One Pass, Any Order: Position-Invariant Listwise Reranking for LLM-Based Recommendation

TL;DR

InvariRank is a novel permutation-invariant reranking framework for LLMs that ensures stable, set-based predictions regardless of candidate order, improving reliability in recommendation systems.

Contribution

It introduces an architectural approach with structured attention and shared positional framing to achieve order-invariant listwise reranking in LLMs.

Findings

Maintains competitive ranking effectiveness

Produces stable rankings across candidate permutations

Demonstrates practical architectural invariance for LLM reranking

Abstract

Large language models (LLMs) are increasingly used for recommendation reranking, but their listwise predictions can depend on the order in which candidates are presented. This creates a mismatch between the set-based nature of recommendation and the sequence-based computation of decoder-only LLMs, where permuting an otherwise identical candidate set can change item scores and final rankings. Such order sensitivity makes LLM-based rerankers difficult to rely on, since rankings may reflect prompt serialization rather than user preference. We propose InvariRank, a permutation-invariant listwise reranking framework that addresses this dependence at the architectural level. InvariRank blocks cross-candidate attention with a structured attention mask and negates position-induced scoring changes through shared positional framing under Rotary Positional Embeddings (RoPE). Combined with a listwise learning-to-rank objective, the model scores all candidates in a single forward pass, avoiding permutation-based invariance training objectives that require multiple permutations of a candidate set. Experiments on recommendation benchmarks show that InvariRank maintains competitive ranking effectiveness while producing stable rankings across candidate permutations. The results suggest that architectural invariance is a practical route to reliable and efficient LLM-based recommendation reranking. The source code is at https://github.com/ejbito/InvariRank.