ULTra: Unveiling Latent Token Interpretability in Transformer Based Understandingby Hesam Hosseini, Ghazal Hosseini Mighan, Amirabbas…
Provable unlearning in topic modeling and downstream tasksby Stanley Wei, Sadhika Malladi, Sanjeev Arora, Amartya…
Mechanism and Emergence of Stacked Attention Heads in Multi-Layer Transformersby Tiberiu MusatFirst submitted to arxiv…
SeqProFT: Applying LoRA Finetuning for Sequence-only Protein Property Predictionsby Shuo Zhang, Jian K. LiuFirst submitted…
CLMIA: Membership Inference Attacks via Unsupervised Contrastive Learningby Depeng Chen, Xiao Liu, Jie Cui, Hong…
Multi-Modal Self-Supervised Learning for Surgical Feedback Effectiveness Assessmentby Arushi Gupta, Rafal Kocielnik, Jiayun Wang, Firdavs…
Towards a framework on tabular synthetic data generation: a minimalist approach: theory, use cases, and…
HELENE: Hessian Layer-wise Clipping and Gradient Annealing for Accelerating Fine-tuning LLM with Zeroth-order Optimizationby Huaqin…
Efficient Alignment of Large Language Models via Data Samplingby Amrit Khera, Rajat Ghosh, Debojyoti DuttaFirst…
Hateful Meme Detection through Context-Sensitive Prompting and Fine-Grained Labelingby Rongxin Ouyang, Kokil Jaidka, Subhayan Mukerjee,…