Soheil Abbasloo

Senior Researcher

Publications

View by:

- Bias Control and Injection for Multi-Agent Aggregation
  
  Sungjae Lee, Soheil Abbasloo, Jabin Koo, Jungseul Ok
  
  October 2026
  
  October 2026
  
  Publication
- Dynamic Expert Pruning for Multi-Agent Systems
  
  Jabin Koo, Soheil Abbasloo, Sungjae Lee, Jungseul Ok
  
  October 2026
  
  October 2026
  
  Publication
- Do AI Agents Understand Computer Architecture?
  
  Ambika Sharan, Grigory Chirkov, Soheil Abbasloo
  
  MLForSys | September 2026
  
  PDF
- CornerCase: Automated Extremal Testing of Protocol Implementations using LLMs
  
  Rathin Singha, Kuan Qian, S. Saikrishnan, Tracy Zhao, Soheil Abbasloo, Ryan Beckett, Siva Kesava Reddy Kakarla, Todd D. Millstein, George Varghese
  
  June 2026
  
  arXiv | June 2026
  
  Publication
- Evaluating LLM Reasoning Beyond Correctness and CoT
  
  Soheil Abbasloo
  
  February 2026
  
  February 2026
  
  Publication
- Beyond Correctness: Learning Robust Reasoning via Transfer
  
  Hyunseok Lee, Soheil Abbasloo, Jihoon Tack, Jinwoo Shin
  
  February 2026
  
  arXiv | February 2026
  
  Publication
- Mahak: An Automated and Efficient Assessment Framework for Internet Control Algorithms
  
  Parsa Pazhooheshy, Soheil Abbasloo, Y. Ganjali
  
  2025 IEEE 33rd International Conference on Network Protocols (ICNP) | September 2025, pp. 1-13
  
  Publication
- Are Language Models Up to Sequential Optimization Problems? From Evaluation to a Hegelian-Inspired Enhancement
  
  Soheil Abbasloo
  
  February 2025
  
  February 2025
  
  Publication

- Mahak: An Automated and Efficient Assessment Framework for Internet Control Algorithms
  
  Parsa Pazhooheshy, Soheil Abbasloo, Y. Ganjali
  
  2025 IEEE 33rd International Conference on Network Protocols (ICNP) | September 2025, pp. 1-13
  
  Publication
- CornerCase: Automated Extremal Testing of Protocol Implementations using LLMs
  
  Rathin Singha, Kuan Qian, S. Saikrishnan, Tracy Zhao, Soheil Abbasloo, Ryan Beckett, Siva Kesava Reddy Kakarla, Todd D. Millstein, George Varghese
  
  June 2026
  
  arXiv | June 2026
  
  Publication
- Mahak: An Automated and Efficient Assessment Framework for Internet Control Algorithms
  
  Parsa Pazhooheshy, Soheil Abbasloo, Y. Ganjali
  
  2025 IEEE 33rd International Conference on Network Protocols (ICNP) | September 2025, pp. 1-13
  
  Publication
- Are Language Models Up to Sequential Optimization Problems? From Evaluation to a Hegelian-Inspired Enhancement
  
  Soheil Abbasloo
  
  February 2025
  
  February 2025
  
  Publication
- Bias Control and Injection for Multi-Agent Aggregation
  
  Sungjae Lee, Soheil Abbasloo, Jabin Koo, Jungseul Ok
  
  October 2026
  
  October 2026
  
  Publication
- Dynamic Expert Pruning for Multi-Agent Systems
  
  Jabin Koo, Soheil Abbasloo, Sungjae Lee, Jungseul Ok
  
  October 2026
  
  October 2026
  
  Publication
- Do AI Agents Understand Computer Architecture?
  
  Ambika Sharan, Grigory Chirkov, Soheil Abbasloo
  
  MLForSys | September 2026
  
  PDF
- Evaluating LLM Reasoning Beyond Correctness and CoT
  
  Soheil Abbasloo
  
  February 2026
  
  February 2026
  
  Publication
- Beyond Correctness: Learning Robust Reasoning via Transfer
  
  Hyunseok Lee, Soheil Abbasloo, Jihoon Tack, Jinwoo Shin
  
  February 2026
  
  arXiv | February 2026
  
  Publication
- Are Language Models Up to Sequential Optimization Problems? From Evaluation to a Hegelian-Inspired Enhancement
  
  Soheil Abbasloo
  
  February 2025
  
  February 2025
  
  Publication
- Do AI Agents Understand Computer Architecture?
  
