Pith. sign in

REVIEW 4 major objections 5 minor 49 references

Chart-CoCa: Self-Improving Chart Understanding of Vision LMs via Code-Driven Synthesis and Candidate-Conditioned Answering

T0 review · 4 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A vision-language model improves its own chart understanding by up to 15.5 accuracy points, using generated code for clean training data and candidate-conditioned answering, with no human labels or external teacher model.

desk verdict Code-executed synthetic labels are a genuinely useful fix for chart-QA self-training; the reasoning gains are real but likely candidate selection, not learned reasoning. read the letter →

arxiv 2508.11975 v1 pith:ZFSV7CZX submitted 2025-08-16 cs.AI

Gongyao Jiang , Qiong Luo This is my paper · ORCID
classification cs.AI
keywords chartunderstandingvisionlanguagemodelsmultimodallargequestionansweringsyntheticdatagenerationcode-drivensynthesiscandidate-conditionedself-improvement
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Chart-CoCa tries to establish that a vision-language model can improve its own chart understanding without human-labeled answers or a stronger external model. It does this in two steps: first, the model turns an unlabeled chart into a description and then into executable Python code, and running that code produces both a new chart and the exact chart elements needed to label simple questions; second, the model is fine-tuned to produce a final answer when it is shown a chart, a question, and several candidate answers it has sampled. On the CharXiv benchmark with k=5 candidates, this lifts InternVL2-8B descriptive accuracy from 54.10 to 69.60 and reasoning accuracy from 23.60 to 31.60, and the same recipe improves other open vision-language models. If the claim holds, a vision-language model can bootstrap its own visual reasoning from code-verified synthetic data rather than relying on costly annotation or teacher models.

What carries the argument

The mechanism that carries the argument is twofold. The first load-bearing object is executable code used as an intermediary: because the chart and its extracted information both come from executing the same code, the synthetic training labels are exact by construction rather than generated from noisy image-to-text predictions. The second is candidate-conditioned answering: the answer model receives the chart, the question, and several sampled responses, and is trained to emit the correct answer, so it learns to weigh and combine candidates instead of trusting majority consensus or a separately trained reward model. Code execution also provides the retry filter that discards failed synthesis attempts, and the candidate loop provides the test-time scaling that lets model accuracy grow with inference budget.

What would settle it

Isolate the CharXiv reasoning questions where all k sampled candidates are wrong, so Pass@K fails for that subset; if Chart-CoCa's accuracy on those questions is near zero or no better than picking a candidate at random, then the method is selecting from candidates rather than synthesizing new answers, and the reasoning gain would come from candidate coverage alone.

Watch

Extended reading notes

Core claim

The paper's central claim is that a single vision-language model can be its own data generator and its own answer synthesizer. Starting from an unlabeled real chart, the VLM writes a description, converts it to executable Matplotlib code, runs that code to draw a new chart, and reads chart elements such as titles, labels, ticks, legends, and lines back from the plotting objects; rule-based matching then turns those elements into accurate question-answer triplets without human annotation or a stronger teacher model. The same VLM is next fine-tuned on chart, question, and k sampled candidate answers to output the ground-truth answer, thereby learning to condition on candidates. At test time the initial model samples k answers and the fine-tuned answer model produces the final prediction. With k=5 on the CharXiv validation set, Chart-CoCa improves InternVL2-8B descriptive accuracy from 54.10 to 69.60 and reasoning accuracy from 23.60 to 31.60, and it outperforms majority voting, self-verification, reward-model selection, and direct fine-tuning.

Load-bearing premise

The load-bearing premise is that candidate-conditioned answering trained on simple, rule-generated descriptive questions about synthetic charts transfers to open-ended reasoning questions about real charts, even though direct fine-tuning on the same synthetic data lowers reasoning accuracy in the paper's own ablation.

Editorial extensions

If this is right

  • On the CharXiv validation set, the full Chart-CoCa pipeline raises InternVL2-8B descriptive accuracy from 54.10 to 69.60 and reasoning accuracy from 23.60 to 31.60, surpassing majority voting, self-verification, reward-model selection, and direct fine-tuning.
  • Accuracy keeps rising as more candidate answers are sampled, with most of the practical gain reached by K=5–10 before the curve flattens.
  • The gains transfer across model families, including LLaVA-1.6-7B, Qwen2VL-7B, and InternVL2 in 7B, 8B, and 26B sizes, all improving over their initial models and over majority voting.
  • Smaller models improve the most in relative terms, so the method is most useful where the base vision-language model is weakest.
  • No human-labeled answers and no external teacher model are required, so the recipe can be rerun on any new vision-language model that can generate and execute code.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same code-as-intermediary design could generalize to any domain where executable code draws the visual input and exposes structured attributes, such as diagrams, maps, or synthetic scenes, giving a general route to noise-free self-training data.
  • The answer model may be learning to combine fragments of correct information across wrong candidates rather than simply selecting; this could be tested by feeding candidates that are all wrong but jointly contain the missing pieces and checking whether the final answer is assembled correctly.
  • Because direct fine-tuning on the synthetic triplets degrades reasoning, an iterative self-training loop should treat the code-execution filter, not the model's own confidence, as the guarantee that new training data stays clean.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper proposes Chart-CoCa, a self-improvement framework for chart understanding in vision language models (VLMs). The method has two main components: (1) a code-driven chart synthesis pipeline in which the VLM describes a real chart, generates Python/Matplotlib code, executes it to produce a new chart, and extracts chart information from plotting objects to obtain accurate chart-question-answer triplets; and (2) a candidate-conditioned answering strategy in which, at training time, the VLM is fine-tuned to produce the correct answer given the chart, question, and multiple sampled candidate answers, and at test time the fine-tuned model converts a fresh set of candidates into a final answer. Experiments on the CharXiv benchmark report consistent improvements over the initial VLM and several baselines, including up to a 15.50-point absolute gain on descriptive tasks and an 8.00-point gain on reasoning tasks for InternVL2-8B. The paper also includes ablations showing the importance of the description and code-mediation components, candidate-count scaling experiments, and an error analysis of code execution.

Significance. If the results hold, the paper offers a practical recipe for bootstrapping chart understanding without manually labeled training answers or an external teacher model for data generation. The code-driven synthesis idea is elegant in that labels are derived from executed code rather than from model predictions, which reduces label noise. The candidate-conditioned answering module is also interesting and shows consistent gains across four models in the ablation-oriented comparison. The paper provides machine-executable pipeline details and makes code public, which supports reproducibility. However, the paper's central claim—fully self-improving without human-labeled data or external models—is qualified by its reliance on CharXiv's human-curated questions as seeds and on a GPT-4o judge for evaluation, and the reasoning-task gain is not yet convincingly separated from candidate selection rather than learned reasoning synthesis.

major comments (4)
  1. [Abstract / Section 3.1] The claim of a "fully self-improving paradigm without either human-labeled data or external models" is overstated. Section 3.1 states that the QA generation uses "the descriptive questions in CharXiv as seeds," and these questions are human-curated (Appendix A). Furthermore, Section 4.1 states that evaluation uses GPT-4o as an automated judge. While no human-annotated answers are used for training and no external model is used to produce training labels, the method does depend on a human-annotated question bank and an external model for evaluation. The authors should either temper the claim or clearly scope it to "no external model or human labels in the training-data generation loop" and explicitly acknowledge the dependence on CharXiv and GPT-4o in the abstract and conclusion.
  2. [Section 4.2 / Section 3.1] The reasoning-task improvement is not yet established as a learned reasoning ability. Synthetic training triplets are generated exclusively from descriptive CharXiv questions (Table 1), so the answer model MANS is never trained on a reasoning question-answer pair. Yet the main reasoning gain in Table 2 (from 23.60 to 31.60) is attributed to candidate-conditioned answering, while direct fine-tuning on the same synthetic data degrades reasoning to 21.90. The paper does not report whether the reasoning gains persist when all K test-time candidates are incorrect, nor does it measure how often the final answer is a verbatim selection or light paraphrase of a correct candidate. The case study in Table 5 shows the final answer elaborating the correct candidate (Candidate 4), which is consistent with selection-plus-paraphrase. To support the claim that the model learns to synthesize a correct answer from candidates, the authors should add an analysis that separates candidate selection from genuine answer synthesis, e.g., by reporting accuracy conditioned on the presence/absence of a correct candidate and by computing the fraction of final answers that contain substantial new content beyond the selected candidate.
  3. [Section 4.2 / Table 2] The absence of error bars, repeated seeds, or any statistical significance measure is a concern given the small margins between some baselines. For example, CoT (26.80) and Majority Voting (26.70) differ by only 0.10 points, and Self-Verification (28.30) is only 0.20 points above Majority Voting on reasoning. The claim that Chart-CoCa "surpasses all baselines by significant margins" is based on single runs. The authors should report variance across at least three seeds or provide a paired significance test, especially for the reasoning differences that are central to the paper's contribution.
  4. [Section 4.3 / Figure 4] The Pass@K curves in Figure 4 provide a useful ceiling analysis, but the paper does not explain how Pass@K is computed for open-ended answers. Specifically, it is unclear whether a candidate is considered correct based on exact string match, GPT-4o judgment, or another criterion. Since the final evaluation uses GPT-4o pairwise comparison, the Pass@K values should be computed under the same protocol to be comparable to the Chart-CoCa and majority-voting curves. The authors should specify the correctness criterion used for Pass@K.
minor comments (5)
  1. [Table 2] The header "Reasoning Quesitons" contains a typo; it should read "Reasoning Questions."
  2. [Section 4.3] The text contains a duplicated phrase: "Pass@K also rises significantly, improving from from 23.6% (K=1) to 55.8%." Remove the second "from."
  3. [Table 3 caption] The caption states "where the values in parentheses are the comparison with the initial model," but the table does not show any values in parentheses. Either add the relative gains or revise the caption.
  4. [Section 4.6 / Table 5] Table 5's final answer text is described as showing a reasoning path, but the path consists of a single clause referencing "the purple line in subplot e." The caption and surrounding text should more carefully characterize this as a minimal justification rather than an extensive chain of reasoning.
  5. [Section 3.1] The maximum number of code generation/execution attempts is informally described as "we simply set a maximum number of attempts" and later in Section 4.7 stated as 5 attempts. This value should be stated once, early in Section 3.1, rather than only appearing in the error analysis.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: synthetic answers come from code execution and evaluation is on an external benchmark, so the self-improvement loop is broken by an external verifier.

full rationale

The central derivation chain is not circular. Synthetic chart-question-answer triplets are produced by generating matplotlib code and executing it, then extracting answers from plotting-object attributes (Equations 1-4); the answer labels therefore come from deterministic code execution, not from the VLM's own predictions. The candidate-conditioned answering model is trained on these externally verified labels and evaluated on CharXiv's validation split, whose human-verified ground truth is independent of the synthetic training data. The only self-citation is [46], used for the generic claim that LLMs can rephrase questions while retaining meaning; this claim is also supported by external references [3,10] and is not load-bearing for the main result. Two legitimate concerns remain but are not circularity: the use of CharXiv test-set charts as unlabeled seeds and CharXiv descriptive question templates for question synthesis creates possible distributional overlap, and the transfer from descriptive-only synthetic training to reasoning questions is an unverified empirical assumption. Neither concern makes any reported accuracy equal its input by construction. The paper is self-contained against an external benchmark, so no circular step is exhibited.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The method rests on the validity of code-generated charts as training distribution, the reliability of the CharXiv and GPT-4o evaluation, and the transfer of candidate-conditioned answering to reasoning questions. These are domain assumptions rather than derived results.

free parameters (2)
  • K (number of answer candidates) = 5 for main experiments; varied from 1 to 30 in Figure 4
    Candidate-conditioned answering depends on K, and the headline gains use K=5, chosen for practical trade-offs rather than fitted to optimize the benchmark.
  • Maximum code generation and execution attempts = 5
    The synthesis pipeline retries code generation and execution up to 5 times; this hand-set engineering choice affects how much training data is produced.
assumptions (4)
  • domain assumption CharXiv validation answers and the GPT-4o pairwise evaluation protocol are reliable ground truth for measuring chart understanding correctness.
    The paper inherits this evaluation standard from CharXiv [37] and does not independently validate it.
  • domain assumption Executing generated Matplotlib code produces a chart whose visual content exactly matches the information extracted from the plotting objects during execution.
    This is the foundation of the noise-free label claim in Section 3.1 Code Execution; the chart and the extracted elements both derive from the same code, but the paper does not verify human-perceived chart-text alignment.
  • domain assumption Training on descriptive synthetic QA pairs, combined with candidate-conditioning, transfers to reasoning questions at test time.
    Direct fine-tuning on the same synthetic data degrades reasoning from 23.60 to 21.90, so the reasoning gain is attributed to candidate-conditioned answering, whose transfer mechanism is not proven (Section 4.2).
  • domain assumption The VLM itself is capable enough to generate descriptions and code that produce useful synthetic charts, with simple retries handling failures.
    The pipeline presupposes that a self-trained VLM can bootstrap from its own chart descriptions and code, and that up to 5 retries suffice to produce a meaningful dataset.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Chart-CoCa: Self-Improving Chart Understanding of Vision LMs via Code-Driven Synthesis and Candidate-Conditioned Answering." pith.science (2026). https://pith.science/paper/ZFSV7CZX

@misc{pith2026250811975,
  author       = {Pith},
  title        = {Pith review of: Chart-CoCa: Self-Improving Chart Understanding of Vision LMs via Code-Driven Synthesis and Candidate-Conditioned Answering},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZFSV7CZX}},
  note         = {Machine review of arXiv:2508.11975}
}
read the original abstract

Vision Language Models (VLMs) often struggle with chart understanding tasks, particularly in accurate chart description and complex reasoning. Synthetic data generation is a promising solution, while usually facing the challenge of noise labels. To address this challenge, we first introduce a chart synthesis pipeline that generates aligned chart-question-answer triplets through code generation and execution, ensuring the reliability of synthetic data without human intervention. Furthermore, inspired by test-time scaling that increases inference budget and thereby improves performance, we design a candidate-conditioned answering process. The VLM first generates multiple responses per query, and then synthesizes the final answer by contextualizing these candidates. Experiments demonstrate significant improvements, with up to 15.50 points accuracy gain over the initial VLM, in a fully self-improving paradigm without either human-labeled data or external models.

Figures

Figures reproduced from arXiv: 2508.11975 by the authors.

Figure 1
Figure 1. (a) Direct Chart-to-Text Generation. (b) Generating [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. The pipeline of data synthesis for VLM training. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. The process of candidate-conditioned generation. [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Performance of Pass@K, Majority Voting, and our [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: The left pie chart (a) illustrates the portions of dif [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: The frequency distribution of various error types [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

49 extracted references · 15 canonical work pages

  1. [1]

    Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Floren- cia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al. 2023. Gpt-4 technical report. arXiv preprint arXiv:2303.08774 (2023)

    arXiv 2023

  2. [2]

    Massih-Reza Amini, Vasilii Feofanov, Loic Pauletto, Lies Hadjadj, Emilie Devijver, and Yury Maximov. 2025. Self-training: A survey. Neurocomputing 616 (2025), 128904

    work page 2025

  3. [3]

    Jinze Bai, Shuai Bai, Shusheng Yang, Shijie Wang, Sinan Tan, Peng Wang, Junyang Lin, Chang Zhou, and Jingren Zhou. 2023. Qwen-vl: A frontier large vision- language model with versatile abilities. arXiv preprint arXiv:2308.12966 (2023)

    arXiv 2023

  4. [4]

    Tran, and Mehran Kazemi

    Hritik Bansal, Arian Hosseini, Rishabh Agarwal, Vinh Q. Tran, and Mehran Kazemi. 2024. Smaller, Weaker, Yet Better: Training LLM Reasoners via Compute- Optimal Sampling. In The 4th Workshop on Mathematical Reasoning and AI at NeurIPS’24. https://openreview.net/forum?id=HuYSURUxs2

    work page 2024

  5. [5]

    Zixiang Chen, Yihe Deng, Huizhuo Yuan, Kaixuan Ji, and Quanquan Gu. 2024. Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models. In ICML 2024

    work page 2024

  6. [6]

    Zixiang Chen, Yihe Deng, Huizhuo Yuan, Kaixuan Ji, and Quanquan Gu. 2024. Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models. In Forty-first International Conference on Machine Learning, ICML 2024, Vienna, Austria, July 21-27, 2024. OpenReview.net. https://openreview.net/forum? id=O4cHTxW9BS

    work page 2024

  7. [7]

    Zhe Chen, Weiyun Wang, Hao Tian, Shenglong Ye, Zhangwei Gao, Erfei Cui, Wenwen Tong, Kongzhi Hu, Jiapeng Luo, Zheng Ma, et al. 2024. How far are we to gpt-4v? closing the gap to commercial multimodal models with open-source suites. arXiv preprint arXiv:2404.16821 (2024)

    arXiv 2024

  8. [8]

    Karl Cobbe, Vineet Kosaraju, Mohammad Bavarian, Mark Chen, Heewoo Jun, Lukasz Kaiser, Matthias Plappert, Jerry Tworek, Jacob Hilton, Reiichiro Nakano, et al. 2021. Training verifiers to solve math word problems. arXiv preprint arXiv:2110.14168 (2021)

    arXiv 2021

Show all 49 references
  1. [9]

    Wenliang Dai, Junnan Li, Dongxu Li, Anthony Tiong, Junqi Zhao, Weisheng Wang, Boyang Li, Pascale Fung, and Steven Hoi. 2023. InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning. In Thirty- seventh Conference on Neural Information Processing Sys...

    2023

  2. [10]

    Yihe Deng, Weitong Zhang, Zixiang Chen, and Quanquan Gu. 2024. Rephrase and Respond: Let Large Language Models Ask Better Questions for Themselves. arXiv:2311.04205 [cs.CL] https://arxiv.org/abs/2311.04205

    2024 arXiv

  3. [11]

    Xiaoyi Dong, Pan Zhang, Yuhang Zang, Yuhang Cao, Bin Wang, Linke Ouyang, Xilin Wei, Songyang Zhang, Haodong Duan, Maosong Cao, et al. 2024. Internlm- xcomposer2: Mastering free-form text-image composition and comprehension in vision-language large model. arXiv preprint arXiv:2...

    2024 arXiv

  4. [12]

    Venkat Gudivada, Amy Apon, and Junhua Ding. 2017. Data quality consid- erations for big data and machine learning: Going beyond data cleaning and transformations. International Journal on Advances in Software 10, 1 (2017), 1–20

    2017

  5. [13]

    Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, et al . 2025. Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning. arXiv preprint arXiv:2501.12948 (2025)

    2025 arXiv

  6. [14]

    Jiayi He, Hehai Lin, Qingyun Wang, Yi Fung, and Heng Ji. 2024. Self-Correction is More than Refinement: A Learning Framework for Visual and Language Rea- soning Tasks. CoRR abs/2410.04055 (2024). doi:10.48550/ARXIV.2410.04055 arXiv:2410.04055

    2024 doi arXiv

  7. [15]

    Wei He, Zhiheng Xi, Wanxu Zhao, Xiaoran Fan, Yiwen Ding, Zifei Shan, Tao Gui, Qi Zhang, and Xuanjing Huang. 2024. Distill Visual Chart Reasoning Ability from LLMs to MLLMs. arXiv preprint arXiv:2410.18798 (2024)

    2024 arXiv

  8. [16]

    Kushal Kafle, Brian Price, Scott Cohen, and Christopher Kanan. 2018. Dvqa: Understanding data visualizations via question answering. In Proceedings of the IEEE conference on computer vision and pattern recognition . 5648–5656

    2018

  9. [17]

    Samira Ebrahimi Kahou, Vincent Michalski, Adam Atkinson, Ákos Kádár, Adam Trischler, and Yoshua Bengio. 2017. Figureqa: An annotated figure dataset for visual reasoning. arXiv preprint arXiv:1710.07300 (2017)

    2017 arXiv

  10. [18]

    Aitor Lewkowycz, Anders Andreassen, David Dohan, Ethan Dyer, Henryk Michalewski, Vinay Ramasesh, Ambrose Slone, Cem Anil, Imanol Schlag, Theo Gutman-Solo, et al. 2022. Solving quantitative reasoning problems with language models. Advances in Neural Information Processing Syste...

    2022

  11. [19]

    Hunter Lightman, Vineet Kosaraju, Yuri Burda, Harrison Edwards, Bowen Baker, Teddy Lee, Jan Leike, John Schulman, Ilya Sutskever, and Karl Cobbe. 2024. Let’s Verify Step by Step. In The Twelfth International Conference on Learning Represen- tations. ICLR, Vienna, Austria. http...

    2024

  12. [20]

    Fuxiao Liu, Xiaoyang Wang, Wenlin Yao, Jianshu Chen, Kaiqiang Song, Sangwoo Cho, Yaser Yacoob, and Dong Yu. 2024. MMC: Advancing Multimodal Chart Understanding with Large-scale Instruction Tuning. In Proceedings of the 2024 Conference of the North American Chapter of the Assoc...

    2024

  13. [21]

    Haotian Liu, Chunyuan Li, Yuheng Li, and Yong Jae Lee. 2024. Improved Baselines with Visual Instruction Tuning. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2024, Seattle, W A, USA, June 16-22, 2024 . IEEE, 26286– 26296. doi:10.1109/CVPR52733.2024.02484

    2024

  14. [22]

    Lin Long, Rui Wang, Ruixuan Xiao, Junbo Zhao, Xiao Ding, Gang Chen, and Haobo Wang. 2024. On LLMs-Driven Synthetic Data Generation, Curation, and Evaluation: A Survey. In Findings of the Association for Computational Linguistics ACL 2024, Lun-Wei Ku, Andre Martins, and Vivek S...

    2024 doi

  15. [23]

    Aman Madaan, Niket Tandon, Prakhar Gupta, Skyler Hallinan, Luyu Gao, Sarah Wiegreffe, Uri Alon, Nouha Dziri, Shrimai Prabhumoye, Yiming Yang, Shashank Gupta, Bodhisattwa Prasad Majumder, Katherine Hermann, Sean Welleck, Amir Yazdanbakhsh, and Peter Clark. 2023. Self-Refine: It...

    2023

  16. [24]

    Ahmed Masry, Do Xuan Long, Jia Qing Tan, Shafiq Joty, and Enamul Hoque. 2022. ChartQA: A Benchmark for Question Answering about Charts with Visual and Logical Reasoning. In Findings of the Association for Computational Linguistics: ACL 2022, Smaranda Muresan, Preslav Nakov, an...

    2022

  17. [26]

    Fanqing Meng, Wenqi Shao, Quanfeng Lu, Peng Gao, Kaipeng Zhang, Yu Qiao, and Ping Luo. 2024. Chartassisstant: A universal chart multimodal language model via chart-to-table pre-training and multitask instruction tuning. arXiv preprint arXiv:2401.02384 (2024)

    2024 arXiv

  18. [27]

    Nitesh Methani, Pritha Ganguly, Mitesh M Khapra, and Pratyush Kumar. 2020. Plotqa: Reasoning over scientific plots. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision . 1527–1536

    2020

  19. [28]

    Ajay Patel, Colin Raffel, and Chris Callison-Burch. 2024. DataDreamer: A Tool for Synthetic Data Generation and Reproducible LLM Workflows. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Lun-Wei Ku, Andre Mar...

    2024 doi

  20. [29]

    Neoklis Polyzotis, Sudip Roy, Steven Euijong Whang, and Martin Zinkevich

  21. [30]

    Samyam Rajbhandari, Jeff Rasley, Olatunji Ruwase, and Yuxiong He. 2020. ZeRO: memory optimizations toward training trillion parameter models. In Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2020, Virtual Event...

    2020

  22. [31]

    Charlie Snell, Jaehoon Lee, Kelvin Xu, and Aviral Kumar. 2024. Scaling llm test- time compute optimally can be more effective than scaling model parameters. arXiv preprint arXiv:2408.03314 (2024)

    2024 arXiv

  23. [32]

    doi:10.1109/SC41405.2020.00024

    2020 arXiv

  24. [33]

    Peng Wang, Shuai Bai, Sinan Tan, Shijie Wang, Zhihao Fan, Jinze Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, Yang Fan, Kai Dang, Mengfei Du, Xuancheng Ren, Rui Men, Dayiheng Liu, Chang Zhou, Jingren Zhou, and Junyang Lin. 2024. Qwen2-VL: Enhancing Vision-Language Mode...

    2024 arXiv

  25. [34]

    Hashimoto

    Rohan Taori, Ishaan Gulrajani, Tianyi Zhang, Yann Dubois, Xuechen Li, Carlos Guestrin, Percy Liang, and Tatsunori B. Hashimoto. 2023. Stanford Alpaca: An Instruction-following LLaMA model. https://github.com/tatsu-lab/stanford_ alpaca

    2023

  26. [35]

    Xuezhi Wang, Jason Wei, Dale Schuurmans, Quoc V Le, Ed H Chi, Sharan Narang, Aakanksha Chowdhery, and Denny Zhou. 2023. Self-Consistency Improves Chain of Thought Reasoning in Language Models. In The Eleventh International Conference on Learning Representations

    2023

  27. [36]

    Xiyao Wang, Jiuhai Chen, Zhaoyang Wang, Yuhang Zhou, Yiyang Zhou, Huaxiu Yao, Tianyi Zhou, Tom Goldstein, Parminder Bhatia, Furong Huang, and Cao Xiao. 2024. Enhancing Visual-Language Modality Alignment in Large Vision Language Models via Self-Improvement. CoRR abs/2405.15973 ...

    2024 doi arXiv

  28. [37]

    Zirui Wang, Mengzhou Xia, Luxi He, Howard Chen, Yitao Liu, Richard Zhu, Kaiqu Liang, Xindi Wu, Haotian Liu, Sadhika Malladi, Alexis Chevalier, Sanjeev Arora, and Danqi Chen. 2024. CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs.CoRR abs/2406.18521 (2...

    2024 doi arXiv

  29. [38]

    Yizhong Wang, Yeganeh Kordi, Swaroop Mishra, Alisa Liu, Noah A Smith, Daniel Khashabi, and Hannaneh Hajishirzi. 2023. Self-Instruct: Aligning Language Models with Self-Generated Instructions. In The 61st Annual Meeting Of The Association For Computational Linguistics . Chart-C...

    2023

  30. [39]

    Yixuan Weng, Minjun Zhu, Fei Xia, Bin Li, Shizhu He, Shengping Liu, Bin Sun, Kang Liu, and Jun Zhao. 2023. Large Language Models are Better Reasoners with Self-Verification. InFindings of the Association for Computational Linguistics: EMNLP 2023. 2550–2575

    2023

  31. [40]

    Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, et al. 2022. Chain-of-thought prompting elicits reasoning in large language models. Advances in neural information processing systems 35 (2022), 24824–24837

    2022

  32. [41]

    Renqiu Xia, Bo Zhang, Hancheng Ye, Xiangchao Yan, Qi Liu, Hongbin Zhou, Zijun Chen, Min Dou, Botian Shi, Junchi Yan, et al . 2024. Chartx & chartvlm: A versatile benchmark and foundation model for complicated chart reasoning. arXiv preprint arXiv:2402.12185 (2024)

    2024 arXiv

  33. [42]

    Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Clement Delangue, Anthony Moi, Pierric Cistac, Tim Rault, Rémi Louf, Morgan Funtowicz, et al

  34. [43]

    Zhengzhuo Xu, Sinan Du, Yiyan Qi, Chengjin Xu, Chun Yuan, and Jian Guo. 2023. Chartbench: A benchmark for complex visual reasoning in charts. arXiv preprint arXiv:2312.15915 (2023)

    2023 arXiv

  35. [44]

    Weizhe Yuan, Richard Yuanzhe Pang, Kyunghyun Cho, Xian Li, Sainbayar Sukhbaatar, Jing Xu, and Jason E Weston. 2024. Self-Rewarding Language Models. In Forty-first International Conference on Machine Learning . https: //openreview.net/forum?id=0NphYCmgua

    2024

  36. [45]

    Guowei Xu, Peng Jin, Hao Li, Yibing Song, Lichao Sun, and Li Yuan. 2024. LLaVA- CoT: Let Vision Language Models Reason Step-by-Step. arXiv:2411.10440 [cs.CV] https://arxiv.org/abs/2411.10440

    2024 arXiv

  37. [46]

    Meishan Zhang, Gongyao Jiang, Shuang Liu, Jing Chen, and Min Zhang. 2024. LLM-Assisted Data Augmentation for Chinese Dialogue-Level Dependency Pars- ing. Computational Linguistics 50, 3 (Sept. 2024), 867–891. doi:10.1162/coli_a_ 00515

    2024 doi

  38. [47]

    Xuanle Zhao, Xianzhen Luo, Qi Shi, Chi Chen, Shuo Wang, Wanxiang Che, Zhiyuan Liu, and Maosong Sun. 2025. ChartCoder: Advancing Multimodal Large Language Model for Chart-to-Code Generation

    2025

  39. [48]

    Simon Zhai, Hao Bai, Zipeng Lin, Jiayi Pan, Peter Tong, Yifei Zhou, Alane Suhr, Saining Xie, Yann LeCun, Yi Ma, and Sergey Levine. 2024. Fine- Tuning Large Vision-Language Models as Decision-Making Agents via Re- inforcement Learning. In Advances in Neural Information Processi...

    2024

  40. [2018]

    ACM SIGMOD Record 47, 2 (2018), 17–28

    Data lifecycle challenges in production machine learning: a survey. ACM SIGMOD Record 47, 2 (2018), 17–28

    2018

  41. [2019]

    arXiv preprint arXiv:1910.03771 (2019)

    Huggingface’s transformers: State-of-the-art natural language processing. arXiv preprint arXiv:1910.03771 (2019)

    2019 arXiv

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.