Pith. sign in

REVIEW 4 major objections 4 minor 30 references

Graph-Based Multimodal Contrastive Learning for Chart Question Answering

T0 review · 4 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read The paper claims that injecting a contrastively trained multimodal scene-graph layer as a decoder soft prompt materially improves chart question answering, reporting a 9.04-point ChartQA gain and a 7.59-point OpenCQA gain over the…

desk verdict Plausible method, unproven headline: the GCL gains aren't causally identifiable without a matched no-graph baseline, and the authors' own ablation says visual-only is often best. read the letter →

arxiv 2501.04303 v2 pith:H65BIUUF submitted 2025-01-08 cs.CL

Yue Dai , Soyeon Caren Han , Wei Liu This is my paper · ORCID
classification cs.CL
keywords chartquestionansweringmultimodalscenegraphcontrastivelearningsoftpromptchain-of-thoughtpromptinghallucinationmitigationunderstanding
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 question answering asks a model to read a chart and answer questions that often require locating values and comparing components, something patch-based vision encoders do poorly because they blur object-level detail. The paper proposes to recover that detail explicitly: it detects chart objects, builds a visual scene graph of spatial neighbours and a textual scene graph of labels and OCR text, aligns the two graphs with contrastive learning, and inserts the aligned graph representations into a transformer decoder as a soft prompt. The central claim is that this graph layer is what lifts performance, and the paper reports gains up to 9.04 points on ChartQA and 7.59 BLEU points on OpenCQA over the released UniChart checkpoint, alongside smaller gains over the original UniChart results. The paper also tries chain-of-thought prompting for multimodal LLMs, observes that it helps on the more diverse ChartX benchmark but not consistently on ChartQA, and takes that as evidence that explicit graph structure still has a role to play.

What carries the argument

The load-bearing object is the multimodal scene-graph soft prompt: a pair of graphs, one visual and one textual, whose node representations are aligned by graph contrastive learning and then injected into a text decoder as trainable prompt tokens. The visual graph encodes spatial proximity between detected chart objects using k-nearest-neighbour edges and distance-based edge weights; the textual graph encodes labels and OCR text. Two graph convolutional encoders compute node representations, a contrastive InfoNCE-style objective maximizes similarity between nodes representing the same object across modalities, and the aligned textual node representations replace special graph tokens prepended to the question, turning the graph into a soft prompt without modifying the backbone encoder. This mechanism is what the paper credits for the reported performance gains, and it is also what the ablations probe when they remove one graph or add intra-modality contrastive loss.

What would settle it

Scramble the pairing between visual and textual graph nodes before contrastive training and rerun the same pipeline: if chart QA performance stays at the same level as the correctly aligned model, then cross-modal alignment is not the cause of the gain. Alternatively, build the graphs from ground-truth chart tables instead of detector and OCR output; if the improvement disappears, the bottleneck is perception quality rather than graph reasoning.

Watch

Extended reading notes

Core claim

The central claim, stated on the paper's own terms, is that multimodal graph contrastive learning improves chart question answering when the learned graph is delivered to the decoder as a soft prompt. In the proposed architecture, a visual graph connects each detected object to its three nearest neighbours with edge weights that decay with bounding-box distance, while a textual graph connects chart labels and OCR text; two graph convolutional encoders produce node representations, and a contrastive objective pulls together node pairs that refer to the same chart component across the two modalities. The textual node representations are then written into special graph tokens prepended to the question, giving the decoder structured information about chart components and their relationships. The paper reports that this addition improves both UniChart and VL-T5 on ChartQA and OpenCQA in most settings, that it outperforms the released UniChart checkpoint by 9.04 points on ChartQA and 7.59 points on OpenCQA, and that dropping either graph or adding intra-modality contrastive loss tends to reduce performance, which the paper reads as evidence that the cross-modal alignment is the active ingredient.

Load-bearing premise

The method assumes that detected chart objects and OCR-extracted text can be aligned into trustworthy visual-textual node pairs; if the detector or OCR produces wrong labels or positions, the contrastive loss connects mismatched components and the graph prompt carries corrupted structure into the decoder.

