Papers with decoder-only language models
BullyBench: Youth & Experts-in-the-loop Framework for Intrinsic and Extrinsic Cyberbullying NLP Benchmarking (2025.emnlp-industry)
Copied to clipboard
Kanishk Verma, Sri Balaaji, Joachim Wagner, Arefeh Kazemi, Darragh Mccashin, Isobel Walsh@dcu, Sayani Basak, Sinan Asci, Yelena Cherkasova, Alexandros Poulis, James Ohiggins Norman, Rebecca Umbach Umbach, Tijana Milosevic, Brian Davis
| Challenge: | Existing youth-focused CB datasets lack conversational realism and ethical youth involvement with little or no evaluation of their social plausibility. |
| Approach: | They propose a youth-in-the-loop dataset “BullyBench” that incorporates a structured intrinsic quality evaluation with experts-in the-looop (social scientists, psychologists, and content moderators) they perform extrinsic baseline evaluation by benchmarking encoder- and decoder-only language models for multi-class CB role classification. |
| Outcome: | The proposed dataset is evaluated by a team of social scientists, psychologists, and content moderators to assess its quality, relevance, and coherence. |
Evaluating the Capabilities of Large Language Models for Multi-label Emotion Understanding (2025.coling-main)
Copied to clipboard
Tadesse Destaw Belay, Israel Abebe Azime, Abinew Ali Ayele, Grigori Sidorov, Dietrich Klakow, Philip Slusallek, Olga Kolesnikova, Seid Muhie Yimam
| Challenge: | Emotion classification is one of the most challenging tasks in large language models. |
| Approach: | They propose to use a multi-label emotion classification dataset for four Ethiopian languages to evaluate their ability to learn and reason. |
| Outcome: | The proposed model improves the understanding of emotions in language models and how people convey emotions through various languages. |
DiffSkip: Differential Layer Skipping in Large Language Models (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing Large Language Models (LLMs) enforce uniform computation across all tokens. |
| Approach: | They propose to dynamically skip FFN blocks based on self-attention difference . they propose to use a lightweight router module to do the same . |
| Outcome: | The proposed model can skip FFN blocks even in token generation tasks. |
Embedded Named Entity Recognition using Probing Classifiers (2024.emnlp-main)
Copied to clipboard
| Challenge: | Streaming text generation requires separate models during inference, which increases computational cost, or destructive fine-tuning of the language model. |
| Approach: | They propose an approach which enables streaming named entity recognition in decoder-only language models without fine-tuning them. |
| Outcome: | The proposed approach maintains high token generation rates with only a negligible decrease in speed of around 1% compared to a baseline of 43.64%. |