Trusting CHATGPT: how minor tweaks in the prompts lead to major differences in sentiment classification

• URL: http://arxiv.org/abs/2504.12180v1
• Date: Wed, 16 Apr 2025 15:37:09 GMT
• Title: Trusting CHATGPT: how minor tweaks in the prompts lead to major differences in sentiment classification
• Abstract summary: This study tests the hypothesis that subtle changes in the structure of prompts do not produce significant variations in the classification results of sentiment polarity analysis.<n>Using a dataset of 100.000 comments in Spanish on four Latin American presidents, the model classified the comments as positive, negative, or neutral on 10 occasions, varying the prompts slightly each time.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: One fundamental question for the social sciences today is: how much can we trust highly complex predictive models like ChatGPT? This study tests the hypothesis that subtle changes in the structure of prompts do not produce significant variations in the classification results of sentiment polarity analysis generated by the Large Language Model GPT-4o mini. Using a dataset of 100.000 comments in Spanish on four Latin American presidents, the model classified the comments as positive, negative, or neutral on 10 occasions, varying the prompts slightly each time. The experimental methodology included exploratory and confirmatory analyses to identify significant discrepancies among classifications. The results reveal that even minor modifications to prompts such as lexical, syntactic, or modal changes, or even their lack of structure impact the classifications. In certain cases, the model produced inconsistent responses, such as mixing categories, providing unsolicited explanations, or using languages other than Spanish. Statistical analysis using Chi-square tests confirmed significant differences in most comparisons between prompts, except in one case where linguistic structures were highly similar. These findings challenge the robustness and trust of Large Language Models for classification tasks, highlighting their vulnerability to variations in instructions. Moreover, it was evident that the lack of structured grammar in prompts increases the frequency of hallucinations. The discussion underscores that trust in Large Language Models is based not only on technical performance but also on the social and institutional relationships underpinning their use.

Related papers

• Inference-Time Steering for Cross-Lingual Factual Consistency in LLMs [51.56484100374058]
  Large Language Models (LLMs) demonstrate remarkable multilingual fluency, but their internal knowledge representations remain disproportionately biased toward high-resource languages.<n>We investigate whether these biases can be mitigated at inference time, forcing an English-prompted model to answer as if it were queried in target languages.<n>Experiments on Gemma 3 12B Instruct reveal persona prompting to be the strongest overall intervention, balancing efficacy, safety, and out-of-domain generalization.
  arXiv  Detail & Related papers  (2026-07-21T16:15:05Z)
• Type and Complexity Signals in Multilingual Question Representations [0.27930955543692815]
  We introduce the Question Type and Complexity dataset with sentences across seven languages.<n>We compare layer-wise probes on frozen Glot500-m representations against subword TF-IDF baselines, and a fine-tuned model.<n>Results show that statistical features classify questions effectively in languages with explicit marking, while neural probes capture fine-grained structural complexity patterns better.
  arXiv  Detail & Related papers  (2025-10-07T17:50:05Z)
• Tokenization and Representation Biases in Multilingual Models on Dialectal NLP Tasks [7.216732751280017]
  We correlate Tokenization Parity (TP) and Information Parity (IP) as measures of representational biases in pre-trained multilingual models.<n>We compare state-of-the-art decoder-only LLMs with encoder-based models across three tasks: dialect classification, topic classification, and extractive question answering.<n>Our analysis reveals that TP is a better predictor of the performance on tasks reliant on syntactic and morphological cues, while IP better predicts performance in semantic tasks.
  arXiv  Detail & Related papers  (2025-09-24T12:13:53Z)
• ExpliCa: Evaluating Explicit Causal Reasoning in Large Language Models [75.05436691700572]
  We introduce ExpliCa, a new dataset for evaluating Large Language Models (LLMs) in explicit causal reasoning. We tested seven commercial and open-source LLMs on ExpliCa through prompting and perplexity-based metrics. Surprisingly, models tend to confound temporal relations with causal ones, and their performance is also strongly influenced by the linguistic order of the events.
  arXiv  Detail & Related papers  (2025-02-21T14:23:14Z)
• Counterfactual Samples Constructing and Training for Commonsense Statements Estimation [17.970740197590693]
  Plausibility Estimation plays a crucial role for enabling language models to objectively comprehend the real world. They lack two key traits of an ideal PE model: language-explainable and commonsense-sensitive. We propose a novel model-agnostic method, referred to as Commonsense Counterfactual Samples Generating.
  arXiv  Detail & Related papers  (2024-12-29T20:18:52Z)
• The Lou Dataset -- Exploring the Impact of Gender-Fair Language in German Text Classification [57.06913662622832]
  Gender-fair language fosters inclusion by addressing all genders or using neutral forms. Gender-fair language substantially impacts predictions by flipping labels, reducing certainty, and altering attention patterns. While we offer initial insights on the effect on German text classification, the findings likely apply to other languages.
  arXiv  Detail & Related papers  (2024-09-26T15:08:17Z)
• How to Handle Different Types of Out-of-Distribution Scenarios in Computational Argumentation? A Comprehensive and Fine-Grained Field Study [59.13867562744973]
  This work systematically assesses LMs' capabilities for out-of-distribution (OOD) scenarios. We find that the efficacy of such learning paradigms varies with the type of OOD. Specifically, while ICL excels for domain shifts, prompt-based fine-tuning surpasses for topic shifts.
  arXiv  Detail & Related papers  (2023-09-15T11:15:47Z)
• Disco-Bench: A Discourse-Aware Evaluation Benchmark for Language Modelling [70.23876429382969]
  We propose a benchmark that can evaluate intra-sentence discourse properties across a diverse set of NLP tasks. Disco-Bench consists of 9 document-level testsets in the literature domain, which contain rich discourse phenomena. For linguistic analysis, we also design a diagnostic test suite that can examine whether the target models learn discourse knowledge.
  arXiv  Detail & Related papers  (2023-07-16T15:18:25Z)
• Simple Linguistic Inferences of Large Language Models (LLMs): Blind Spots and Blinds [59.71218039095155]
  We evaluate language understanding capacities on simple inference tasks that most humans find trivial. We target (i) grammatically-specified entailments, (ii) premises with evidential adverbs of uncertainty, and (iii) monotonicity entailments. The models exhibit moderate to low performance on these evaluation sets.
  arXiv  Detail & Related papers  (2023-05-24T06:41:09Z)
• Mitigating Bias in Text Classification via Prompt-Based Text Transformation [0.2864713389096699]
  We investigate whether prompting ChatGPT to rewrite text can reduce demographic signals while preserving meaning.<n>Results suggest that prompt-based rewriting offers a practical and generalisable approach for mitigating bias in text classification.
  arXiv  Detail & Related papers  (2023-05-09T13:10:23Z)
• Cross-Lingual Transfer of Cognitive Processing Complexity [11.939409227407769]
  We use sentence-level eye-tracking patterns as a cognitive indicator for structural complexity. We show that the multilingual model XLM-RoBERTa can successfully predict varied patterns for 13 typologically diverse languages.
  arXiv  Detail & Related papers  (2023-02-24T15:48:23Z)
• Prompting Large Language Model for Machine Translation: A Case Study [87.88120385000666]
  We offer a systematic study on prompting strategies for machine translation. We examine factors for prompt template and demonstration example selection. We explore the use of monolingual data and the feasibility of cross-lingual, cross-domain, and sentence-to-document transfer learning.
  arXiv  Detail & Related papers  (2023-01-17T18:32:06Z)
• Montague semantics and modifier consistency measurement in neural language models [13.187353456418204]
  This work proposes a novel methodology for measuring compositional behavior in contemporary language embedding models.<n> Specifically, we focus on adjectival modifier phenomena in adjective-noun phrases.
  arXiv  Detail & Related papers  (2022-10-10T18:43:16Z)
• Estimating Confidence of Predictions of Individual Classifiers and Their Ensembles for the Genre Classification Task [0.0]
  Genre identification is a subclass of non-topical text classification. Nerve models based on pre-trained transformers, such as BERT or XLM-RoBERTa, demonstrate SOTA results in many NLP tasks.
  arXiv  Detail & Related papers  (2022-06-15T09:59:05Z)
• More Than Words: Towards Better Quality Interpretations of Text Classifiers [16.66535643383862]
  We show that token-based interpretability, while being a convenient first choice given the input interfaces of the ML models, is not the most effective one in all situations. We show that higher-level feature attributions offer several advantages: 1) they are more robust as measured by the randomization tests, 2) they lead to lower variability when using approximation-based methods like SHAP, and 3) they are more intelligible to humans in situations where the linguistic coherence resides at a higher level.
  arXiv  Detail & Related papers  (2021-12-23T10:18:50Z)
• AES Systems Are Both Overstable And Oversensitive: Explaining Why And Proposing Defenses [66.49753193098356]
  We investigate the reason behind the surprising adversarial brittleness of scoring models. Our results indicate that autoscoring models, despite getting trained as "end-to-end" models, behave like bag-of-words models. We propose detection-based protection models that can detect oversensitivity and overstability causing samples with high accuracies.
  arXiv  Detail & Related papers  (2021-09-24T03:49:38Z)
• ABNIRML: Analyzing the Behavior of Neural IR Models [45.74073795558624]
  Pretrained language models such as BERT and T5 have established a new state-of-the-art for ad-hoc search. We present a new comprehensive framework for Analyzing the Behavior of Neural IR ModeLs (ABNIRML) We conduct an empirical study that yields insights into the factors that contribute to the neural model's gains.
  arXiv  Detail & Related papers  (2020-11-02T03:07:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.