Yuval Ran-Milo

4 papers · 2024–2026 · 2 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓

🌍 Conference Polyglot · Bronze 🗺️ Taxonomy Completionist (10) 🏆 Keyword Champion (2)

Venues

ACL (3) NeurIPS (1)

Top co-authors

Raja Giryes (2) Itay Tshuva (1) Amir Globerson (1) Nir Endy (1) Yonatan Slutzky (1) Edo Cohen-Karlik (1) Shahar Mendel (1) Nadav Cohen (1) Eden Lumbroso (1) Idan Daniel Grosbard (1)

Research topics

Deep Learning (4) Architectures (4) Artificial Intelligence (3) Core AI (3) Machine Learning (1) Models (1) Optimization & Theory · Deep Learning (1) Optimization & Theory · Machine Learning (1)

Keywords

attention sink (2) state space model (2) factual knowledge (1) positional encoding (1) self-attention model (1) causal intervention (1) information flow (1) layer-wise dynamics (1) complex parameterization (1) diagonal structure (1) softmax transformer (1) trigger-conditional task (1) relu attention (1) learning theory (1) query bia (1) attention mechanism (1)

Papers

A Mechanistic Account of Attention Sinks in GPT-2: One Circuit, Broader Implications for Mitigation ACL 2026 Attention Sinks Are Provably Necessary in Softmax Transformers: Evidence from Trigger-Conditional Tasks ACL 2026 Mamba Knockout for Unraveling Factual Information Flow ACL 2025 Provable Benefits of Complex Parameterizations for Structured State Space Models NeurIPS 2024