PoLi-RL: A Point-to-List Reinforcement Learning Framework for Conditional Semantic Textual Similarity

Zixin Song, Bowen Zhang, Qian-Wen Zhang, di yin, Xing Sun, Chunping Li

Keywords: Conditional Semantic Textual Similarity, Reinforcement Learning, Large Language Models, Natural Language Processing, Curriculum Learning

TL;DR: We propose PoLi-RL, a two-stage RL framework that enables effective listwise optimization for the C-STS task via a progressive curriculum and a Parallel Slice Ranking Reward mechanism , achieving SOTA performance for the cross encoder architecture.

Abstract: Conditional Semantic Textual Similarity (C-STS) measures the semantic proximity between text segments under a specific condition, thereby overcoming the ambiguity inherent in traditional STS. However, existing methods are largely confined to discriminative models, failing to fully leverage recent breakthroughs in the NLP community involving Large Language Models (LLMs) and Reinforcement Learning (RL). RL is a particularly well-suited paradigm for this task, as it can directly optimize the non-differentiable Spearman ranking metric and guide the reasoning process required by C-STS. Nevertheless, we find that naively applying listwise RL fails to produce meaningful improvements, as the model struggles with complex, coarse-grained reward signals, leading to optimization difficulties. To address this challenge, we introduce PoLi-RL, a novel Point-to-List Reinforcement Learning framework. PoLi-RL employs a two-stage curriculum: it first trains the model with a simple pointwise reward to establish fundamental scoring capabilities, then transitions to a hybrid reward that combines pointwise, pairwise, and listwise objectives to refine the model's ability to discern subtle semantic distinctions. Crucially, we propose an innovative Parallel Slice Ranking Reward (PSRR) mechanism that computes ranking rewards in parallel slices, where each slice consists of completions with the same index from different samples. This provides a precise, differentiated learning signal for each individual completion, enabling granular credit assignment and effective optimization. On the official C-STS benchmark, PoLi-RL achieves a Spearman correlation coefficient of 48.18, establishing a new SOTA for the cross-encoder architecture. Furthermore, PoLi-RL also maintains SOTA status on the re-annotated C-STS dataset, confirming its robust generalization capabilities. As the first work to successfully apply RL to C-STS, our study introduces a powerful paradigm for aligning LLMs for complex, ranking-based conditional judgment tasks. Our code and checkpoints are available at https://github.com/ZBWpro/PoLi-RL.

Primary Area: applications to computer vision, audio, language, and other modalities

Submission Number: 18363