Year: 2025
Type: conference-paper
Abstract: Model distillation-using outputs from a large teacher model to teach a small student modelis a practical means of creating efficient models for a particular task.We ask: Can we identify a students' teacher based on its outputs?Such "footprints" left by teacher LLMs would be interesting artifacts.Beyond this, reliable teacher inference may have practical implications as actors seek to distill specific capabilities of massive proprietary LLMs into deployed smaller LMs, potentially violating terms of service. 1 We consider practical task distillation targets including summarization, question answering, and instruction-following.We assume a finite set of candidate teacher models, which we treat as blackboxes.We design discriminative models that operate over lexical features.We find that n-gram similarity alone is unreliable for identifying teachers, but part-of-speech (PoS) templates (Shaib et al., 2024b) preferred by student models mimic those of their teachers.
Institution: Northeastern University
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: gold
OpenAlex ID: W4412888874