Adaptive Originality Filtering: Rejection Based Prompting and RiddleScore for Culturally Grounded Multilingual Riddle Generation
Not on the map. Why?
Abstract (arXiv)
Language models are increasingly tested on multilingual creativity, demanding culturally grounded, abstract generations. Standard prompting methods often produce repetitive or shallow outputs. We introduce Adaptive Originality Filtering (AOF), a prompting strategy that enforces novelty and cultural fidelity via semantic rejection. To assess quality, we propose RiddleScore, a metric combining novelty, diversity, fluency, and answer alignment. AOF improves Distinct-2 (0.915 in Japanese), reduces Self-BLEU (0.177), and raises RiddleScore (up to +57.1% in Arabic). Human evaluations confirm fluency, creativity, and cultural fit gains. However, improvements vary: Arabic shows greater RiddleScore gains than Distinct-2; Japanese sees similar changes. Though focused on riddles, our method may apply to broader creative tasks. Overall, semantic filtering with composite evaluation offers a lightweight path to culturally rich generation without fine-tuning.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 32 of its 89 references that are in the map. First posted on arXiv 26 Aug 2025.
Builds on
32 of 89 references.
References · 32
- 2020Language Models are Few-Shot Learnerscited by 32,419
- 2019BERTScore: Evaluating Text Generation with BERTcited by 5,512
- 2022Large Language Models are Zero-Shot Reasonerscited by 4,374
- 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,276
- 2019Unsupervised Cross-lingual Representation Learning at Scalecited by 3,935
- 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,442
- 2020MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformerscited by 1,463
- 2020BLEURT: Learning Robust Metrics for Text Generationcited by 1,203
- 2018Hierarchical Neural Story Generationcited by 1,193
- 2019Cross-lingual Language Model Pretrainingcited by 1,090
- 2015A Diversity-Promoting Objective Function for Neural Conversation Modelscited by 1,077
- 2020COMET: A Neural Framework for MT Evaluationcited by 1,033
20 more
- 2019Universal Adversarial Triggers for Attacking and Analyzing NLPcited by 722
- 2022Using DeepSpeed and Megatron to Train Megatron-Turing NLG 530B, A Large-Scale Generative Language Modelcited by 559
- 2020XCOPA: A Multilingual Dataset for Causal Commonsense Reasoningcited by 333
- 2022COLD Decoding: Energy-based Constrained Text Generation with Langevin Dynamicscited by 161
- 2025Sketch-of-Thought: Efficient LLM Reasoning with Adaptive Cognitive-Inspired Sketchingcited by 117
- 2024Survey of Cultural Awareness in Language Models: Text and Beyondcited by 103
- 2020Trading Off Diversity and Quality in Natural Language Generationcited by 96
- 2022Testing the Ability of Language Models to Interpret Figurative Languagecited by 67
- 2018Controllable Neural Story Plot Generation via Reward Shapingcited by 49
- 2021MERMAID: Metaphor Generation with Symbolism and Discriminative Decodingcited by 49
- 2023Multi-lingual and Multi-cultural Figurative Language Understandingcited by 47
- 2025Diversity-Aware Policy Optimization for Large Language Model Reasoningcited by 41
- 2022Large Language Models and the Reverse Turing Testcited by 38
- 2024Knowledge Editing through Chain-of-Thoughtcited by 26
- 2024CS4: Measuring the Creativity of Large Language Models Automatically by Controlling the Number of Story-Writing Constraintscited by 24
- 2024Spotting AI's Touch: Identifying LLM-Paraphrased Spans in Textcited by 16
- 2022Human Judgement as a Compass to Navigate Automatic Metrics for Formality Transfercited by 12
- 2021BiRdQA: A Bilingual Dataset for Question Answering on Tricky Riddlescited by 10
- 2022CC-Riddle: A Question Answering Dataset of Chinese Character Riddlescited by 6
- 2024MetaMetrics-MT: Tuning Meta-Metrics for Machine Translation via Human Preference Calibrationcited by 6
Built on by
Citing papers per year: 2025: 1.
1 paper.