papers

Publications (11)

cs.LG2026

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…

cs.SD2023

In-Context Prompt Editing For Conditional Audio Generation

Ernie Chang, Pin-Jie Lin, Yang Li +6

Distributional shift is a central challenge in the deployment of machine learning models as they can be ill-equipped for real-world data. This is particularly evident in text-to-au…

cs.CL2025

Efficient Model Development through Fine-tuning Transfer

Pin-Jie Lin, Rishab Balasubramanian, Fengyuan Liu +2

Modern LLMs struggle with efficient updates, as each new pretrained model version requires repeating expensive alignment processes. This challenge also applies to domain- or langua…

cs.CL2023

Revisiting Sample Size Determination in Natural Language Understanding

Ernie Chang, Muhammad Hassan Rashid, Pin-Jie Lin +4

Knowing exactly how many data points need to be labeled to achieve a certain model performance is a hugely beneficial step towards reducing the overall budgets for annotation. It p…

cs.CL2025

Self-Vocabularizing Training for Neural Machine Translation

Pin-Jie Lin, Ernie Chang, Yangyang Shi +1

Past vocabulary learning techniques identify relevant vocabulary before training, relying on statistical and entropy-based assumptions that largely neglect the role of model traini…

cs.CL2023

Low-Resource Cross-Lingual Adaptive Training for Nigerian Pidgin

Pin-Jie Lin, Muhammed Saeed, Ernie Chang +1

Developing effective spoken language processing systems for low-resource languages poses several challenges due to the lack of parallel data and limited resources for fine-tuning m…

cs.CL2024

Scaling Parameter-Constrained Language Models with Quality Data

Ernie Chang, Matteo Paltenghi, Yang Li +7

Scaling laws in language modeling traditionally quantify training loss as a function of dataset size and model parameters, providing compute-optimal estimates but often neglecting…

cs.CL2024

Modeling Orthographic Variation Improves NLP Performance for Nigerian Pidgin

Pin-Jie Lin, Merel Scholman, Muhammed Saeed +1

Nigerian Pidgin is an English-derived contact language and is traditionally an oral language, spoken by approximately 100 million people. No orthographic standard has yet been adop…

cs.CL2024

Exploring the Effectiveness and Consistency of Task Selection in Intermediate-Task Transfer Learning

Pin-Jie Lin, Miaoran Zhang, Marius Mosbach +1

Identifying beneficial tasks to transfer from is a critical step toward successful intermediate-task transfer learning. In this work, we experiment with 130 source-target task comb…

cs.CL2024

Target-Aware Language Modeling via Granular Data Sampling

Ernie Chang, Pin-Jie Lin, Yang Li +6

Language model pretraining generally targets a broad range of use cases and incorporates data from diverse sources. However, there are instances where we desire a model that excels…

cs.SD2023

On The Open Prompt Challenge In Conditional Audio Generation

Ernie Chang, Sidd Srinivasan, Mahi Luthra +8

Text-to-audio generation (TTA) produces audio from a text description, learning from pairs of audio samples and hand-annotated text. However, commercializing audio generation is ch…