- Rating
- 3
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
The paper studies consensus-based test-time RL for VLMs (TTRV) on free-form VQA and finds two problems. Most exact-match gains are answer-format fixes: 72.3% of corrections are form-only, and exact match rises 3.36 points while ground-truth containment falls 0.41. In addition, 69% of incorrect responses don't depend on the image, the question, or both. The proposed method TTIQ, measures how much a sampled response's token likelihoods drop when the image is zeroed or the question is masked. Its reward is the within-group rank of min(image dependence, question dependence, confidence cap). Token-level credit goes mostly to tokens supported by both inputs, and training uses GRPO on 20 unlabeled examples per task. On eight VQA tasks and three InternVL backbones, TTIQ has the best average exact match at each scale: 58.77 / 62.41 / 66.94, against 57.48 / 61.36 / 66.67 for the best of TTRL, SCOPE and TTRV. The paper also reports ablations, cross-dataset transfer and results on Qwen2.5-VL.
Critical strengths
- Splitting exact-match gains into form-only and content corrections exposes a measurement artifact that probably affects other TTRL-style work. It also gives the community a simple check that can be applied to any label-free method.
- TTIQ costs two extra teacher-forced passes (about 1.6× time per update, same memory) and needs only 20 unlabeled examples. It beats TTRV in all 24 backbone–task cells, and code is released.
Critical weaknesses
-
The claim that TTIQ corrects content rather than answer form is not well supported. e.g. 68.4% of TTIQ's own corrections are form-only, versus 72.3% for TTRV (Fig. 4). Ground-truth containment, the metric used against TTRV, is never reported for TTIQ. Gains concentrate where Base exact match is limited by format. On InternVL2.5-4B, COCO-QA-N goes from 36.0 to 74.7, and training only on TextVQA already reaches 72.4 (Table 3). This points to a task-agnostic format change, not better grounding. In Table 2, an image-only reward scores 57.66. That is above the full reward with the same credit (56.40) and close to min(I, Q) with joint-token credit (57.79), so the "joint" design adds little.
-
The evaluation protocol likely inflates small margins. \delta is chosen per task by test accuracy, while the baselines are not tuned. As a result, TTIQ beats the \delta = 0 and \delta = 1 rows of Table 2 by construction. With the best fixed \delta (read from Fig. 5a), confidence adds only about 0.3 points. No seeds or variance are reported. Margins over the strongest baseline are +1.29, +1.05 and +0.27 at 2B, 4B and 8B; the 4B margin is +0.54 without COCO-QA-N. On Qwen2.5-VL-3B, TTIQ is only +0.24 over Base.
Questions for authors
- Was \delta selected per task on the evaluation split? Which run produced the 4B TextVQA result of 79.52, given that Fig. 5a peaks at about 79.1?
- Were the tables produced with the released code, which uses a sustained weight of 0.1 and a token-credit floor of 0.2?
- In Fig. 2, what threshold defines "lacking dependence," and what is the same breakdown for correct responses?
4 comments
Response to Reviewer w5KU (2/3)
3. Joint dependence addresses different task requirements. Image-only dependence is competitive. However, on some dataset like GQA-Rel., adding question dependence improves the average from 47.96 to 56.40.
We interpret this difference in light of task requirements. Image-only rewards may be particularly effective when salient visual content strongly constrains the answer, as in some OCR questions (such as TextVQA, InfoVQA). However, strong image dependence alone does not ensure that a response addresses the specific question. Question dependence provides an additional signal when the question determines which object, attribute, or relation is relevant, or which reasoning operation is required, as in GQA. Under minimum fusion, adding this signal can downweight responses that use the image but weakly depend on the specific question, substantially changing their ranking and the resulting training preferences.
These findings support the complementary roles of image and question dependence across different task requirements. The complete TTIQ configuration combines these signals with confidence and joint-token credit assignment, achieving the highest average accuracy in Table 2.
W2 & Q1: Evaluation Protocol, Confidence Selection, and Statistical Reliability.
W2: The evaluation protocol likely inflates small margins. \delta is chosen per task by test accuracy, while the baselines are not tuned. As a result, TTIQ beats the \delta = 0 and \delta = 1 rows of Table 2 by construction. With the best fixed \delta (read from Fig. 5a), confidence adds only about 0.3 points. No seeds or variance are reported. Margins over the strongest baseline are +1.29, +1.05 and +0.27 at 2B, 4B and 8B; the 4B margin is +0.54 without COCO-QA-N. On Qwen2.5-VL-3B, TTIQ is only +0.24 over Base. Q1: Was \delta selected per task on the evaluation split? Which run produced the 4B TextVQA result of 79.52, given that Fig. 5a peaks at about 79.1?
