- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper proposes Switch-Reasoner, which treats thinking as a tool and adaptively selects reasoning modes for MLLMs. Specifically, Switch-Reasoner has global mode balance control and sample-level fine-grained optimization. This dual-level regulation mechanism balances the Thinking and Direct Modes.
Critical strengths
- This paper proposed a dual-level regulation mechanism to balance the overall use of Thinking Mode and Direct Mode. It's somewhat new.
Critical weaknesses
- This paper fails to provide a sufficiently detailed analysis. Somehow, Think Mode is only a format, and even when the model selects the Direct Mode, it may still present a lengthy reasoning process in Chain-of-Thought (CoT) form. In addition to reporting the "think rate", I suggest that the authors provide statistics on output length, as this is crucial.
- To the best of my knowledge, the "thinking-as-tool" concept was first mentioned in the Anthropic blog: https://www.anthropic.com/engineering/claude-think-tool, which has been widely discussed within the community. There are also several highly relevant works that the authors should discuss in the "Related Work" section: https://arxiv.org/html/2505.13379v2; https://arxiv.org/html/2508.12265v2.
- The baseline comparisons in this work are insufficient; it only compares against GRPO. It should incorporate response efficiency (response length) and include comparisons with existing work on efficient reasoning.
- The ablation study variant removed the sample-level fine-grained correction. What would be the effect if the sample-level fine-grained correction were retained while the global balance control was removed? Additionally, I am particularly concerned about the fact that global balance is heavily influenced by batch sampling; how do extreme sampling outcomes—such as a batch consisting entirely of tasks requiring high inference effort followed by a batch of simple tasks—affect training stability?
- Following up on the previous point, I believe that training stability is heavily influenced by the data distribution; therefore, I think this aspect should be described in greater detail in the main text of the paper. What is the composition of your training data, and does it affect the final model's performance and think ratio?
Questions for authors
See Weaknesses.
5 comments
Response to Reviewer PfBQ (2/4)
W3: Additional efficiency statistics and comparisons with existing efficient reasoning methods are needed.
Thank you for this suggestion. We have incorporated thinking rate, generated token count, and end-to-end latency into a unified comparison, and added two adaptive reasoning baselines, Thinkless/DeGRPO and AdaptThink.
| Method | Accuracy | Thinking Rate | Avg. Generated Tokens | Avg. Latency (s) |
|---|---|---|---|---|
| Switch-Reasoner (Full) | 71.60 | 51.53% | 142.55 | 2.94 |
| Thinkless/DeGRPO | 59.02 | 0.00% | 11.76 | 0.35 |
| AdaptThink | 62.23 | 12.33% | 65.26 | 0.73 |
The full Switch-Reasoner achieves an accuracy of 71.60, outperforming Thinkless/DeGRPO and AdaptThink by 12.58 and 9.37 percentage points, respectively.
Meanwhile, Thinkless/DeGRPO exhibits a thinking rate of 0.00%, indicating complete collapse into Direct. Although AdaptThink does not completely collapse, its thinking rate of only 12.33% indicates a pronounced preference for Direct.
In contrast, our method keeps both Direct and Thinking active, avoiding the substantial accuracy degradation associated with reducing computational costs primarily by suppressing reasoning.
We also acknowledge that the two baselines achieve lower token consumption and latency. Therefore, Switch-Reasoner does not dominate all efficiency metrics.
A within-model comparison between default routing and forced Thinking provides a more direct measurement of the efficiency gains from routing.
Default routing achieves an accuracy of 71.60 with 142.55 generated tokens and 2.94 seconds of latency, whereas forced Thinking requires 251.45 tokens and 5.49 seconds to achieve an accuracy of 71.80.
This corresponds to a 43.3% reduction in generated tokens and a 46.4% reduction in latency, with only a 0.20-percentage-point accuracy gap.
This experiment provides more direct evidence that the dual-level regulation mechanism reduces unnecessary reasoning while maintaining high accuracy.
Response to Reviewer PfBQ (3/4)
W4: What happens when Global regulation is removed? Does extreme alternation between task groups affect training stability?
Thank you for raising this question.
We address it through two complementary analyses: an ablation of Global regulation and a controlled experiment on task-order sensitivity.
Global regulation ablation.
