English

You Only Forward Once: An Efficient Compositional Judging Paradigm

Artificial Intelligence 2026-02-03 v3

Abstract

Multimodal large language models (MLLMs) show strong potential as judges. However, existing approaches face a fundamental trade-off: adapting MLLMs to output a single score misaligns with the generative nature of MLLMs and limits fine-grained requirement understanding, whereas autoregressively generating judging analyses is prohibitively slow in high-throughput settings. Observing that judgment reduces to verifying whether inputs satisfy a set of structured requirements, we propose YOFO, a template-conditioned method that judges all requirements in a single forward pass. Built on an autoregressive model, YOFO accepts a structured requirement template and, in one inference step, produces a binary yes/no decision for each requirement by reading the logits of the final token associated with that requirement. This design yields orders-of-magnitude speedups while preserving interpretability. Extensive experiments show that YOFO not only achieves state-of-the-art results on standard recommendation datasets, but also supports dependency-aware analysis -- where subsequent judgments are conditioned on previous ones -- and further benefits from post-hoc CoT.

Keywords

Cite

@article{arxiv.2511.16600,
  title  = {You Only Forward Once: An Efficient Compositional Judging Paradigm},
  author = {Tianlong Zhang and Hongwei Xue and Shilin Yan and Di Wu and Chen Xu and Guannan Zhang and Yunyun Yang},
  journal= {arXiv preprint arXiv:2511.16600},
  year   = {2026}
}
R2 v1 2026-07-01T07:47:43.985Z