Michal Lukasik @ ACL 2025

I will be at ACL 2025 in Vienna next week. Let me know if you'd like to meet and chat!

Ever tried to balance multiple goals in training a ranking model, like helpfulness vs. factuality in LLM reward models, or relevance and clicks in retrieval? 🤔 Our #ICML2025 paper shows that common loss aggregation methods for multi-objective learning to rank can cause "label…

🚀 New Paper Alert! 🚀 Want better LLM-as-a-Judge? TRACT: 🧠 CoT + Regression-Aware Fine-tuning (RAFT) = Better numerical predictions! 📊 arxiv.org/abs/2503.04381 🧵👇 A thread on TRACT: