AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Parallel-Probe: Towards Efficient Parallel Thinking via 2D Probing

Tong Zheng, Chengsong Huang, Runpeng Dai, Yun He, Rui Liu, Xin Ni, Huiwen Bao, Kaishen Wang, Hongtu Zhu, Jiaxin Huang, Furong Huang, Heng Huang

ICML 2026 · Architectures and efficiency › RNNs and normalization · approach: Distributed LLM inference systems · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add global 2D probing signals (vs local SAC early stopping) (ablation)Parallel-Probe (Qwen3-0.6B) · AIME24/AIME25 average (ablation)accuracy: 22.4 -> 25.8Improved
add global 2D probing signals (vs local SAC early stopping) (ablation)Parallel-Probe (Qwen3-0.6B) · AIME24/AIME25 average (ablation)sequential tokens: +33.7%Improved
add deviation-based branch pruning (ablation)Parallel-Probe (Qwen3-0.6B) · AIME24/AIME25 average (ablation)accuracy: 23.9 -> 25.8No clear effect
add deviation-based branch pruning (ablation)Parallel-Probe (Qwen3-0.6B) · AIME24/AIME25 average (ablation)total tokens: +14.7%Improved
add consensus-based early stopping (ablation)Parallel-Probe (Qwen3-0.6B) · AIME24/AIME25 average (ablation)accuracy: 25.5 -> 25.8No clear effect
add consensus-based early stopping (ablation)Parallel-Probe (Qwen3-0.6B) · AIME24/AIME25 average (ablation)sequential tokens: +13.1%Improved
add warmup stage (ablation)Parallel-Probe (Qwen3-0.6B) · AIME24/AIME25 average (ablation)accuracy: 23.5 -> 25.8Improved
add warmup stage (ablation)Parallel-Probe (Qwen3-0.6B) · AIME24/AIME25 average (ablation)total tokens: -19.2%Hurt
replace SC@64 with Parallel-ProbeQwen3-8B · AIME24/AIME25/HMMT25 averageaccuracy: 68.6 -> 68.5No clear effect
replace SC@64 with Parallel-ProbeQwen3-0.6B · AIME24/AIME25/HMMT25 averagesequential tokens: -35.8%Improved
replace SC@64 with Parallel-ProbeQwen3-1.7B · AIME24/AIME25/HMMT25 averagetotal tokens: -25.8%Improved
apply per-trajectory SAC early stopping to SC@64Qwen3-8B · AIME24/AIME25/HMMT25 averageaccuracy: 68.6 -> 63.2Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy