AIDG: Evaluating Asymmetry Between Information Extraction and Containment in Multi-Turn Dialogue
arXiv:2602.17443v1 Announce Type: new Abstract: Evaluating the strategic reasoning capabilities of Large Language Models (LLMs) requires moving beyond static benchmarks to dynamic, multi-turn interactions. We …
Adib Sakhawat, Fardeen Sadab, Rakin Shahriar
조회수 104회