Igor Korostil (@eeghor) header
Igor Korostil (@eeghor) profile picture

Co-Founder & CTO @ Mirmi AI | AI Strategy, GenAI & Automation

Sydney, New South Wales
Joined July 2016
Photo by Igor Korostil (@eeghor)Photo by Igor Korostil (@eeghor)Photo by Igor Korostil (@eeghor)Photo by Igor Korostil (@eeghor)Photo by Igor Korostil (@eeghor)Photo by Igor Korostil (@eeghor)Photo by Igor Korostil (@eeghor)Photo by Igor Korostil (@eeghor)Photo by Igor Korostil (@eeghor)Photo by Igor Korostil (@eeghor)
Commit or Abstain? System One Decision Models, #LLMs, and Missing Legal Authority: arxiv.org/pdf/2610.29961 #LegalAI #GenAI #LegalTech
206
Spilled Energy in Large Language Models: arxiv.org/pdf/2602.18671 #GenAI #LLMs #AI
178
Retrieval-Constrained Decoding Reveals Underestimated Parametric Knowledge in Language Models: arxiv.org/pdf/2509.23417 #LLM #GenAI #ArtificialIntelligence
2
203
A Comprehensive Survey of Self-Evolving #AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systems: arxiv.org/pdf/2508.07407 #LLMs #DataScience
196
A Technical Survey of Reinforcement Learning Techniques for Large Language Models: arxiv.org/pdf/2507.04136 #AI #GenAI #DataScience
1
134
Igor Korostil retweeted
The Illusion of Thinking in LLMs Apple researchers discuss the strengths and limitations of reasoning models. Apparently, reasoning models "collapse" beyond certain task complexities. Lots of important insights on this one. (bookmark it!) Here are my notes:
136
601
128
4,382
956,777
#AI simulations of 1000 people accurately replicate their behaviour: newscientist.com/article/245…
216