  Ambika Sharan, Grigory Chirkov, Soheil Abbasloo
  
  MLForSys | September 2026
  
  PDF
- Beyond Correctness: Learning Robust Reasoning via Transfer
  
  Hyunseok Lee, Soheil Abbasloo, Jihoon Tack, Jinwoo Shin
  
  February 2026
  
  arXiv | February 2026
  
  Publication

- Mahak: An Automated and Efficient Assessment Framework for Internet Control Algorithms
  
  Parsa Pazhooheshy, Soheil Abbasloo, Y. Ganjali
  
  2025 IEEE 33rd International Conference on Network Protocols (ICNP) | September 2025, pp. 1-13
  
  Publication
- Do AI Agents Understand Computer Architecture?
  
  Ambika Sharan, Grigory Chirkov, Soheil Abbasloo
  
  MLForSys | September 2026
  
  PDF
- Mahak: An Automated and Efficient Assessment Framework for Internet Control Algorithms
  
  Parsa Pazhooheshy, Soheil Abbasloo, Y. Ganjali
  
  2025 IEEE 33rd International Conference on Network Protocols (ICNP) | September 2025, pp. 1-13
  
  Publication
- Bias Control and Injection for Multi-Agent Aggregation
  
  Sungjae Lee, Soheil Abbasloo, Jabin Koo, Jungseul Ok
  
  October 2026
  
  October 2026
  
  Publication
- Dynamic Expert Pruning for Multi-Agent Systems
  
  Jabin Koo, Soheil Abbasloo, Sungjae Lee, Jungseul Ok
  
  October 2026
  
  October 2026
  
  Publication
- CornerCase: Automated Extremal Testing of Protocol Implementations using LLMs
  
  Rathin Singha, Kuan Qian, S. Saikrishnan, Tracy Zhao, Soheil Abbasloo, Ryan Beckett, Siva Kesava Reddy Kakarla, Todd D. Millstein, George Varghese
  
  June 2026
  
  arXiv | June 2026
  
  Publication
- Evaluating LLM Reasoning Beyond Correctness and CoT
  
  Soheil Abbasloo
  
  February 2026
  
  February 2026
  
  Publication
- Beyond Correctness: Learning Robust Reasoning via Transfer
  
  Hyunseok Lee, Soheil Abbasloo, Jihoon Tack, Jinwoo Shin
  
  February 2026
  
  arXiv | February 2026
  
  Publication
- Are Language Models Up to Sequential Optimization Problems? From Evaluation to a Hegelian-Inspired Enhancement
  
  Soheil Abbasloo
  
  February 2025
  
  February 2025
  
  Publication

Soheil Abbasloo

Publications

Bias Control and Injection for Multi-Agent Aggregation

Dynamic Expert Pruning for Multi-Agent Systems

Do AI Agents Understand Computer Architecture?

CornerCase: Automated Extremal Testing of Protocol Implementations using LLMs

Evaluating LLM Reasoning Beyond Correctness and CoT

Beyond Correctness: Learning Robust Reasoning via Transfer

Mahak: An Automated and Efficient Assessment Framework for Internet Control Algorithms

Are Language Models Up to Sequential Optimization Problems? From Evaluation to a Hegelian-Inspired Enhancement

Mahak: An Automated and Efficient Assessment Framework for Internet Control Algorithms

CornerCase: Automated Extremal Testing of Protocol Implementations using LLMs

Mahak: An Automated and Efficient Assessment Framework for Internet Control Algorithms

Are Language Models Up to Sequential Optimization Problems? From Evaluation to a Hegelian-Inspired Enhancement

Bias Control and Injection for Multi-Agent Aggregation

Dynamic Expert Pruning for Multi-Agent Systems

Do AI Agents Understand Computer Architecture?

Evaluating LLM Reasoning Beyond Correctness and CoT

Beyond Correctness: Learning Robust Reasoning via Transfer

Are Language Models Up to Sequential Optimization Problems? From Evaluation to a Hegelian-Inspired Enhancement

Do AI Agents Understand Computer Architecture?

Beyond Correctness: Learning Robust Reasoning via Transfer

Mahak: An Automated and Efficient Assessment Framework for Internet Control Algorithms

Do AI Agents Understand Computer Architecture?

Mahak: An Automated and Efficient Assessment Framework for Internet Control Algorithms

Bias Control and Injection for Multi-Agent Aggregation

Dynamic Expert Pruning for Multi-Agent Systems

CornerCase: Automated Extremal Testing of Protocol Implementations using LLMs

Evaluating LLM Reasoning Beyond Correctness and CoT

Beyond Correctness: Learning Robust Reasoning via Transfer

Are Language Models Up to Sequential Optimization Problems? From Evaluation to a Hegelian-Inspired Enhancement

Contact Soheil Abbasloo

Microsoft Research Asia