---
title: "Mitigating Coordinate Prediction Bias from Positional Encoding"
description: "The authors show this drift is not random. 51 references. Among them: 0."
authors: ["Xingjian Tao", "Yiwei Wang", "Yujun Cai", "Yihong Luo", "Kai Han", "Jing Tang"]
year: 2026
venue: "Findings of the Association for Computational Linguistics: ACL 2026"
source: "https://aclanthology.org/2026.findings-acl.1034/"
doi: "10.18653/v1/2026.findings-acl.1034"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/tao-2026-mitigating-coordinate-prediction-bias-positional/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/tao-2026-mitigating-coordinate-prediction-bias-positional/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/41/) / Mitigating Coordinate Prediction Bias from Positional Encoding Failures

# Mitigating Coordinate Prediction Bias from Positional Encoding Failures

**Authors:** Xingjian Tao, [Yiwei Wang](https://md.research.elamur.ai/free/en/authors/yiwei-wang/), [Yujun Cai](https://md.research.elamur.ai/free/en/authors/yujun-cai/), Yihong Luo, [Kai Han](https://md.research.elamur.ai/free/en/authors/kai-han/) and [Jing Tang](https://md.research.elamur.ai/free/en/authors/jing-tang/) · **Year:** 2026 · **Venue:** [Findings of the Association for Computational Linguistics: ACL 2026](https://md.research.elamur.ai/free/en/venues/findings/) ([Findings of the Association for Computational Linguistics 2026](https://md.research.elamur.ai/free/en/venues/findings/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.findings-acl.1034/) · [PDF](https://aclanthology.org/2026.findings-acl.1034.pdf) · [DOI 10.18653/v1/2026.findings-acl.1034](https://doi.org/10.18653/v1/2026.findings-acl.1034) · [OpenAlex](https://openalex.org/W7166815036)

**Citations:** 0 citations (OpenAlex, 2026-10-14)

## Summary
Multimodal LLMs that output click coordinates as digit strings often drift on high-resolution screenshots, because visual positional encodings (VPEs) lose grounding strength in long visual contexts. The authors show this drift is not random. When VPEs are deliberately shuffled, models collapse toward repeatable, directional coordinate guesses. These guesses resemble the clustered errors seen on real high-resolution inputs. Using a causal graph, they interpret these default numeric habits as a spurious pathway that dominates once positional signals weaken.

Their remedy, Vision-PE Shuffle Guidance (VPSG), requires no training. At inference time it runs several auxiliary passes with shuffled VPEs and combines them in log space to estimate the position-free prior. It then uses this estimate as negative evidence against the main pass's logits. A small finite-state machine restricts the correction to digit tokens. The guidance weight shrinks across successive digits and restarts for the y value.

On ScreenSpot-Pro, VPSG raises localization accuracy for Qwen2.5-VL models at moderate extra latency. This makes it a practical plug-in for GUI agents.

## Key points
- **Method components:** VPSG combines three elements: multi-seed shuffled routes aggregated as a geometric mean, an FSM-gated correction applied only to digits, and a geometrically decaying, per-coordinate coefficient schedule. It is model-agnostic and works without fine-tuning or architectural changes.
- **Evaluation:** experiments use the ScreenSpot-Pro high-resolution GUI grounding benchmark with Qwen2.5-VL-3B and Qwen2.5-VL-7B.
- **Ablation on coefficient decay:** removing it lowers average accuracy from 13.3 to 11.9 (3B) and from 19.1 to 18.2 (7B). This drop is larger than the one from removing seed aggregation.
- **Latency:** with S = 3 auxiliary routes, per-case time grows from 3.64s to 5.27s (3B) and from 4.20s to 5.50s (7B). That is roughly a 1.3–1.45× slowdown, well below the naive (S+1)× estimate.