Editorial extensions

If this is right

  • If the graph soft prompt is the cause of the gains, existing chart QA backbones can be upgraded by adding a contrastively trained scene-graph layer without retraining from scratch.
  • The ablations imply that inter-modality alignment, not extra intra-modality structure, is what helps, so future graph-based chart models should concentrate on cross-modal node alignment.
  • Since adding intra-modality contrastive loss hurts both backbones, the reported results argue against simply stacking more contrastive objectives in generation tasks.
  • The CoT experiments indicate that prompting alone is unreliable for chart reasoning on diverse chart types, pointing to explicit graph structure as a complementary route to reduce hallucination.

Reading between the lines

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

  • One testable extension is to build the graphs from the chart's underlying data table rather than from detected objects and OCR text; if the gain grows, perception noise is the current bottleneck.
  • The design does not isolate the contrastive objective from the graph prompt itself, so a comparison against a graph prompt trained with plain reconstruction or matching loss would reveal whether the contrastive alignment is necessary or merely sufficient.
  • The same patch-loss problem exists for tables, infographics, and other structured documents, so the decoder soft-prompt mechanism may transfer beyond charts if the node alignment can be supplied from structured sources.
  • The ChartX results with the chart-to-table prompt suggest that grouping results by chart type could identify which chart families still need perceptual structure rather than reasoning prompts.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 4 minor

Summary. This manuscript proposes a multimodal scene-graph framework for chart question answering. The method extracts visual objects with Mask R-CNN and text with Google OCR, constructs visual and textual graphs, aligns their node representations with a graph contrastive InfoNCE objective, and injects the learned graph representations into the decoder of a vision-language backbone as soft prompts. The authors evaluate the approach on ChartQA and OpenCQA for fine-tuned backbones (UniChart, VL-T5) and on ChartQA and ChartX for zero-shot MLLMs with chain-of-thought prompts. They report gains over published baselines, including up to a 9.04-point improvement on ChartQA and 7.59 on OpenCQA for UniChart relative to a Hugging Face checkpoint, and they present ablations on intra-modality contrastive learning and on visual versus textual graph prompts.

Significance. If the central claim were supported, the contribution would be useful: injecting graph-structured scene information as decoder soft prompts is a plausible way to recover object-level information lost by patch-based encoders, and the GCL objective offers a principled alignment mechanism. The paper also provides a compact CoT-prompting study for MLLMs on chart tasks, which is of practical interest. The framework is described in enough detail to reimplement in broad strokes, and the ablation structure shows awareness of the key design choices. However, the experimental design does not currently isolate the effect of the proposed graph layer and GCL from the effect of fine-tuning itself, and the ablations partly contradict the multimodal-alignment premise.

major comments (4)
  1. [Section 3.2, Table 2] The central claim that graph contrastive learning 'significantly boosts the performance of backbone models' is not causally identifiable from the reported comparisons. UniChart(O) is compared against UniChart's original-paper result, a Hugging Face checkpoint, and the previous MSG-Chart system of reference [2], none of which was trained under the authors' fine-tuning protocol. The differences of 9.04 points on ChartQA and 7.59 points on OpenCQA are computed against the checkpoint, not against a matched no-graph baseline. The row labeled UniChart[2] is the only graph-without-GCL comparison, but the paper does not state whether it was rerun under identical conditions or is quoted from reference [2]. A matched baseline consisting of the same backbone fine-tuned on the same splits with the same optimizer, epochs, batch size, learning rate, and seeds, but without any graph prompt or GCL loss, is required to attribute the gains to the proposed method.
  2. [Section 3.3, Table 5] The full multimodal VG+TG configuration is not consistently the best, which undercuts the premise that cross-modal alignment is what drives the improvement. On OpenCQA, visual-graph-only prompts outperform the full multimodal prompt for both backbones (16.68 vs 16.35 for UniChart; 19.03 vs 18.42 for VL-T5), and on VL-T5 ChartQA the visual-only variant is better on both the human set (38.16 vs 34.96) and the average (65.04 vs 63.60). The authors themselves concede that predicted labels from Mask R-CNN are 'highly noisy, particularly on OpenCQA.' Since the positive pairs in Eq. (2) are defined by aligning detected objects with OCR-derived textual nodes, noisy detections can corrupt the contrastive objective. The manuscript should provide evidence that the node alignments are reliable, for example by measuring alignment accuracy against ground-truth chart annotations or by ablating with ground-truth object locations.
  3. [Section 3.2 and Section 3.3] All reported scores are single numbers with no variance or significance testing. Several key comparisons are small (e.g., 67.44 vs 66.64 average on ChartQA; 16.35 vs 16.38 BLEU on OpenCQA in Table 5), and without repeated runs or error bars these differences are not distinguishable from noise. The paper should report means and standard deviations over multiple seeds and, where appropriate, a significance test for the headline gains in Table 2.
  4. [Section 2.4 and Table 5] The decoder integration is described only for the textual graph: the paper says the embeddings of 36 special tokens are replaced with 'object node representations from the textual graph.' Yet Table 5 reports prompts built from the visual graph only, and the abstract claims a joint multimodal graph is integrated. The manuscript should specify how the visual-graph representation is injected into the decoder and whether the 36-token prompt is used for both modalities, as this is essential to reproducing the reported configurations.
minor comments (4)
  1. [Abstract and Section 3.1] The abstract says evaluations include ChartX, but Section 3.1 clarifies that GCL is evaluated on ChartQA and OpenCQA while ChartX is used only for CoT prompting of MLLMs; this distinction should be stated more explicitly in the abstract or introduction.
  2. [Section 3.2] The phrase '9.04% increase' refers to an absolute percentage-point difference in relaxed accuracy (67.44 vs 58.40), not a relative increase; the paper should use 'percentage points' or give relative percentages for clarity.
  3. [Section 2.2 and Section 2.3] Hyperparameters for the graph and contrastive components are only partially specified: K is stated as three in the text but not defined as a hyperparameter, and the temperature tau in Eq. (1), GCN hidden dimensions, number of training epochs, learning rate, and batch size are not reported. Reporting these values is necessary for reproducibility.
  4. [General] The manuscript contains ACM template placeholder text, including the CCS Concepts line 'Do Not Use This Code' and the keywords 'Do, Not, Us, This, Code...'; these must be replaced with the correct metadata.