Thanks for raising this question. On selection. We'd like to clarify that the gains are not dependent on selecting the best for each task. Across the 24 model–dataset combinations, TTIQ outperforms the most closely related baseline, TTRV, in all 24 cells when reporting the best . More importantly, using the same fixed for every cell still yields improvements in 23/24 cells. Even if we take the worst result among all evaluated values for each cell, TTIQ remains better in 20/24 cells. Thus, selection affects the magnitude of the gain but does not determine the overall conclusion. All baselines use the default settings recommended by their papers or official implementations.
On the contribution of confidence. The average gain under a fixed does not fully capture the value of confidence across diverse tasks. Although confidence contributes little on some tasks, it can be useful when model likelihood helps distinguish reliable responses from unreliable ones. For example, Figure 5(a) shows that improves TextVQA by approximately 1.2 points over . Our goal is to develop a general method that accommodates diverse tasks, where different signals may be useful to different degrees. Therefore, we therefore retain confidence as an adjustable constraint that provides additional guidance when informative.
On the TextVQA discrepancy. The 79.52 TextVQA result and Figure 5(a) used different randomly sampled sets of 20 training examples; we apologize for the unclear reporting. We will refine the figure to make it more consistent.
Here, we additionally show that our method is robust to different training sets. We report results across 3 randomly sampled training sets below. TTIQ achieves the highest mean accuracy, supporting its effectiveness beyond a particular training sample.
| Model | Method | GQA-Rel | GQA-Attr | ST-VQA | COCO-QA (Num) |
|---|---|---|---|---|---|
| InternVL3-2B | TTRL | 32.24 ± 0.84 | 52.84 ± 1.66 | 76.25 ± 0.37 | 72.93 ± 0.19 |
| SCOPE | 32.77 ± 0.89 | 52.79 ± 1.62 | 76.24 ± 0.35 | 72.98 ± 0.39 | |
| TTRV | 27.47 ± 3.59 | 52.50 ± 0.75 | 76.06 ± 0.10 | 74.08 ± 0.42 | |
| TTIQ (ours) | 33.63 ± 1.53 | 53.76 ± 0.60 | 77.08 ± 0.61 | 74.66 ± 0.56 | |
| InternVL2.5-4B | TTRL | 35.24 ± 0.57 | 56.74 ± 0.63 | 82.98 ± 1.05 | 72.13 ± 2.91 |
| SCOPE | 33.73 ± 1.42 | 56.61 ± 1.08 | 82.89 ± 0.88 | 72.82 ± 2.34 | |
| TTRV | 28.52 ± 4.90 | 56.95 ± 1.30 | 81.53 ± 0.34 | 72.22 ± 2.31 | |
| TTIQ (ours) | 36.37 ± 0.70 | 57.69 ± 0.55 | 83.13 ± 0.62 | 74.64 ± 0.21 |
On training seeds and variance. We have also added two random seeds and now report mean and standard deviation across three seeds for every method and have updated our Table 1 in the manuscript. We found that variability is comparable across methods, and the three-seed averages retain the improvements, with better performance compared to baselines in several settings.
Response to Reviewer w5KU (3/3)
Q2: Implementation Settings and Sensitivity.
Were the tables produced with the released code, which uses a sustained weight of 0.1 and a token-credit floor of 0.2?
Thank you for your careful checking. Yes, all main-table TTIQ results used the released settings: for sustained dependence and for the token-credit floor. These values were set initially and kept fixed across all datasets without tuning. We are sorry for missing this detail and we have clarified it in the implementation section in the updated manuscript.
Specifically, the sustained term supplements overall dependence by measuring whether support persists across response tokens. The floor preserves positive credit for low-support tokens and limits the maximum token-weight ratio to . Subsequent mean-one normalization preserves this ratio.
Additional sensitivity checks show little variation across nearby settings:
| Floor | CLEVR-H, 4B | TextVQA, 8B |
|---|---|---|
| 0.1 | 78.67 | 84.34 |
| 0.2 | 78.81 | 84.48 |
| 0.3 | 78.77 | 84.49 |
Varying from to changes pairwise reward-ranking accuracy by at most 0.05 percentage points in our sensitivity analysis.
Q3: Dependence Threshold and Correct-Response Breakdown.
In Fig. 2, what threshold defines "lacking dependence," and what is the same breakdown for correct responses?
