2606.10794v1 Jun 09, 2026 cs.AI

READER: Robust Evidence-based Authorship Decoding via Extracted Representations

Jing Shao
Jing Shao
Citations: 21
h-index: 2
Dongqi Huang
Dongqi Huang
Citations: 0
h-index: 0
Jiewei Zhang
Jiewei Zhang
Citations: 286
h-index: 10
Liuyin Wang
Liuyin Wang
Citations: 65
h-index: 3
Jiaxu Liu
Jiaxu Liu
Citations: 86
h-index: 5
Su Mu
Su Mu
Citations: 1
h-index: 1

As agentic applications increasingly route user tasks through official and third-party LLM APIs, provenance becomes an operational question: which model generated a given black-box response? We study Dynamic Black-Box LLM Provenance: identifying the source LLM from generations elicited by query-varying, non-predefined prompts rather than a fixed input set or benchmark suite. This setting is difficult because prompt semantics dominate the text, while model-specific authorship traces are weak and inconsistent at the surface level. We introduce READER (Robust Evidence-based Authorship Decoding via Extracted Representations), a lightweight provenance framework that treats a frozen proxy LLM as a reader of hidden authorship evidence. READER maps black-box outputs into proxy activation space, temporally filters token states within each response, and performs Bayesian Evidence Accumulation by summing single-response log-posterior evidence across independently sampled prompts. This avoids fragile mean-pooling of prompt-specific representations while preserving the query-wise evidence needed for calibrated confidence. On Agent500, a 50-target dataset built from agent-style prompts, READER reaches $31.0$-$42.4\%$ top-1 accuracy from a single response and $70.0$-$84.0\%$ from 50 responses, substantially outperforming sentence-encoder fingerprints. Scaling across nine proxy readers further shows that stronger LLMs expose more linearly decodable authorship structure, suggesting that authorship perception is already present in frozen LLM representations and can be converted into reliable multi-query attribution.

0 Citations
0 Influential
5 Altmetric
25.0 Score
Original PDF

No Analysis Report Yet

This paper hasn't been analyzed by Gemini yet.

Log in to request an AI analysis.

댓글

댓글을 작성하려면 로그인하세요.

아직 댓글이 없습니다. 첫 번째 댓글을 남겨보세요!