- **Limitations:** only a few, smaller models were tested, malformed outputs are not handled in scoring, and experiments are English-only. Code is available at github.com/taoxj2001/VPSG.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

- [Hong Kong University of Science and Technology](https://md.research.elamur.ai/free/en/universities/hong-kong-university-of-science-and-technology/) — University, Hong Kong, China · 170 works · 26,341 citations
- [University of California, Los Angeles](https://md.research.elamur.ai/free/en/universities/university-of-california-los-angeles/) — University, Los Angeles, United States · 99 works · 56,333 citations
- [The Hong Kong University of Science and Technology (Guangzhou)](https://md.research.elamur.ai/free/en/universities/the-hong-kong-university-of-science-and-technology-guangzhou/) — University, Guangzhou, China · 67 works · 1,765 citations
- [The University of Queensland](https://md.research.elamur.ai/free/en/universities/the-university-of-queensland/) — University, Brisbane, Australia · 14 works · 4,603 citations
- [Shanghai University of Finance and Economics](https://md.research.elamur.ai/free/en/universities/shanghai-university-of-finance-and-economics/) — University, Shanghai, China · 15 works · 825 citations

📚 Shelves (2): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [👁️ Computer Vision and Pattern Recognition](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-vision-and-pattern-recognition/) · 3,486 works

## Cite

Cite as: Xingjian Tao, Yiwei Wang, Yujun Cai, Yihong Luo, Kai Han and Jing Tang. 2026. Mitigating Coordinate Prediction Bias from Positional Encoding Failures. Findings of the Association for Computational Linguistics: ACL 2026.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{tao2026mitigating,
  title = {{Mitigating Coordinate Prediction Bias from Positional Encoding Failures}},
  author = {Xingjian Tao, Yiwei Wang, Yujun Cai, Yihong Luo, Kai Han and Jing Tang},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.1034},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.1034}
}
```

- [← previous Mitigating Context Interference for Reliable and Efficient Search Agents](https://md.research.elamur.ai/free/en/papers/xue-2026-mitigating-context-interference-reliable-efficient/)
- [next → Mitigating Cultural Bias in LLMs via Multi-Agent Cultural Debate](https://md.research.elamur.ai/free/en/papers/tan-2026-mitigating-cultural-bias-llms-multi/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/41/)

## References

51 references, 22 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al. 2023. [Gpt-4 technical report](https://md.research.elamur.ai/free/en/papers/openai-2023-gpt-4-technical-report/). arXiv preprint arXiv:2303.08774.
2. Jean-Baptiste Alayrac, Jeff Donahue, Pauline Luc, Antoine Miech, Iain Barr, Yana Hasson, Karel Lenc, Arthur Mensch, Katherine Millican, Malcolm Reynolds, et al. 2022. [Flamingo: a visual language model for few-shot learning](https://md.research.elamur.ai/free/en/papers/alayrac-2022-flamingo-visual-language-model-few/). Advances in neural information processing systems, 35:23716–23736.
3. Ebtesam Almazrouei, Hamza Alobeidli, Abdulaziz Alshamsi, Alessandro Cappelli, Ruxandra Cojocaru, Merouane Debbah, Etienne Goffinet, Daniel Heslow, Julien Launay, Quentin Malartic, Badreddine Noune, Baptiste Pannier, and Guilherme Penedo. 2023. Falcon-40B: an open large language model with state-of-t… [find](https://scholar.google.com/scholar?q=Ebtesam+Almazrouei%2C+Hamza+Alobeidli%2C+Abdulaziz+Alshamsi%2C+Alessandro+Cappelli%2C+Ruxandra+Cojocaru%2C+Merouane+Debbah%2C+Etienne+Goffinet%2C+Daniel+Heslow%2C+Julien+Launay%2C+Quentin+Malartic%2C+Badreddine+Noune%2C+Ba)
4. Jinze Bai, Shuai Bai, Yunfei Chu, Zeyu Cui, Kai Dang, Xiaodong Deng, Yang Fan, Wenbin Ge, Yu Han, Fei Huang, et al. 2023. [Qwen technical report](https://md.research.elamur.ai/free/en/papers/bai-2023-qwen-technical-report/). arXiv preprint arXiv:2309.16609.
