Batch Prompting Suppresses Overthinking Reasoning Under Constraint: How Batch Prompting Suppresses Overthinking in Reasoning Models
Saurabh Srivastava, Janit Bidhan, Hao Yan, Abhishek Dey, Tanu Kansal, Paras Kath, Sina Mansouri, Mohit Marvania, Vamsi Shankar Simhadri, Gaurav Singh

TL;DR
Batch prompting significantly reduces overthinking in reasoning models, cutting reasoning tokens by 76% while maintaining or improving accuracy, by inducing beneficial behavioral effects during inference.
Contribution
This paper demonstrates that batch prompting suppresses overthinking in reasoning models at inference time, improving efficiency and reliability without modifying the models.
Findings
Reduces reasoning tokens by 76% on average
Enables pattern induction from multiple queries
Suppresses metacognitive hedging behaviors
Abstract
Large Reasoning Models (LRMs) achieve strong performance through explicit chain-of-thought reasoning but suffer from \textit{overthinking}: generating excessive reasoning tokens even for trivial queries. {Beyond inflating cost, overthinking can be self-defeating: models enter recursive self-doubt loops that exhaust token budgets without producing an answer, causing API timeouts that directly hurt accuracy.} We present an empirical study showing that \textbf{batch prompting}, originally introduced for throughput optimization, effectively suppresses overthinking at inference time. Across 13 diverse benchmarks with DeepSeek-R1 and OpenAI-o1, batch prompting {reduces reasoning tokens by 76\% (2{,}950710), on average, while preserving or improving accuracy}. Through behavioral analysis, we find that batching induces three beneficial effects: (1) it reduces per-query reasoning effort…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsConstraint Satisfaction and Optimization · Explainable Artificial Intelligence (XAI) · Bayesian Modeling and Causal Inference
