developing robust detectors that can detect such distributional shifts caused by a task-oriented constraint in the instruction.

OUTFOX: LLM-generated Essay Detection through In-context Learning with Adversarially Generated Examples

Large Language Models (LLMs) have achieved human-level fluency in text generation, making it difficult to distinguish between human-written and LLM-generated texts. This poses a growing risk of misuse of LLMs and demands the development of detectors to identify LLM-generated texts. However, existing detectors degrade detection accuracy by simply paraphrasing LLM-generated texts. Furthermore, the effectiveness of these detectors in real-life situations, such as when students use LLMs for writing homework assignments (e.g., essays) and quickly learn how to evade these detectors, has not been explored. In this paper, we propose OUTFOX, a novel framework that improves the robustness of LLM-generated-text detectors by allowing both the detector and the attacker to consider each other's output and apply this to the domain of student essays. In our framework, the attacker uses the detector's prediction labels as examples for in-context learning and adversarially generates essays that are harder to detect. While the detector uses the adversarially generated essays as examples for in-context learning to learn to detect essays from a strong attacker. Our experiments show that our proposed detector learned in-context from the attacker improves the detection performance on the attacked dataset by up to +41.3 point F1-score. While our proposed attacker can drastically degrade the performance of the detector by up to -57.0 point F1-score compared to the paraphrasing method.