5. Shuai Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, Sibo Song, Kai Dang, Peng Wang, Shijie Wang, Jun Tang, Humen Zhong, Yuanzhi Zhu, Mingkun Yang, Zhaohai Li, Jianqiang Wan, Pengfei Wang, Wei Ding, Zheren Fu, Yiheng Xu, Jiabo Ye, Xi Zhang, Tianbao Xie, Zesen Cheng, Hang Zhang, Zhibo Yang, Ha… — [Qwen2.5-VL Technical Report](https://md.research.elamur.ai/free/en/papers/bai-2025-qwen2-5-vl-technical/)
6. Xi Chen, Shiyang Zhou, Muqi Huang, Jiaxu Feng, Yun Xiong, Kun Zhou, Biao Yang, Yuhui Zhang, Huishuai Bao, Sijia Peng, et al. 2025a. A 2d semantic-aware position encoding for vision transformers. arXiv preprint arXiv:2505.09466. — [arXiv:2505.09466](https://arxiv.org/abs/2505.09466) [find](https://scholar.google.com/scholar?q=Xi+Chen%2C+Shiyang+Zhou%2C+Muqi+Huang%2C+Jiaxu+Feng%2C+Yun+Xiong%2C+Kun+Zhou%2C+Biao+Yang%2C+Yuhui+Zhang%2C+Huishuai+Bao%2C+Sijia+Peng%2C+et+al%2E+2025a%2E+A+2d+semantic%2Daware+position+encoding+for+vision+transformers%2E+arXiv)
7. Zhanpeng Chen, Mingxiao Li, Ziyang Chen, Nan Du, Xiaolong Li, and Yuexian Zou. 2025b. Advancing general multimodal capability of vision-language models with pyramid-descent visual position encoding. arXiv preprint arXiv:2501.10967. [find](https://scholar.google.com/scholar?q=Zhanpeng+Chen%2C+Mingxiao+Li%2C+Ziyang+Chen%2C+Nan+Du%2C+Xiaolong+Li%2C+and+Yuexian+Zou%2E+2025b%2E+Advancing+general+multimodal+capability+of+vision%2Dlanguage+models+with+pyramid%2Ddescent+visual+position+encoding%2E+a)
8. Kanzhi Cheng, Qiushi Sun, Yougang Chu, Fangzhi Xu, Yantao Li, Jianbing Zhang, and Zhiyong Wu. 2024. [Seeclick: Harnessing gui grounding for advanced visual gui agents](https://md.research.elamur.ai/free/en/papers/cheng-2024-seeclick-harnessing-gui-grounding-advanced/). arXiv preprint arXiv:2401.10935.
9. Wei-Lin Chiang, Zhuohan Li, Zi Lin, Ying Sheng, Zhanghao Wu, Hao Zhang, Lianmin Zheng, Siyuan Zhuang, Yonghao Zhuang, Joseph E. Gonzalez, Ion Stoica, and Eric P. Xing. 2023. Vicuna: An open-source chatbot impressing gpt-4 with 90%* chatgpt quality. [find](https://scholar.google.com/scholar?q=Wei%2DLin+Chiang%2C+Zhuohan+Li%2C+Zi+Lin%2C+Ying+Sheng%2C+Zhanghao+Wu%2C+Hao+Zhang%2C+Lianmin+Zheng%2C+Siyuan+Zhuang%2C+Yonghao+Zhuang%2C+Joseph+E%2E+Gonzalez%2C+Ion+Stoica%2C+and+Eric+P%2E+Xing%2E+2023%2E+Vicuna%3A+An+open%2Dsource+cha)
10. Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, et al. 2020. [An image is worth 16x16 words: Transformers for image recognition at scale](https://md.research.elamur.ai/free/en/papers/dosovitskiy-2020-image-worth-16x16-words-transformers/). arXiv preprint arXiv:2010.11929.
11. Hiroki Furuta, Kuang-Huei Lee, Ofir Nachum, Yutaka Matsuo, Aleksandra Faust, Shixiang Shane Gu, and Izzeddin Gur. 2023. Multimodal web navigation with instruction-finetuned foundation models. arXiv preprint arXiv:2305.11854. — [arXiv:2305.11854](https://arxiv.org/abs/2305.11854) [find](https://scholar.google.com/scholar?q=Hiroki+Furuta%2C+Kuang%2DHuei+Lee%2C+Ofir+Nachum%2C+Yutaka+Matsuo%2C+Aleksandra+Faust%2C+Shixiang+Shane+Gu%2C+and+Izzeddin+Gur%2E+2023%2E+Multimodal+web+navigation+with+instruction%2Dfinetuned+foundation+models%2E+arXiv+pr)
12. Muhan Gao, TaiMing Lu, Kuai Yu, Adam Byerly, and Daniel Khashabi. 2024. Insights into llm long-context failures: when transformers know but don’t tell. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 7611–7625. [find](https://scholar.google.com/scholar?q=Muhan+Gao%2C+TaiMing+Lu%2C+Kuai+Yu%2C+Adam+Byerly%2C+and+Daniel+Khashabi%2E+2024%2E+Insights+into+llm+long%2Dcontext+failures%3A+when+transformers+know+but+don%E2%80%99t+tell%2E+In+Findings+of+the+Association+for+Computational)
13. Junqi Ge, Ziyi Chen, Jintao Lin, Jinguo Zhu, Xihui Liu, Jifeng Dai, and Xizhou Zhu. 2024. V2pe: Improving multimodal long-context capability of vision-language models with variable visual position encoding. arXiv preprint arXiv:2412.09616. — [arXiv:2412.09616](https://arxiv.org/abs/2412.09616) [find](https://scholar.google.com/scholar?q=Junqi+Ge%2C+Ziyi+Chen%2C+Jintao+Lin%2C+Jinguo+Zhu%2C+Xihui+Liu%2C+Jifeng+Dai%2C+and+Xizhou+Zhu%2E+2024%2E+V2pe%3A+Improving+multimodal+long%2Dcontext+capability+of+vision%2Dlanguage+models+with+variable+visual+position+enc)
14. Google. 2023. An important next step on our ai journey. [find](https://scholar.google.com/scholar?q=Google%2E+2023%2E+An+important+next+step+on+our+ai+journey%2E)
15. Boyu Gou, Ruohan Wang, Boyuan Zheng, Yanan Xie, Cheng Chang, Yiheng Shu, Huan Sun, and Yu Su. 2025. [Navigating the digital world as humans do: Universal visual grounding for GUI agents](https://md.research.elamur.ai/free/en/papers/gou-2024-navigating-digital-world-humans-universal/). In The Thirteenth International Conference on Learning Representations.
16. Byeongho Heo, Song Park, Dongyoon Han, and Sangdoo Yun. 2024. Rotary position embedding for vision transformer. In European Conference on Computer Vision, pages 289–305. Springer. [find](https://scholar.google.com/scholar?q=Byeongho+Heo%2C+Song+Park%2C+Dongyoon+Han%2C+and+Sangdoo+Yun%2E+2024%2E+Rotary+position+embedding+for+vision+transformer%2E+In+European+Conference+on+Computer+Vision%2C+pages+289%E2%80%93305%2E+Springer%2E)
17. Jonathan Ho and Tim Salimans. 2022. [Classifier-free diffusion guidance](https://md.research.elamur.ai/free/en/papers/ho-2022-classifier-free-diffusion-guidance/). arXiv preprint arXiv:2207.12598.
18. Wenyi Hong, Weihan Wang, Qingsong Lv, Jiazheng Xu, Wenmeng Yu, Junhui Ji, Yan Wang, Zihan Wang, Yuxiao Dong, Ming Ding, et al. 2024. [Cogagent: A visual language model for gui agents](https://md.research.elamur.ai/free/en/papers/hong-2023-cogagent-visual-language-model/). In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 14281–14290.
19. Cheng-Yu Hsieh, Yung-Sung Chuang, Chun-Liang Li, Zifeng Wang, Long T Le, Abhishek Kumar, James Glass, Alexander Ratner, Chen-Yu Lee, Ranjay Krishna, et al. 2024. Found in the middle: Calibrating positional attention bias improves long context utilization. arXiv preprint arXiv:2406.16008. [find](https://scholar.google.com/scholar?q=Cheng%2DYu+Hsieh%2C+Yung%2DSung+Chuang%2C+Chun%2DLiang+Li%2C+Zifeng+Wang%2C+Long+T+Le%2C+Abhishek+Kumar%2C+James+Glass%2C+Alexander+Ratner%2C+Chen%2DYu+Lee%2C+Ranjay+Krishna%2C+et+al%2E+2024%2E+Found+in+the+middle%3A+Calibrating+posit)
20. Ermo Hua, Che Jiang, Xingtai Lv, Kaiyan Zhang, Youbang Sun, Yuchen Fan, Xuekai Zhu, Biqing Qi, Ning Ding, and Bowen Zhou. 2024. Fourier position embedding: Enhancing attention’s periodic extension for length generalization. arXiv preprint arXiv:2412.17739. [find](https://scholar.google.com/scholar?q=Ermo+Hua%2C+Che+Jiang%2C+Xingtai+Lv%2C+Kaiyan+Zhang%2C+Youbang+Sun%2C+Yuchen+Fan%2C+Xuekai+Zhu%2C+Biqing+Qi%2C+Ning+Ding%2C+and+Bowen+Zhou%2E+2024%2E+Fourier+position+embedding%3A+Enhancing+attention%E2%80%99s+periodic+extension+for)
21. Wonjun Lee, Bumsub Ham, and Suhyun Kim. 2025. Maximizing the position embedding for vision transformers with global average pooling. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 39, pages 18154–18162. [find](https://scholar.google.com/scholar?q=Wonjun+Lee%2C+Bumsub+Ham%2C+and+Suhyun+Kim%2E+2025%2E+Maximizing+the+position+embedding+for+vision+transformers+with+global+average+pooling%2E+In+Proceedings+of+the+AAAI+Conference+on+Artificial+Intelligence%2C+v)
22. Chunyuan Li, Cliff Wong, Sheng Zhang, Naoto Usuyama, Haotian Liu, Jianwei Yang, Tristan Naumann, Hoifung Poon, and Jianfeng Gao. 2023. [Llava-med: Training a large language-and-vision assistant for biomedicine in one day](https://md.research.elamur.ai/free/en/papers/li-2023-llava-med-training-large-language/). Advances in Neural Information Processing Systems, 36:28541–28564.
23. Kaixin Li, Ziyang Meng, Hongzhan Lin, Ziyang Luo, Yuchen Tian, Jing Ma, Zhiyong Huang, and Tat-Seng Chua. 2025. Screenspot-pro: Gui grounding for professional high-resolution computer use. arXiv preprint arXiv:2504.07981. — [arXiv:2504.07981](https://arxiv.org/abs/2504.07981) [find](https://scholar.google.com/scholar?q=Kaixin+Li%2C+Ziyang+Meng%2C+Hongzhan+Lin%2C+Ziyang+Luo%2C+Yuchen+Tian%2C+Jing+Ma%2C+Zhiyong+Huang%2C+and+Tat%2DSeng+Chua%2E+2025%2E+Screenspot%2Dpro%3A+Gui+grounding+for+professional+high%2Dresolution+computer+use%2E+arXiv+prepr)
24. Wenhui Liao, Jiapeng Wang, Hongliang Li, Chengyu Wang, Jun Huang, and Lianwen Jin. 2025. Do-clayllm: An efficient multi-modal extension of large language models for text-rich document understanding. In Proceedings of the Computer Vision and Pattern Recognition Conference, pages 4038–4049. [find](https://scholar.google.com/scholar?q=Wenhui+Liao%2C+Jiapeng+Wang%2C+Hongliang+Li%2C+Chengyu+Wang%2C+Jun+Huang%2C+and+Lianwen+Jin%2E+2025%2E+Do%2Dclayllm%3A+An+efficient+multi%2Dmodal+extension+of+large+language+models+for+text%2Drich+document+understanding%2E+I)
25. Kevin Qinghong Lin, Linjie Li, Difei Gao, Zhengyuan Yang, Shiwei Wu, Zechen Bai, Weixian Lei, Lijuan Wang, and Mike Zheng Shou. 2024. Showui: One vision-language-action model for gui visual agent. arXiv preprint arXiv:2411.17465. — [arXiv:2411.17465](https://arxiv.org/abs/2411.17465) [find](https://scholar.google.com/scholar?q=Kevin+Qinghong+Lin%2C+Linjie+Li%2C+Difei+Gao%2C+Zhengyuan+Yang%2C+Shiwei+Wu%2C+Zechen+Bai%2C+Weixian+Lei%2C+Lijuan+Wang%2C+and+Mike+Zheng+Shou%2E+2024%2E+Showui%3A+One+vision%2Dlanguage%2Daction+model+for+gui+visual+agent%2E+arX)
26. Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee. 2023a. [Visual instruction tuning](https://md.research.elamur.ai/free/en/papers/liu-2023-visual-instruction-tuning/). Advances in neural information processing systems, 36:34892– 34916.
27. Shilong Liu, Hao Cheng, Haotian Liu, Hao Zhang, Feng Li, Tianhe Ren, Xueyan Zou, Jianwei Yang, Hang Su, Jun Zhu, Lei Zhang, Jianfeng Gao, and Chunyuan Li. 2023b. Llava-plus: Learning to use tools for creating multimodal agents. [find](https://scholar.google.com/scholar?q=Shilong+Liu%2C+Hao+Cheng%2C+Haotian+Liu%2C+Hao+Zhang%2C+Feng+Li%2C+Tianhe+Ren%2C+Xueyan+Zou%2C+Jianwei+Yang%2C+Hang+Su%2C+Jun+Zhu%2C+Lei+Zhang%2C+Jianfeng+Gao%2C+and+Chunyuan+Li%2E+2023b%2E+Llava%2Dplus%3A+Learning+to+use+tools+for+)
28. Yadong Lu, Jianwei Yang, Yelong Shen, and Ahmed Awadallah. 2024. Omniparser for pure vision based gui agent. arXiv preprint arXiv:2408.00203. — [arXiv:2408.00203](https://arxiv.org/abs/2408.00203) [find](https://scholar.google.com/scholar?q=Yadong+Lu%2C+Jianwei+Yang%2C+Yelong+Shen%2C+and+Ahmed+Awadallah%2E+2024%2E+Omniparser+for+pure+vision+based+gui+agent%2E+arXiv+preprint+arXiv%3A2408%2E00203%2E+%E2%80%94+%5BarXiv%3A2408%2E00203%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2408%2E00203%29)
29. Xinyin Ma, Gongfan Fang, and Xinchao Wang. 2023. [Llm-pruner: On the structural pruning of large language models](https://md.research.elamur.ai/free/en/papers/ma-2023-llm-pruner-structural-pruning-large/). Advances in neural information processing systems, 36:21702–21720.
