NaCTeM

NaCTeM success at EMNLP 2025 - 7/7 papers accepted

2025-08-23

We are delighted to announce that all seven papers that were co-authored by PhD students at NaCTeM and submitted to EMNLP have been accepted for publication.

We congratulate our students Kailai Yang, Mohsinul Kabir, Zeping Yu, Xin Zhang, Polydoros Giannouris, Triantafillos Papadopoulos, Zhiwei Liu and Jimin Huang for their excellent achievements

We also thank our collaborators Qianqian Xie, Erxue Min, Efstathia Soufleri, Yonatan Belinkov, Xueqing Peng and Ruoyu Xiang for their contributions.

The accepted papers are as follows:

  1. From n-gram to Attention: How Model Architectures Learn and Propagate Bias in Language Modeling
  2. Break the Checkbox: Challenging Closed-Style Evaluations of Cultural Alignment in LLMs
  3. Back Attention: Understanding and Enhancing Multi-Hop Reasoning in Large Language Models
  4. Locate-then-Merge: Neuron-Level Parameter Fusion for Mitigating Catastrophic Forgetting in Multimodal LLMs
  5. THCM-CAL: Temporal-Hierarchical Causal Modelling with Conformal Calibration for Clinical Risk Prediction
  6. Selective Preference Optimization via Token-Level Reward Function Estimation
  7. Plutus: Benchmarking Large Language Models in Low-Resource Greek Finance in collaboration with ARCHIMEDES Research on AI, Data Science, and Algorithms

Next item
Back to news summary page