Circularity Check

0 steps flagged · score 0.0 of 10

No definitional circularity found; the central comparison is against external benchmarks, though causal attribution would need a matched no-graph baseline.

full rationale

The paper's claims are experimental rather than derivational: it fine-tunes UniChart and VL-T5 with a graph layer and graph contrastive learning, then evaluates on public benchmarks (ChartQA, OpenCQA, ChartX). The reported improvements are relative to the original UniChart paper results and a Hugging Face checkpoint, not to quantities defined in terms of the method's own outputs. The graph construction is indeed taken from the authors' prior MSG-Chart paper [2], and Table 2 reuses rows from that self-cited work as baselines; however, this is a normal comparison to prior work, not a case where a prediction is equivalent to its input by construction. Section 2.3's contrastive objective is standard InfoNCE, and the positive pairs are defined by detected object alignment. The paper explicitly concedes in Section 3.3 that Mask R-CNN labels are noisy on OpenCQA, but that is a data-quality limitation, not circularity. The main weakness is that no matched no-graph fine-tuning baseline is provided, so the statement that GCL 'significantly boosts' performance is causally underdetermined; that is an experimental-control concern, not a definitional or self-citation circularity. Under the stated rules, no circular step can be quoted and exhibited, so the honest finding is no significant circularity.

Assumptions & free parameters 4 free parameters · 6 assumptions · 0 invented entities

The model's operation relies on pretrained perception components (Mask R-CNN, OCR, BERT, backbone) and standard graph contrastive learning assumptions. Several hyperparameters (K, temperature, prompt-token count, edge-drop ratio) are chosen by hand and not ablated. No new physical entities are introduced; the multimodal graph prompt is an internal architectural construct.

free parameters (4)
  • K (number of nearest neighbors in visual graph) = 3
    Chosen without reported sensitivity analysis; controls message passing topology in G_v (Section 2.2).
  • InfoNCE temperature tau
    Appears in Eq. 1 and controls cross-modal alignment sharpness, but its value is never reported.
  • Number of decoder graph prompt tokens = 36
    The 36 <G> tokens (Section 2.4) are set without justification or ablation; this capacity may affect how much graph information reaches the decoder.
  • Edge drop ratio p for intra-modality contrastive augmentation = 0.30
    Used in ablation (Section 3.3) to create augmented graphs; fixed at 0.30 without sensitivity analysis.
assumptions (6)
  • domain assumption Mask R-CNN fine-tuned by [15] detects chart objects with usable bounding boxes.
    Section 2.1: all visual nodes come from these detections; failures propagate to the graph.
  • domain assumption Google OCR returns accurate text that can be aligned to detected objects.
    Section 2.1: textual features come from OCR and labels; alignment is needed to build positive pairs.
  • domain assumption BERT CLS embeddings carry enough semantic information for chart labels and OCR text.
    Section 2.1: node initialization uses BERT CLS; no chart-specific verification.
  • standard math InfoNCE contrastive loss with same-object pairs yields semantically aligned cross-modal node representations.
    Section 2.3, Eqs. 1-2: relies on the standard assumption that contrastive learning aligns positive pairs, but no analysis shows the alignment is correct for chart objects.
  • domain assumption Pre-trained backbones (UniChart, VL-T5) can be fine-tuned with an added graph soft prompt without breaking their internal representations.
    Section 2.4: the prompt tokens replace 36 input embeddings; the paper does not study interference with pretraining.
  • standard math GCN encoders produce useful node representations for the visual and textual graphs.
    Section 2.3: GCN is treated as a black box encoder; no architecture details or validation provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Graph-Based Multimodal Contrastive Learning for Chart Question Answering." pith.science (2026). https://pith.science/paper/H65BIUUF

@misc{pith2026250104303,
  author       = {Pith},
  title        = {Pith review of: Graph-Based Multimodal Contrastive Learning for Chart Question Answering},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/H65BIUUF}},
  note         = {Machine review of arXiv:2501.04303}
}
read the original abstract

Chart question answering (ChartQA) is challenged by the heterogeneous composition of chart elements and the subtle data patterns they encode. This work introduces a novel joint multimodal scene graph framework that explicitly models the relationships among chart components and their underlying structures. The framework integrates both visual and textual graphs to capture structural and semantic characteristics, while a graph contrastive learning strategy aligns node representations across modalities enabling their seamless incorporation into a transformer decoder as soft prompts. Moreover, a set of tailored Chain of Thought (CoT) prompts is proposed to enhance multimodal large language models (MLLMs) in zero-s ot scenarios by mitigating hallucinations. Extensive evaluations on benchmarks including ChartQA, OpenCQA, and ChartX demonstrate significant performance improvements and validate the efficacy of the proposed approach.

Figures

Figures reproduced from arXiv: 2501.04303 by the authors.

Figure 1
Figure 1. Multimodal Graph Contrastive Learning Prompt [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Multimodal Graph Contrastive Learning Prompt [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

30 extracted references · 10 canonical work pages

  1. [2]

    Yue Dai, Soyeon Caren Han, and Wei Liu. 2024. MSG-Chart: Multimodal Scene Graph for ChartQA. In Proceedings of the 33rd ACM International Conference on Information and Knowledge Management . 3709–3713

    work page 2024

  2. [1]

    Jaemin Cho, Jie Lei, Hao Tan, and Mohit Bansal. 2021. Unifying Vision-and- Language Tasks via Text Generation. In Proceedings of the 38th International Conference on Machine Learning, ICML 2021, 18-24 July 2021, Virtual Event (Pro- ceedings of Machine Learning Research, Vol. 139) , Marina Meila and Tong Zhang (Eds.). PMLR, 1931–1942. http://proceedings.ml...

    work page 2021

  3. [3]

    Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Associa- tion for Computational Linguistics: Human Language Technologies, NAACL-HLT 2019, Minneapolis, MN, USA, June 2-7, 2019, ...

    2019

  4. [4]

    Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ah- mad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Amy Yang, Angela Fan, Anirudh Goyal, Anthony Hartshorn, Aobo Yang, Archi Mitra, Archie Sra- vankumar, Artem Korenev, Arthur Hinsvark, Arun Rao, Aston Zhang, Aurélien Rodriguez, Austen Gregerson, Ava Spataru, Baptiste Rozière, Bet...

    arXiv doi:10.48550/arxiv.2407.21783 2024

  5. [5]

    Difei Gao, Ke Li, Ruiping Wang, Shiguang Shan, and Xilin Chen. 2020. Multi- Modal Graph Neural Network for Joint Reasoning on Vision and Scene Text. In 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2020, Seattle, W A, USA, June 13-19, 2020 . Computer Vision Foundation / IEEE, 12743–12753. doi:10.1109/CVPR42600.2020.01276

    arXiv 2020

  6. [6]

    Yucheng Han, Chi Zhang, Xin Chen, Xu Yang, Zhibin Wang, Gang Yu, Bin Fu, and Hanwang Zhang. 2023. Chartllama: A multimodal llm for chart understanding and generation. arXiv preprint arXiv:2311.16483 (2023)

    arXiv 2023

  7. [7]

    Girshick

    Kaiming He, Georgia Gkioxari, Piotr Dollár, and Ross B. Girshick. 2017. Mask R-CNN. In IEEE International Conference on Computer Vision, ICCV 2017, Venice, Italy, October 22-29, 2017. IEEE Computer Society, 2980–2988. doi:10.1109/ICCV. 2017.322

    doi:10.1109/iccv 2017

  8. [8]

    Ziwei Ji, Nayeon Lee, Rita Frieske, Tiezheng Yu, Dan Su, Yan Xu, Etsuko Ishii, Yejin Bang, Andrea Madotto, and Pascale Fung. 2023. Survey of Hallucination in Natural Language Generation. ACM Comput. Surv. 55, 12 (2023), 248:1–248:38. doi:10.1145/3571730

    doi:10.1145/3571730 2023

Show all 30 references
  1. [9]

    Shankar Kantharaj, Xuan Long Do, Rixie Tiffany Ko Leong, Jia Qing Tan, Enamul Hoque, and Shafiq R. Joty. 2022. OpenCQA: Open-ended Question Answering with Charts. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, EMNLP 2022, Abu Dhabi, ...

    2022 doi

  2. [10]

    Zhuowan Li, Bhavan Jasani, Peng Tang, and Shabnam Ghadar. 2024. Synthesize Step-by-Step: Tools, Templates and LLMs as Data Generators for Reasoning-Based Chart VQA. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2024, Seattle, W A, USA, June 16-22, 202...

    2024

  3. [11]

    Fangyu Liu, Julian Martin Eisenschlos, Francesco Piccinno, Syrine Krichene, Chenxi Pang, Kenton Lee, Mandar Joshi, Wenhu Chen, Nigel Collier, and Yasemin Altun. 2023. DePlot: One-shot visual language reasoning by plot-to-table trans- lation. In Findings of the Association for ...

    2023 doi

  4. [12]

    Fangyu Liu, Francesco Piccinno, Syrine Krichene, Chenxi Pang, Kenton Lee, Mandar Joshi, Yasemin Altun, Nigel Collier, and Julian Martin Eisenschlos. 2023. MatCha: Enhancing Visual Language Pretraining with Math Reasoning and Chart Derendering. In Proceedings of the 61st Annual...

    2023

  5. [13]

    Jintao Liu, Kaiwen Wei, and Chenglong Liu. 2024. Multimodal Event Causality Reasoning with Scene Graph Enhanced Interaction Network. In Thirty-Eighth AAAI Conference on Artificial Intelligence, AAAI 2024, Thirty-Sixth Conference on Innovative Applications of Artificial Intelli...

    2024

  6. [14]

    Ahmed Masry, Parsa Kavehzadeh, Do Xuan Long, Enamul Hoque, and Shafiq Joty. 2023. UniChart: A Universal Vision-language Pretrained Model for Chart Comprehension and Reasoning. InProceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, EMNLP 2023,...

    2023

  7. [15]

    Joty, and Enamul Hoque

    Ahmed Masry, Do Xuan Long, Jia Qing Tan, Shafiq R. Joty, and Enamul Hoque

  8. [16]

    Fanqing Meng, Wenqi Shao, Quanfeng Lu, Peng Gao, Kaipeng Zhang, Yu Qiao, and Ping Luo. 2024. ChartAssistant: A Universal Chart Multimodal Language Model via Chart-to-Table Pre-training and Multitask Instruction Tuning. In Findings of the Association for Computational Linguisti...

    2024

  9. [17]

    Khapra, and Pratyush Kumar

    Nitesh Methani, Pritha Ganguly, Mitesh M. Khapra, and Pratyush Kumar. 2020. PlotQA: Reasoning over Scientific Plots. InIEEE Winter Conference on Applications of Computer Vision, W ACV 2020, Snowmass Village, CO, USA, March 1-5, 2020 . IEEE, 1516–1525. doi:10.1109/WACV45572.202...

    2020

  10. [18]

    Matt Post. 2018. A Call for Clarity in Reporting BLEU Scores. InProceedings of the Third Conference on Machine Translation: Research Papers, WMT 2018, Belgium, Brussels, October 31 - November 1, 2018 , Ondrej Bojar, Rajen Chatterjee, Christian Federmann, Mark Fishel, Yvette Gr...

    2018

  11. [19]

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, and Ilya Sutskever. 2021. Learning Transferable Visual Models From Natural Language Supervision. In Proceedings...

    2021

  12. [20]

    Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, and Peter J. Liu. 2020. Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer.J. Mach. Learn. Res. 21 (2020), 140:1–140:67. http://jmlr.org/p...

    2020

  13. [21]

    Tang, Angie Boggust, and Arvind Satyanarayan

    Benny J. Tang, Angie Boggust, and Arvind Satyanarayan. 2023. VisText: A Benchmark for Semantically Rich Chart Captioning. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2023, Toronto, Canada, July 9-14, 2...

    2023 doi

  14. [22]

    Hamilton, Pietro Liò, Yoshua Bengio, and R

    Petar Velickovic, William Fedus, William L. Hamilton, Pietro Liò, Yoshua Bengio, and R. Devon Hjelm. 2019. Deep Graph Infomax. In 7th International Conference on Learning Representations, ICLR 2019, New Orleans, LA, USA, May 6-9, 2019 . OpenReview.net. https://openreview.net/f...

    2019

  15. [23]

    Peng Wang, Shuai Bai, Sinan Tan, Shijie Wang, Zhihao Fan, Jinze Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, et al. 2024. Qwen2-vl: Enhancing vision-language model’s perception of the world at any resolution.arXiv preprint arXiv:2409.12191 (2024). Conference acronym ’...

    2024 arXiv

  16. [24]

    Zhecan Wang, Haoxuan You, Liunian Harold Li, Alireza Zareian, Suji Park, Yiqing Liang, Kai-Wei Chang, and Shih-Fu Chang. 2022. SGEITL: Scene Graph Enhanced Image-Text Learning for Visual Commonsense Reasoning. In Thirty-Sixth AAAI Conference on Artificial Intelligence, AAAI 20...

    2022

  17. [25]

    Chi, Quoc V

    Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed H. Chi, Quoc V. Le, and Denny Zhou. 2022. Chain-of-Thought Prompt- ing Elicits Reasoning in Large Language Models. In Advances in Neural Infor- mation Processing Systems 35: Annual Conference on ...

    2022

  18. [26]

    Renqiu Xia, Bo Zhang, Hancheng Ye, Xiangchao Yan, Qi Liu, Hongbin Zhou, Zijun Chen, Min Dou, Botian Shi, Junchi Yan, and Yu Qiao. 2024. ChartX & ChartVLM: A Versatile Benchmark and Foundation Model for Complicated Chart Reasoning. CoRR abs/2402.12185 (2024). doi:10.48550/ARXIV...

    2024 doi arXiv

  19. [27]

    Yanqiao Zhu, Yichen Xu, Qiang Liu, and Shu Wu. 2021. An Empirical Study of Graph Contrastive Learning. In Proceedings of the Neural Information Pro- cessing Systems Track on Datasets and Benchmarks 1, NeurIPS Datasets and Benchmarks 2021, December 2021, virtual , Joaquin Vansc...

    2021

  20. [28]

    Yanqiao Zhu, Yichen Xu, Feng Yu, Qiang Liu, Shu Wu, and Liang Wang. 2021. Graph Contrastive Learning with Adaptive Augmentation. In WWW ’21: The Web Conference 2021, Virtual Event / Ljubljana, Slovenia, April 19-23, 2021 , Jure Leskovec, Marko Grobelnik, Marc Najork, Jie Tang,...

    2021

  21. [2022]

    InFindings of the Association for Computational Linguistics: ACL 2022, Dublin, Ireland, May 22-27, 2022 , Smaranda Muresan, Preslav Nakov, and Aline Villavicencio (Eds.)

    ChartQA: A Benchmark for Question Answering about Charts with Visual and Logical Reasoning. InFindings of the Association for Computational Linguistics: ACL 2022, Dublin, Ireland, May 22-27, 2022 , Smaranda Muresan, Preslav Nakov, and Aline Villavicencio (Eds.). Association fo...

    2022

  22. [2279]

    doi:10.18653/V1/2022.FINDINGS-ACL.177

    2022 doi

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.