29 citations · 36 across the 14 of their papers we have counts for
6 papers · 1 filter
Long-Context Drafting Does Not Require Long-Context Drafters
Aref Jafari, Vansh Bhatia, Matt Qin +5
Speculative decoding pairs a cheap drafter with parallel verification against the target. At ultra-long context it stops paying, because a drafter that attends over the full prefix…
SCOUT: Toward Sub-Quadratic Attention via Segment Compression for Optimized Utility in Transformers
Aref Jafari, Yuhe Fan, Benyamin Jamialahmadi +3
Transformers have demonstrated strong performance across a wide range of sequence modeling tasks, but their quadratic attention complexity limits scalability to long sequences. Lin…
DTRNet: Dynamic Token Routing Network to Reduce Quadratic Costs in Transformers
Aman Sharma, Saeed Najafi, Parsa Farinneya +6
Transformers achieve state-of-the-art results across many tasks, but their uniform application of quadratic self-attention to every token at every layer makes them computationally…
Improved knowledge distillation by utilizing backward pass knowledge in neural networks
Aref Jafari, Mehdi Rezagholizadeh, Ali Ghodsi
Knowledge distillation (KD) is one of the prominent techniques for model compression. In this method, the knowledge of a large network (teacher) is distilled into a model (student)…
Continuation KD: Improved Knowledge Distillation through the Lens of Continuation Optimization
Aref Jafari, Ivan Kobyzev, Mehdi Rezagholizadeh +2
Knowledge Distillation (KD) has been extensively used for natural language understanding (NLU) tasks to improve a small model's (a student) generalization by transferring the knowl…
Do we need Label Regularization to Fine-tune Pre-trained Language Models?
Ivan Kobyzev, Aref Jafari, Mehdi Rezagholizadeh +5
Knowledge Distillation (KD) is a prominent neural model compression technique that heavily relies on teacher network predictions to guide the training of a student model. Consideri…