◍wovepaper
SearchResearchersInstitutions
Sign in
researcher

Mark Rofin

EPFL

4 papers hereh-index 4126 citations7 works total

Matching runs newest-first, so older work may not be attached to this profile yet.

author position
  • first author2
  • middle author2

Across the 4 of 4 papers where every author was matched, so the position is known.

fields
  • cs.LG4
affiliations
  • EPFL
Homepage

identity via Semantic Scholar / OpenAlex

collaborators

4 papers

cs.LG2026

(How) Learning Rates Regulate Catastrophic Overtraining

Mark Rofin, Aditya Varre, Nicolas Flammarion

Supervised fine-tuning (SFT) is a common first stage of LLM post-training, teaching the model to follow instructions and shaping its behavior as a helpful assistant. At the same ti…

cs.LG2026

Understanding the Emergence of Seemingly Useless Features in Next-Token Predictors

Mark Rofin, Jalal Naghiyev, Michael Hahn

Trained Transformers have been shown to compute abstract features that appear redundant for predicting the immediate next token. We identify which components of the gradient signal…

cs.LG2026

Gradient Flow Polarizes Softmax Outputs towards Low-Entropy Solutions

Aditya Varre, Mark Rofin, Nicolas Flammarion

Understanding the intricate non-convex training dynamics of softmax-based models is crucial for explaining the empirical success of transformers. In this article, we analyze the gr…

cs.LG2025

Lower Bounds for Chain-of-Thought Reasoning in Hard-Attention Transformers

Alireza Amiri, Xinting Huang, Mark Rofin +1

Chain-of-thought reasoning and scratchpads have emerged as critical tools for enhancing the computational capabilities of transformers. While theoretical results show that polynomi…

◍wovepaper

Papers, researchers and institutions, woven together.

Explore
  • Search
  • Researchers
  • Institutions
Account
  • Library
  • Chat
Data
  • arXiv.org
  • Semantic Scholar
  • OpenAlex
  • Latest RSS
AboutContactPrivacyDevelopersllms.txtopenapi.json
Not affiliated with arXiv. Researcher data from Semantic Scholar (ODC-BY) and OpenAlex.