Author

Yuval Ran-Milo

Observed names: Ran-Milo, Yuval, Yuval Ran-milo

Institution: Tel Aviv University

Observed institution: Tel Aviv University

ORCID: 0009-0006-8008-8170

Observed ORCID: 0009-0006-8008-8170

OpenAlex ID: A5114635302

A Generative AI-Empowered Digital Tutor for Higher Education Courses

2025 · Hila Reicher, Yarden Frenkel, et al. · Information · 8

Provable Benefits of Complex Parameterizations for Structured State Space Models

2024 · Yuval Ran-Milo, Eden Lumbroso, et al. · neural information processing systems · 1

PDF

A Memory Efficient Unified Algorithm for Online Learning of Linear Dynamical Systems

2026 · Ran-Milo, Yuval, Assos, Angelos, et al. · arXiv (Cornell University)

PDF

A Mechanistic Account of Attention Sinks in GPT-2: One Circuit, Broader Implications for Mitigation

2026 · Association for Computational Linguistics 2026, Mendel, Shahar, et al. · Underline Science Inc.

Attention Sinks Are Provably Necessary in Softmax Transformers: Evidence from Trigger-Conditional Tasks

2026 · Association for Computational Linguistics 2026, Ran-Milo, Yuval · Underline Science Inc.

A Mechanistic Account of Attention Sinks in GPT-2: One Circuit, Broader Implications for Mitigation

2026 · Ran-Milo, Yuval, Ofek, Hila, et al. · arXiv (Cornell University)

A Mechanistic Account of Attention Sinks in GPT-2: One Circuit, Broader Implications for Mitigation

2026 · Ran-Milo, Yuval, Ofek, Hila, et al. · arXiv (Cornell University)

PDF

Attention Sinks Are Provably Necessary in Softmax Transformers: Evidence from Trigger-Conditional Tasks

2026 · Ran-Milo, Yuval · arXiv (Cornell University)

Attention Sinks Are Provably Necessary in Softmax Transformers: Evidence from Trigger-Conditional Tasks

2026 · Ran-Milo, Yuval · arXiv (Cornell University)

PDF

Outcome-Based RL Provably Leads Transformers to Reason, but Only With the Right Data

2026 · Ran-Milo, Yuval, Alexander, Yotam, et al. · arXiv (Cornell University)

PDF

Attention Sinks Are Provably Necessary in Softmax Transformers: Evidence from Trigger-Conditional Tasks

2026 · Yuval Ran-Milo · Annual Meeting of the Association for Computational Linguistics (ACL)

PDF

A Mechanistic Account of Attention Sinks in GPT-2: One Circuit, Broader Implications for Mitigation

2026 · Yuval Ran-Milo, Hila Ofek, et al. · Annual Meeting of the Association for Computational Linguistics (ACL)

PDF

Mamba Knockout for Unraveling Factual Information Flow

2025 · Endy, Nir, Grosbard, Idan Daniel, et al. · arXiv (Cornell University)

PDF

Mamba Knockout for Unraveling Factual Information Flow

2025 · Nir Endy, Idan Daniel Grosbard, et al. · Annual Meeting of the Association for Computational Linguistics (ACL)

Do Neural Networks Need Gradient Descent to Generalize? A Theoretical Study

2025 · ‪Yotam Alexander‬‏, Yonatan Slutzky, et al. · neural information processing systems

PDF

Works count: 16

Citation count: 9

H-index: 1

I10-index: