- Rating
- 3
- Confidence
- 1
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
This paper introduces CoReL, a BAGEL-based pretraining approach that reconstructs masked infographic elements through three tasks: Numerical Reasoning, Numerical Rendering, and Contextual Synthesis. It also presents InfoContext, a large corpus with automatically detected or transcribed elements, and InfoContext-Bench, a benchmark for numerical and graphical reconstruction. The reported results show gains over evaluated open models on reconstruction and downstream chart QA.
Critical strengths
- The paper targets meaningful relationships among chart values, visual marks, labels, and infographic context, which are not directly tested by many whole-image QA tasks.
- InfoContext could provide useful element-level training data at scale, and InfoContext-Bench evaluates both numerical inference and visual completion.
- The paper includes downstream QA experiments across two backbones and an out-of-domain benchmark, providing some evidence that numerical reconstruction pretraining can transfer to chart understanding.
Critical weaknesses
-
Some of the evaluation results need to be enhanced. Most reported image metrics measure similarity to a single reference, although graphical elements can have multiple valid completions. The paper acknowledges this ambiguity but provides no human evaluation or direct measure of whether generated content is appropriate for the chart’s numerical and semantic context.
-
The contribution of the three-task formulation is not isolated. The paper does not compare joint training of the three tasks against the individual tasks. The downstream transfer experiments use Numerical Reasoning pretraining alone, so they do not establish that the three-task framework—or the image-generation tasks in particular—improves chart understanding. It is therefore difficult to distinguish the value of the proposed formulation from that of additional task-specific training.
Questions for authors
See the weaknesses above
5 comments
Thank you for your response.
Q1. How did you construct the set of valid completions used to compute the set-level metrics? Does this set include all possible valid completions?
Q2. I remain unconvinced by the indirect evidence, though I recognize that conducting an additional experiment would require more time than the rebuttal period allows.
Thank you for the follow-up.
Q1 (Follow-up): Reference-free evaluation for multiple valid completions
The reference set consists of the checked images provided by InfoDet (ICLR 2026) [1], which we treat as valid completions. We agree that it does not cover all valid completions; FID and CLIP-FID measure alignment with this empirical reference distribution.
To address this, we additionally adopt the human-aligned MLLM-as-a-Judge framework of [2], which was validated against human judgments for fine-grained image editing, and apply it to our completion task. It evaluates whether a completion satisfies the task and its context without reference images, which suits settings with multiple valid completions.
We randomly sample 50 semantic examples and evaluate Ours, Qwen, FireRed, and LongCat with three independent judges: GPT-5.5, Claude Opus 5, and Gemini 3.8 Flash (200 assessments per judge, 600 in total). Each judge receives the chart context with the target icon fully masked, the generated chart, and an enlarged crop of the generated target region, but no reference image.
We adapt the framework to six dimensions: context relevance (CR), information expression (IE), style consistency (SC), layout fit (LF), rendering quality (RQ), and outside-region preservation (OP). Each is scored from 1 to 7 (higher is better) and averaged over the 50 examples. Filling the region with background, chart marks, or text does not count as generating an icon; the rubric scores missing icons 1 on CR, IE, and RQ.
| Judge | Method | CR | IE | SC | LF | RQ | OP |
|---|---|---|---|---|---|---|---|
| GPT-5.5 | Ours | 6.18 | 5.82 | 5.52 | 6.32 | 5.92 | 6.92 |
| Qwen | 1.00 | 1.00 | 1.02 | 1.02 | 1.04 | 4.56 | |
| FireRed | 1.00 | 1.00 | 1.12 | 1.24 | 1.08 | 2.62 | |
| LongCat | 1.00 | 1.00 | 1.04 | 1.00 | 1.00 | 5.66 | |
| Claude Opus 5 | Ours | 5.26 | 4.58 | 4.20 | 5.60 | 4.06 | 6.86 |
| Qwen | 1.02 | 1.00 | 1.50 | 1.28 | 1.04 | 3.46 | |
| FireRed | 1.04 | 1.00 | 1.22 | 1.32 | 1.00 | 1.84 | |
| LongCat | 1.02 | 1.00 | 1.78 | 1.18 | 1.04 | 4.14 | |
| Gemini 3.8 Flash | Ours | 5.14 | 4.78 | 4.60 | 5.90 | 4.94 | 6.60 |
| Qwen | 1.34 | 1.34 | 1.54 | 1.50 | 1.44 | 4.44 | |
| FireRed | 1.28 | 1.28 | 1.34 | 1.40 | 1.32 | 2.68 | |
| LongCat | 1.28 | 1.28 | 1.44 | 1.42 | 1.38 | 5.34 |
Baseline editors mostly fill the masked region with background or surrounding marks instead of generating an icon, which explains their near-minimum CR, IE, and RQ scores (see also the third row of Figure 4 and Figure 8). Ours scores highest on all six dimensions under all three judges. Higher CR and IE indicate more contextually appropriate and meaningful icons. Higher SC, LF, and RQ indicate better visual integration and rendering, and OP of 6.60–6.92 indicates strong preservation of the surrounding chart. The advantage therefore holds when completions are judged on their own merits rather than by resemblance to a single reference.
Together with the ChartEval results (Q4 in our response to Reviewer twar) and Tables 4 and 15–17, these complementary evaluations provide substantial evidence of our method's effectiveness while directly addressing the non-uniqueness of valid completions. We welcome any further questions.
[1] InfoDet: A Dataset for Infographic Element Detection. ICLR 2026.
[2] Human-Aligned MLLM Judges for Fine-Grained Image Editing Evaluation: A Benchmark, Framework, and Analysis. arXiv:2602.13028, February 2026.
We thank the reviewer for recognizing that CoReL targets meaningful relationships among chart values, visual marks, labels, and context, that InfoContext and InfoContext-Bench are useful at scale, and that the two-backbone, out-of-domain QA experiments support transfer.
Q1: Evaluation results need to be enhanced: metrics measure similarity to a single reference despite multiple valid completions, with no direct measure of contextual appropriateness.
R: Our evaluation covers both points: set-level metrics (FID, CLIP-FID) account for multiple valid completions, and the newly added ChartEval directly measures contextual appropriateness.
Q2: The contribution of the three-task formulation is not isolated.
R: Our current evidence already supports that joint training is better than individual tasks, as detailed below. A matched individual-task comparison would make this more complete, and we will include it in the revision; each such run takes roughly 64 H20 GPUs for a week on BAGEL, which exceeds the rebuttal period.
Fig. S4 in the newly uploaded supplementary material: Numerical Rendering alone generates a plausible appearance but large numerical errors; adding Numerical Reasoning significantly reduces them.
The context ablation (Figure 5; more examples in Fig. S1 of the newly uploaded supplementary material) shows that the jointly trained model learns contextual relationships within the image, which single tasks cannot provide.
As Reviewer XuFw notes, the same masked inputs supervise both text prediction and image generation, making the joint formulation a natural fit for unified models.
We would be very happy to answer any further questions.