TEMPO: Scaling Test-time Training for Large Reasoning Models
Summary
TEMPO introduces a test-time training framework that alternates policy refinement with critic recalibration to prevent diversity collapse and sustain performance gains in large reasoning models, boosting AIME 2024 scores for Qwen3-14B from 42.3% to 65.8%.
View Cached Full Text
Cached at: 04/22/26, 06:17 AM
Paper page - TEMPO: Scaling Test-time Training for Large Reasoning Models
Source: https://huggingface.co/papers/2604.19295
Abstract
TEMPO is a test-time training framework that alternates policy refinement with critic recalibration to sustain performance improvements in language models without diversity collapse.
Test-time training(TTT) adapts model parameters on unlabeled test instances during inference time, which continuously extends capabilities beyond the reach of offline training. Despite initial gains, existing TTT methods for LRMs plateau quickly and do not benefit from additional test-time compute. Without external calibration, the self-generated reward signal increasingly drifts as the policy model evolves, leading to both performance plateaus anddiversity collapse. We propose TEMPO, a TTT framework that interleavespolicy refinementon unlabeled questions with periodiccritic recalibrationon a labeled dataset. By formalizing this alternating procedure through the Expectation-Maximization (EM) algorithm, we reveal that prior methods can be interpreted as incomplete variants that omit the crucial recalibration step. Reintroducing this step tightens theevidence lower bound(ELBO) and enables sustained improvement. Across diverse model families (Qwen3 and OLMO3) and reasoning tasks, TEMPO improves OLMO3-7B onAIME 2024from 33.0% to 51.1% and Qwen3-14B from 42.3% to 65.8%, while maintaining high diversity.
View arXiv pageView PDFProject pageGitHub0Add to collection
Get this paper in your agent:
hf papers read 2604\.19295
Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2604.19295 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2604.19295 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2604.19295 in a Space README.md to link it from this page.
Collections including this paper1
Similar Articles
TEMPO: Temporal Enforcement via Mode-Separated Policy Optimization for Trustworthy LLM Backtesting
Proposes TEMPO, a policy optimization method that trains LLMs to reason exclusively from pre-cutoff information by using a two-mode reward and GRPO-based training, reducing knowledge leakage by 2–13% while improving task performance by 6–13%.
REVES: REvision and VErification--Augmented Training for Test-Time Scaling
Proposes REVES, a two-stage iterative framework that alternates between data augmentation and policy optimization to improve LLM reasoning by leveraging intermediate correction steps, achieving superior performance on coding benchmarks and constraint satisfaction problems.
I trained TIME: short context-triggered thinking on Qwen model instead of overthinking
A personal project led to an ACL 2026 paper introducing TIME, a method training Qwen3 models to engage in short, context-triggered thinking rather than excessive reasoning. The work uses QLoRA and a four-phase curriculum, with all data and code released open-source.
TMAS: Scaling Test-Time Compute via Multi-Agent Synergy
TMAS introduces a multi-agent framework that enhances large language model reasoning by scaling test-time compute through structured collaboration and hierarchical memory systems. The approach uses specialized agents, cross-trajectory information flow, and hybrid reward reinforcement learning to improve iterative scaling and stability on challenging reasoning benchmarks.
ThinkBooster: A Unified Framework for Seamless Test-Time Scaling of LLM Reasoning
ThinkBooster is a unified framework for test-time compute scaling of LLM reasoning, providing a modular Python library, a performance-efficiency benchmark, an OpenAI-compatible proxy service, and a visual debugger. Empirical results on math and coding tasks demonstrate practical gains with quality-cost trade-offs.