We introduce a Sample-only variant that retains sample-level counterfactual routing supervision while disabling Global regulation and the extreme-imbalance safeguard.
Following the protocol of averaging the last three validation checkpoints, the additional ablation results are as follows:
| Method (4B) | Overall Score | Δ vs. Full Model |
|---|---|---|
| Switch-Reasoner (Full) | 71.60 | — |
| Sample-only | 69.82 | −1.78 |
Without Global, Sample-only reaches a thinking rate of 99.18%, compared with 51.53% for the full model, indicating a strong preference for Thinking and near-complete mode collapse.
Meanwhile, the overall score decreases from 71.60 to 69.82.
These results support the role of Global in mitigating mode collapse under the evaluated setting, in terms of both routing behavior and task performance.
Task-order sensitivity.
We further investigated whether changes in batch composition affect controller stability.
We fixed the initialization, 3,520 training samples, batch size, random seed, and execution configuration, varying only the presentation order of task groups.
Under the randomly mixed setting, samples from different tasks were interleaved. Under the alternating setting, task groups representing relatively higher and lower reasoning demands were alternated during the first 80 steps, followed by 30 steps of training on lower-demand task groups.
These task groups serve only as proxies for reasoning demand and are not sample-level difficulty annotations.
Both configurations were evaluated on the same 1,100 validation samples.
We define an extreme-mode batch as a training batch with a thinking rate of at least 98% or at most 2%.
| Sampling Order (Seed 2) | Accuracy | Thinking Rate | Extreme-mode Batches |
|---|---|---|---|
| Randomly Mixed Tasks | 71.60 | 51.53% | 0/110 |
| Alternating Task Groups | 69.96 | 65.55% | 16/110 |
Under the alternating task-group schedule, both reasoning modes remain active. However, accuracy decreases from 71.60 to 69.96 (a reduction of 1.64 percentage points), and the number of extreme-mode batches increases.
This indicates that task order can affect routing stability even when the overall training-data composition remains unchanged.
We will present the corresponding training dynamics and controller-signal changes in Supplementary Figure S2.
Additionally, we observed that another task-alternation experiment resulted in a final thinking rate of 0%. However, because its execution configuration was not consistent with the randomly mixed control setting, we do not include it in the paired quantitative comparison.
We will clarify the task-mixing protocol and acknowledge that, although Global regulation mitigates mode collapse under the evaluated settings, it does not guarantee stability under arbitrary task sequences.
Response to Reviewer PfBQ (1/4)
We sincerely thank the reviewer for the detailed feedback regarding routing behavior, related work, efficiency measurements, and training stability.
W1: Can the Direct mode still generate lengthy reasoning processes?
Thank you for highlighting this important distinction.
The operational definition of Direct is generating an answer without invoking the virtual thinking tool. It does not semantically guarantee that the model performs no internal reasoning or generates no explanatory text.
To examine whether Direct actually produces shorter outputs, we analyzed the generated token distributions of a supplementary 4B checkpoint trained for 110 steps.
The analysis covers 5,500 evaluation samples. Token counts include both model-generation stages and model-generated tool calls. Injected tool-response context is recorded separately and excluded from the generated-output count.
The Direct branch generates only 16.21 tokens on average, with a maximum of 44 tokens among the evaluated samples.
In contrast, the Thinking branch generates 305.25 tokens on average and exhibits a substantially longer tail.
These measurements demonstrate that Direct outputs are indeed concise in the evaluated implementation.
We will clarify that the thinking rate measures the frequency of explicit thinking-tool invocation, whereas generated token statistics provide an independent and more direct characterization of actual output length.
W2: Thinking-as-Tool has precedents, and the related work should be expanded.
We agree with the reviewer and appreciate this suggestion.
We will expand the related work section to discuss representative studies on tool-based reasoning and adaptive reasoning allocation, including:
These studies demonstrate that explicit reasoning control and reasoning-tool interfaces are not new concepts.
Our work builds upon these foundations while focusing on a different training challenge: stabilizing adaptive reasoning-mode selection during GRPO optimization across heterogeneous multimodal tasks.
Specifically, our contribution integrates:
We will revise our contribution statement to emphasize the dual-level training mechanism rather than claiming novelty in the tool abstraction itself.