Papers with compounding
Old is Gold: Linguistic Driven Approach for Entity and Relation Linking of Short Text (N19-1)
Copied to clipboard
Ahmad Sakor, Isaiah Onando Mulang’, Kuldeep Singh, Saeedeh Shekarpour, Maria Esther Vidal, Jens Lehmann, Sören Auer
| Challenge: | Short texts challenge NLP tasks because they lack context or are partially malformed. |
| Approach: | They propose a method which maps entities and relations within a short text to Wikipedia mentions. |
| Outcome: | The proposed approach outperforms state-of-the-art methods for short text query inventories. |
Through the Valley: Path to Effective Long CoT Training for Small Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Long chain-of-thought (CoT) supervision is effective for large language models . but small models trained on limited long CoT data experience performance degradation . |
| Approach: | They identify a phenomenon called Long CoT Degradation in small language models . long CoT data can be used to generate long chain-of-thought (CoT) responses . |
| Outcome: | The results show that models trained on 8k long CoT examples lose up to 75% of their original performance before fine-tuning. |
A Novel Multi-Task Learning Approach for Context-Sensitive Compound Type Identification in Sanskrit (2022.coling-1)
Copied to clipboard
Jivnesh Sandhan, Ashish Gupta, Hrishikesh Terdalkar, Tushar Sandhan, Suvendu Samanta, Laxmidhar Behera, Pawan Goyal
| Challenge: | Previously, compounding is a problem of identifying semantic relations between components of a word. |
| Approach: | They propose a multi-task learning architecture which incorporates contextual information and enriches syntactic information using morphological tagging and dependency parsing as auxiliary tasks. |
| Outcome: | The proposed architecture shows 6.1 points accuracy and 7.7 points (F1-score) absolute gain in English and Marathi languages. |
Massively Translingual Compound Analysis and Translation Discovery (L18-1)
Copied to clipboard
| Challenge: | Morphological compounding is one of the most common and productive methods of word formation across the world's languages. |
| Approach: | They propose a model for compounding using bilingual dictionaries and no annotated training data . they also release a massively multilingual dataset of compound words and their decompositions . |
| Outcome: | The proposed model generates novel translations of English concepts on a multilingual dataset . the model can be applied to a wide range of languages and is highly reproducible. |
An Annotated Social Media Corpus for German (2020.lrec-1)
Copied to clipboard
| Challenge: | Hate Speech (HS) against ethnic, religious and national minorities is a growing concern in online discourse. |
| Approach: | They present the German Twitter section of a large (2 billion word) bilingual Social Media corpus for Hate Speech research. |
| Outcome: | The proposed parser achieved F-scores of 97% for morphology and 92% for syntax on a cross-section of tweets. |