Scholar
Seyedarmin Azizi
Google Scholar ID: kONzybkAAAAJ
University of Southern California
Machine Learning
Efficient Deep Learning
LLMs
Model Compression
Follow
Homepage
↗
Google Scholar
↗
Citations & Impact
All-time
Citations
92
H-index
5
i10-index
4
Publications
19
Co-authors
0
Publications
12 items
One Word Opens the Gate: The Option-Channel Attack on Typed Decision Models as Agent Guardrails
2026
Cited
0
Sharpen Without Search: On-Policy Distillation of Sequence-Level Power Distribution
2026
Cited
0
Labels Override Definitions in Jev-Style Typed Decision Models
2026
Cited
0
Chopthin-Consensus Power Sampling: A Diversity-Preserving Approach to LLM Decoding
2026
Cited
0
CircuitSteer: Geometrically Aligned Multi-Layer Steering via Sparse Autoencoder Circuits
2026
Cited
0
One Token Away from Collapse: The Fragility of Instruction-Tuned Helpfulness
2026
Cited
0
Power-SMC: Low-Latency Sequence-Level Power Sampling for Training-Free LLM Reasoning
2026
Cited
0
SkipKV: Selective Skipping of KV Generation and Storage for Efficient Inference with Large Reasoning Models
2025
Cited
0
Load more
