P2R builds LLM-generated structured expertise profiles and uses coarse-to-fine hybrid retrieval plus rubric-scoring committees to match papers to reviewers, outperforming paper-to-paper baselines on NeurIPS, SIGIR, and SciRepEval.
Rethinking graph contrastive learning through relative similarity preservation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IR 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Beyond Paper-to-Paper: Structured Profiling and Rubric Scoring for Paper-Reviewer Matching
P2R builds LLM-generated structured expertise profiles and uses coarse-to-fine hybrid retrieval plus rubric-scoring committees to match papers to reviewers, outperforming paper-to-paper baselines on NeurIPS, SIGIR, and SciRepEval.