PaCoRe: Learning to Scale Test-Time Compute with Parallel Coordinated Reasoning

ACL ARR 2026 January Submission2205 Authors

Keywords: test-time-scaling, reasoning, parallel search

Abstract:

We introduce Parallel Coordinated Reasoning (PaCoRe), a training-and-inference framework designed to overcome a central limitation of contemporary language models: their inability to scale test-time compute (TTC) far beyond sequential reasoning under a fixed context window.
PaCoRe departs from the traditional sequential paradigm by driving TTC through massive parallel exploration coordinated via a message-passing architecture in multiple rounds.
Each round launches many parallel reasoning trajectories, compacts their findings into context-bounded messages, and synthesizes these messages to guide the next round and ultimately produce the final answer.
Trained end-to-end with large-scale, outcome-based reinforcement learning, the model masters the synthesis abilities required by PaCoRe and scales to multi-million-token effective TTC without exceeding context limits.
The approach yields strong improvements across diverse domains and notably pushes reasoning beyond frontier systems in mathematics: an 8B model reaches 94.5\% on HMMT 2025, surpassing GPT-5’s 93.2\% by scaling effective TTC to roughly two million tokens.
We open-source model checkpoints, training data, and the full inference pipeline to accelerate follow-up work.

Paper Type: Long

Research Area: Language Models

Research Area Keywords: scaling, chain-of-thought, reinforcement learning

Contribution Types: NLP engineering experiment

Languages Studied: English, Chinese

Submission Number: 2205