ParaTempo: Efficient Parallel Reasoning via Temporal Confidence
Summary
ParaTempo is a training-free asynchronous parallel reasoning framework that uses temporal confidence to dynamically manage reasoning branches, reducing latency and token usage while maintaining accuracy in mathematical and scientific reasoning benchmarks.
View Cached Full Text
Cached at: 08/24/26, 12:28 PM
Paper page - ParaTempo: Efficient Parallel Reasoning via Temporal Confidence
Source: https://huggingface.co/papers/2608.16425
Abstract
ParaTempo improves parallel reasoning efficiency by using temporal confidence to dynamically prune, retire, and reallocate reasoning branches without synchronization.
Parallel reasoningimproves the accuracy and robustness of large reasoning models by exploring multiple solution paths, but its computational cost grows with reasoning depth and branch count. Existing methods for managing these parallel paths typically rely on final-answer consensus, local token confidence, or isolated intermediate probes. However, these signals are often delayed, weakly tied to actual reasoning progress, or too noisy for dynamic,branch-level control. To address these limitations, we introduce ParaTempo, a training-free asynchronousparallel reasoningframework. ParaTempo is driven bytemporal confidence, a branch-local measure ofanswer-space convergence. Each branch is periodically probed for a tentative answer probability distribution, andtemporal confidencequantifies how sharply the recent intermediate probes concentrate on a dominant answer. Once sufficient evidence has accumulated, ParaTempo drives its entire control process from this single signal: low-confidence branches are pruned, branches that persistently commit to their dominant answer are retired early, freed computation is reallocated by forking new branches, and generation stops globally once theconfidence-weighted voteconcentrates. Without requiring synchronization among reasoning trajectories, ParaTempo adaptively allocates computation based on branch-level convergence. Experiments on challenging mathematical and scientific reasoning benchmarks show that ParaTempo reduces average latency by 21.8-32.2% and total token usage by 18.1-30.3% while maintaining competitive accuracy. Moreover,temporal confidenceexhibits stronger temporal stability and predictive power for future branch convergence than token-level and instantaneous signals.
View arXiv pageView PDFAdd to collection
Get this paper in your agent:
hf papers read 2608\.16425
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/2608.16425 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2608.16425 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2608.16425 in a Space README.md to link it from this page.
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
TEMPO: Scaling Test-time Training for Large Reasoning Models
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%.
Second Thought: Reasoning in Parallel as LLM Agents Act and Observe
Second Thought is a training-free framework that runs auxiliary reasoning branches in parallel during LLM agent action-observation waits to reduce sequential decoding and turn counts without harming accuracy.
CAPS: Cascaded Adaptive Pairwise Selection for Efficient Parallel Reasoning
CAPS introduces a cascaded adaptive selection framework for efficient parallel reasoning, reducing verifier compute costs by over 75% while outperforming existing pairwise verification methods across multiple LLM benchmarks.
PI-Mem: Pushing Long-Context Reasoning to 3.6M Tokens with Parallel-Iterative Memory
PI-Mem is a parallel-iterative memory mechanism that pushes long-context reasoning to 3.6M tokens, outperforming recurrent-memory baselines while achieving significant inference speedups.
NoisyCoconut: Counterfactual Consensus via Latent Space Reasoning
This paper introduces NoisyCoconut, an inference-time method that improves LLM reliability by injecting noise into latent trajectories to generate diverse reasoning paths. The approach enables models to abstain when uncertain, significantly reducing error rates in mathematical reasoning tasks without requiring retraining.