83 citations · 86 across the 6 of their papers we have counts for
6 papers
The Master Key Hypothesis: Unlocking Cross-Model Capability Transfer via Linear Subspace Alignment
Rishab Balasubramanian, Pin-Jie Lin, Rituraj Sharma +6
We investigate whether post-trained capabilities can be transferred across models without retraining, with a focus on transfer across different model scales. We propose the Master…
PHLoRA: data-free Post-hoc Low-Rank Adapter extraction from full-rank checkpoint
Bhoomit Vasani, Jack FitzGerald, Anjie Fang +1
We introduce PHLoRA (Pronounced "flora"). (Post-hoc LoRA), a simple yet powerful method to extract low-rank adaptation adapters from full-rank fine-tuned models without requiring a…
Follow-on Question Suggestion via Voice Hints for Voice Assistants
Besnik Fetahu, Pedro Faustini, Giuseppe Castellucci +3
The adoption of voice assistants like Alexa or Siri has grown rapidly, allowing users to instantly access information via voice search. Query suggestion is a standard feature of sc…
HTEC: Human Transcription Error Correction
Hanbo Sun, Jian Gao, Xiaomin Wu +3
High-quality human transcription is essential for training and improving Automatic Speech Recognition (ASR) models. Recent study~\cite{libricrowd} has found that every 1% worse tra…
Reinforced Question Rewriting for Conversational Question Answering
Zhiyu Chen, Jie Zhao, Anjie Fang +3
Conversational Question Answering (CQA) aims to answer questions contained within dialogues, which are not easily interpretable without context. Developing a model to rewrite conve…
MultiCoNER: A Large-scale Multilingual dataset for Complex Named Entity Recognition
Shervin Malmasi, Anjie Fang, Besnik Fetahu +2
We present MultiCoNER, a large multilingual dataset for Named Entity Recognition that covers 3 domains (Wiki sentences, questions, and search queries) across 11 languages, as well…