Thank you for asking for this clarification. The threshold is 0. We applied the same criterion to correct responses using InternVL2.5-4B on eight datasets, with 200 questions per dataset and four pre-update rollouts per question:
| Response subset | only | only | Neither | Lacking at least one | |
|---|---|---|---|---|---|
| Incorrect () | 1,212 (36.2%) | 363 (10.8%) | 737 (22.0%) | 1,039 (31.0%) | 2,312 (69.0%) |
| Correct () | 494 (16.2%) | 260 (8.5%) | 97 (3.2%) | 2,198 (72.1%) | 851 (27.9%) |
Joint dependence is present in 72.1% of correct responses, compared with 31.0% of incorrect responses. This contrast supports its usefulness as a proxy for correctness during label-free test-time training, while showing that it is not a guarantee of correctness.
A gentle follow-up on our rebuttal
Dear reviewer w5KU,
We sincerely thank you for your time and effort in reviewing our paper, as well as your thoughtful feedback and constructive suggestions throughout the discussion.
In our rebuttal, we (1) provided a dataset-level analysis of corrections, showing that TTIQ produces 84.0% more content-changing corrections than TTRV, and explained how TTIQ encourages better use of both the image and the question, a general VQA ability that supports cross-dataset transfer; (2) addressed concerns regarding confidence weight by providing additional results showing that the overall gains remain robust without selection, and reported results across randomly sampled training sets and additional training seeds; (3) clarified the implementation settings and provided additional sensitivity analyses; and (4) clarified the dependence threshold in Figure 2 and provided the same breakdown for correct responses, showing that 72.1% exhibit joint image–question dependence, compared with 31.0% of incorrect responses.
With less than 13 hours remaining in the discussion period, we would greatly appreciate hearing whether our responses have adequately addressed your concerns. If any questions or concerns remain, we would be happy to provide further clarifications and continue the discussion before the deadline. Thank you again for your valuable feedback, time, and consideration, which have been greatly appreciated and helpful in improving our work.
Response to Reviewer w5KU (1/3)
We appreciate the reviewer’s detailed and constructive comments, which have prompted further analysis and improvements to the paper. We address the concerns below with additional evidence and explanations, and would be happy to provide further clarification or discuss any remaining questions.
W1: Content Corrections, Cross-Dataset Transfer, and Joint Dependence.
Thank you for your careful review and constructive feedback. Below, we address your concerns regarding 1) content corrections, 2) cross-dataset transfer, and 3) the joint design below.
1. TTIQ corrects substantially more content errors. We would like to clarify that correcting answer form also contributes to accuracy, particularly on datasets where answer-form mismatches account for many evaluation errors. Both form-only and content-changing corrections contribute to improved accuracy. Our goal is to correct more errors overall, especially content errors that existing baselines leave unresolved, while preserving the benefits of answer-form refinement.
Figure 4 reports percentages relative to each method’s own total corrections, which can obscure the difference in absolute content corrections. The table below reports content-changing correction counts for each dataset, summed across the three model backbones, together with TTIQ’s relative increase over TTRV. In total, TTIQ produces 2,105 content-changing corrections, compared with 1,144 for TTRV, an increase of 84.0%, with gains on all eight datasets.
These results show that, alongside answer-form improvements, TTIQ delivers substantially more content corrections than TTRV.
2. TTIQ reinforces input use that transfers across tasks. We want to emphasize that TTIQ encourages a general ability to answer using both the image and the question. This ability is useful across datasets with different tasks and answer requirements, providing a basis for the strong transfer results.
The comparison with TTRV supports this interpretation. On InternVL2.5-4B, direct COCO-QA-N adaptation with TTRV reaches 69.58, but TextVQA adaptation reaches only 33.53, below Base at 36.01. Thus, TTRV’s substantial answer-form improvements do not successfully transfer. TTIQ, however, reaches 72.39 after TextVQA adaptation, close to 74.70 with direct adaptation. Its transfer also extends to GQA-Relation and GQA-Attribute, reaching 30.16 and 55.67, compared with 16.91 and 48.91 for TTRV.
Our transferable improvements can include both content and form corrections because TTIQ encourages a general ability to answer from the inputs. This does not exclude improvements in answer form. Specifically, responses expressing the same correct answer in different forms can receive different rewards because their image–question dependence differs under the given prompt. These reward differences can favor a more appropriate answer form. Furthermore, sustained dependence and joint-token credit favor expressions supported by both inputs, which may also help responses align with the information and answer format requested in the prompt. Thus, TTIQ can transfer improvements in both answer content and form by reinforcing a general ability to answer using the image and question.
3. Joint dependence addresses different task requirements. (Continue on next block)