SPIRAL: Learning to Search and Aggregate

• URL: http://arxiv.org/abs/2606.23595v1
• Date: Mon, 22 Jun 2026 17:02:09 GMT
• Title: SPIRAL: Learning to Search and Aggregate
• Abstract summary: We introduce SPIRAL, a framework in which a language model is trained to use all three primitives as part of a unified inference compute pipeline.<n>We show that SPIRAL effectively scales with inference compute, outperforming GRPO by up to 11$times$ scaling efficiency and 15% higher performance when all three compute primitives are scaled.
• Score: 63.26414719363803
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Language model reasoning can be substantially improved at test time via scaffolds that scale inference compute across different primitives -- sequential reasoning within a trace, independently sampled parallel traces, and aggregation of multiple reasoning traces into a final response. During post-training, however, language models are optimized only for sequential reasoning within a single trace. We introduce Sequential-Parallel-Aggregative Reinforcement Learning (SPIRAL), a framework in which a language model is trained to use all three primitives, as part of a unified inference compute pipeline. Concretely, the language model first samples a set of independent traces in parallel, each produced through sequential chain-of-thought reasoning, and then generates a final aggregation trace conditioned on those traces; all components are optimized end-to-end against the reward of the final aggregated response. To train this system, SPIRAL uses set reinforcement learning to teach models to produce a set of traces that are collectively useful for an aggregator and standard reinforcement learning to teach models to aggregate the set into improved final responses. Our experiments on reasoning tasks show that SPIRAL effectively scales with inference compute, outperforming GRPO by up to 11$\times$ scaling efficiency and 15% higher performance when all three compute primitives are scaled.

Related papers

• UniT: Unified Multimodal Chain-of-Thought Test-time Scaling [85.590774707406]
  Unified models can handle both multimodal understanding and generation within a single architecture, yet they typically operate in a single pass without iteratively refining their outputs.<n>We introduce UniT, a framework for multimodal test-time scaling that enables a single unified model to reason, verify, and refine across multiple rounds.
  arXiv  Detail & Related papers  (2026-02-12T18:59:49Z)
• Coupled Variational Reinforcement Learning for Language Model General Reasoning [83.82392089177841]
  We propose textitbCoupled bVari bReinforcement bLearning (CoVRL) to bridge variational inference and reinforcement learning.<n>CoVRL improves performance by 12.4% over the base model and achieves an additional 2.3% improvement over strong state-of-the-art verifier-free RL baselines.
  arXiv  Detail & Related papers  (2025-12-14T07:03:51Z)
• Mind the Gap: Bridging Thought Leap for Improved Chain-of-Thought Tuning [54.65050470296886]
  We propose the CoT Thought Leap Bridge Task, which aims to automatically detect leaps and generate missing intermediate reasoning steps.<n>We demonstrate that models fine-tuned on bridged datasets consistently outperform those trained on original datasets.<n>Our approach effectively enhances distilled data and provides better starting points for reinforcement learning.
  arXiv  Detail & Related papers  (2025-05-20T17:59:31Z)
• PRefLexOR: Preference-based Recursive Language Modeling for Exploratory Optimization of Reasoning and Agentic Thinking [0.0]
  PRefLexOR combines preference optimization with concepts from Reinforcement Learning to enable models to self-teach. We focus on applications in biological materials science and demonstrate the method in a variety of case studies.
  arXiv  Detail & Related papers  (2024-10-16T08:46:26Z)
• Tree-Averaging Algorithms for Ensemble-Based Unsupervised Discontinuous Constituency Parsing [23.091613114955543]
  We propose to build an ensemble of different runs of the existing discontinuous by averaging the predicted trees. We then develop an efficient exact algorithm to tackle the task, which runs in a reasonable time for all samples. Results on three datasets show our method outperforms all baselines in all metrics.
  arXiv  Detail & Related papers  (2024-02-29T21:49:31Z)
• Infinite forecast combinations based on Dirichlet process [9.326879672480413]
  This paper introduces a deep learning ensemble forecasting model based on the Dirichlet process. It offers substantial improvements in prediction accuracy and stability compared to a single benchmark model.
  arXiv  Detail & Related papers  (2023-11-21T06:41:41Z)
• Scalable Learning of Latent Language Structure With Logical Offline Cycle Consistency [71.42261918225773]
  Conceptually, LOCCO can be viewed as a form of self-learning where the semantic being trained is used to generate annotations for unlabeled text. As an added bonus, the annotations produced by LOCCO can be trivially repurposed to train a neural text generation model.
  arXiv  Detail & Related papers  (2023-05-31T16:47:20Z)
• COLO: A Contrastive Learning based Re-ranking Framework for One-Stage Summarization [84.70895015194188]
  We propose a Contrastive Learning based re-ranking framework for one-stage summarization called COLO. COLO boosts the extractive and abstractive results of one-stage systems on CNN/DailyMail benchmark to 44.58 and 46.33 ROUGE-1 score.
  arXiv  Detail & Related papers  (2022-09-29T06:11:21Z)
• Parameter Decoupling Strategy for Semi-supervised 3D Left Atrium Segmentation [0.0]
  We present a novel semi-supervised segmentation model based on parameter decoupling strategy to encourage consistent predictions from diverse views. Our method has achieved a competitive result over the state-of-the-art semisupervised methods on the Atrial Challenge dataset.
  arXiv  Detail & Related papers  (2021-09-20T14:51:42Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.