AI Paper Atlas

Research map / Large language models / Reasoning and chain of thought

AdaThinking-E: One-Token Entropy Regulation for Adaptive Thinking

Zining Wang, Tongkun Guan, Boming Chen, Zhentao Guo, Jianqiang Liu, chao jin, Chen Duan, Kai zhou, Pengfei Yan, Wei Shen, Xiaokang Yang

ECCV 2026 · Large language models › Reasoning and chain of thought · approach: Efficient and adaptive thinking

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 RL stage with entropy regulation after cold start (ablation)AdaThinking-E Qwen2.5-VL-7B · ChartQAaccuracy: 87.6 -> 89.1Improved
add RL stage with entropy regulation after cold start (ablation)AdaThinking-E Qwen2.5-VL-7B · OCR-Reasoningaccuracy: 21.6 -> 26.5Improved
use two-stage entropy curriculum instead of constant high entropy (ablation)AdaThinking-E · ChartQAaccuracy: 88.1 -> 89.1Improved
use two-stage entropy curriculum instead of constant high entropy (ablation)AdaThinking-E · OCR-Reasoningaccuracy: 23.7 -> 26.5Improved
use two-stage entropy curriculum instead of constant low entropy (ablation)AdaThinking-E · OCR-Reasoningaccuracy: 18.7 -> 26.5Improved
use two-stage entropy curriculum instead of constant low entropy (ablation)AdaThinking-E · ChartQAaccuracy: 87.8 -> 89.1Improved
add accuracy and propensity guidance terms to decision feedback (ablation)AdaThinking-E · ChartQAthinking ratio: 39.9 -> 24.3Improved
add accuracy and propensity guidance terms to decision feedback (ablation)AdaThinking-E · OCR-Reasoningaccuracy: 25.7 -> 26.5Improved
lower inflection point gamma from 0.6 to 0.4 (beta 10) (ablation)AdaThinking-E · ChartQAthinking ratio: 39.3 -> 24.3Improved
raise smoothing coefficient beta from 5 to 10 (gamma 0.4) (ablation)AdaThinking-E · OCR-Reasoningaccuracy: 26.1 -> 26.5Improved
lower propensity threshold Delta_r from 0.625 to 0.375 (ablation)AdaThinking-E · ChartQAaccuracy: 88.4 -> 89.1Improved
replace GRPO/DAPO uniform token reward with decoupled entropy reward (ablation)AdaThinking-E · training dynamicsthinking ratioImproved

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