Mohammad Mohammadi Amiri

Assistant Professor@Rensselaer Polytechnic Institute·美国·OpenReview

研究方向

Federated learning · Supervised machine learning · Distributed machine learning · Distributed computing · Information and coding theory · Data valuation · Large language models

5.5

Scalable, Explainable and Provably Robust Anomaly Detection with One-Step Flow Matching

NeurIPS 2025Poster

6.3

SentenceKV: Efficient LLM Inference via Sentence-Level Semantic KV Caching

COLM 2025Poster

通讯

5.0

Sparse Gradient Compression for Fine-Tuning Large Language Models

ICLR 2025Withdrawn

二作

3.0

Latent Diffusion with LLMs for Reasoning

ICLR 2025Rejected

二作

2.0

Language Models for Textual Data Valuation

合作者 (20)

Bryan Kian Hsiang Low

合作者1 篇

Mohammad Mohammadi Amiri

OFMU: OPTIMIZATION-DRIVEN FRAMEWORK FOR MACHINE UNLEARNING

Data Diversity for Compositional Generalization

A Median Perspective on Unlabeled Data for Out-of-Distribution Detection

Towards Reversible Model Merging For Low-rank Weights

OjaKV: Context-Aware Online Low-Rank KV Cache Compression with Oja’s Rule

Scalable, Explainable and Provably Robust Anomaly Detection with One-Step Flow Matching

SentenceKV: Efficient LLM Inference via Sentence-Level Semantic KV Caching

Sparse Gradient Compression for Fine-Tuning Large Language Models

Latent Diffusion with LLMs for Reasoning

Language Models for Textual Data Valuation