Simran Kaur

5 papers · 2021–2025 · 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 🌉 Interdisciplinary Bridge · Bronze 🗺️ Taxonomy Completionist (12) ❓ The Questioner

Venues

ICLR (4) NeurIPS (1)

Top co-authors

Anirudh Goyal (3) Sanjeev Arora (3) Dingli Yu (2) Arushi Gupta (1) Haoyu Zhao (1) Zachary Novack (1) Yuanzhi Li (1) Tanya Marwah (1) Simon Park (1) Ameet Talwalkar (1)

Research topics

Machine Learning (4) Optimization & Theory · Machine Learning (3) Artificial Intelligence (2) Natural Language Processing (2) Core AI (2) Generation (2) Deep Learning (1) Application Areas (1) Optimization & Theory · Deep Learning (1) Resources & Methods (1)

Keywords

stochastic gradient descent (1) instruction following (1) compositional generalization (1) instruction tuning (1) gradient descent (1) fisher information matrix (1) hessian eigenvalue (1) implicit regularization (1) language model (1) gradient norm (1) model fine-tuning (1) supervised fine-tuning (1) generalization gap (1) skill composition (1) edge of stability (1) data generation (1) emergent behavior (1) skill extraction (1) neural network (1) stochastic parrot (1)

Papers

Instruct-SkillMix: A Powerful Pipeline for LLM Instruction Tuning ICLR 2025 Can Models Learn Skill Composition from Examples? NeurIPS 2024 SKILL-MIX: a Flexible and Expandable Family of Evaluations for AI Models ICLR 2024 Disentangling the Mechanisms Behind Implicit Regularization in SGD ICLR 2023 Gradient Descent on Neural Networks Typically Occurs at the Edge of Stability ICLR 2021