30. MosaicML. 2023. Introducing mpt-7b: A new standard for open-source, commercially usable llms. Accessed: 2023-05-05. [find](https://scholar.google.com/scholar?q=MosaicML%2E+2023%2E+Introducing+mpt%2D7b%3A+A+new+standard+for+open%2Dsource%2C+commercially+usable+llms%2E+Accessed%3A+2023%2D05%2D05%2E)
31. OpenAI. 2022. OpenAI: Introducing ChatGPT. [find](https://scholar.google.com/scholar?q=OpenAI%2E+2022%2E+OpenAI%3A+Introducing+ChatGPT%2E)
32. Judea Pearl. 2018. Causal and counterfactual inference. The Handbook of Rationality, pages 1–41. [find](https://scholar.google.com/scholar?q=Judea+Pearl%2E+2018%2E+Causal+and+counterfactual+inference%2E+The+Handbook+of+Rationality%2C+pages+1%E2%80%9341%2E)
33. Judea Pearl, Madelyn Glymour, and Nicholas P Jewell. 2016. Causal inference in statistics: A primer. John Wiley & Sons. [find](https://scholar.google.com/scholar?q=Judea+Pearl%2C+Madelyn+Glymour%2C+and+Nicholas+P+Jewell%2E+2016%2E+Causal+inference+in+statistics%3A+A+primer%2E+John+Wiley+%26+Sons%2E)
34. Ziqi Wang, Hanlin Zhang, Xiner Li, Kuan-Hao Huang, Chi Han, Shuiwang Ji, Sham M. Kakade, Hao Peng, and Heng Ji. 2025. Eliminating position bias of language models: A mechanistic approach. [find](https://scholar.google.com/scholar?q=Ziqi+Wang%2C+Hanlin+Zhang%2C+Xiner+Li%2C+Kuan%2DHao+Huang%2C+Chi+Han%2C+Shuiwang+Ji%2C+Sham+M%2E+Kakade%2C+Hao+Peng%2C+and+Heng+Ji%2E+2025%2E+Eliminating+position+bias+of+language+models%3A+A+mechanistic+approach%2E)
35. Hang Wu, Hongkai Chen, Yujun Cai, Chang Liu, Qingwen Ye, Ming-Hsuan Yang, and Yiwei Wang. 2025. [Dimo-gui: Advancing test-time scaling in gui grounding via modality-aware visual reasoning](https://md.research.elamur.ai/free/en/papers/wu-2025-dimo-gui-advancing-test-time/). arXiv preprint arXiv:2507.00008.
36. Jianing Qi, Jiawei Liu, Hao Tang, and Zhigang Zhu. 2025. Beyond semantics: Rediscovering spatial awareness in vision-language models. arXiv preprint arXiv:2503.17349. — [arXiv:2503.17349](https://arxiv.org/abs/2503.17349) [find](https://scholar.google.com/scholar?q=Jianing+Qi%2C+Jiawei+Liu%2C+Hao+Tang%2C+and+Zhigang+Zhu%2E+2025%2E+Beyond+semantics%3A+Rediscovering+spatial+awareness+in+vision%2Dlanguage+models%2E+arXiv+preprint+arXiv%3A2503%2E17349%2E+%E2%80%94+%5BarXiv%3A2503%2E17349%5D%28https%3A%2F%2Farxi)
37. Zhiyong Wu, Zhenyu Wu, Fangzhi Xu, Yian Wang, Qiushi Sun, Chengyou Jia, Kanzhi Cheng, Zichen Ding, Liheng Chen, Paul Pu Liang, et al. 2024. [Os-atlas: A foundation action model for generalist gui agents](https://md.research.elamur.ai/free/en/papers/wu-2024-os-atlas-foundation-action-model/). arXiv preprint arXiv:2410.23218.
38. Jianlin Su, Murtadha Ahmed, Yu Lu, Shengfeng Pan, Wen Bo, and Yunfeng Liu. 2024. [Roformer: Enhanced transformer with rotary position embedding](https://md.research.elamur.ai/free/en/papers/su-2021-roformer-enhanced-transformer-rotary-position/). Neurocomputing, 568:127063.
39. Jin Xu, Zhifang Guo, Jinzheng He, Hangrui Hu, Ting He, Shuai Bai, Keqin Chen, Jialin Wang, Yang Fan, Kai Dang, Bin Zhang, Xiong Wang, Yunfei Chu, and Junyang Lin. 2025. [Qwen2.5-omni technical report](https://md.research.elamur.ai/free/en/papers/xu-2025-qwen2-5-omni-technical/).
40. Kaihua Tang, Yulei Niu, Jianqiang Huang, Jiaxin Shi, and Hanwang Zhang. 2020. Unbiased scene graph generation from biased training. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 3716–3725. [find](https://scholar.google.com/scholar?q=Kaihua+Tang%2C+Yulei+Niu%2C+Jianqiang+Huang%2C+Jiaxin+Shi%2C+and+Hanwang+Zhang%2E+2020%2E+Unbiased+scene+graph+generation+from+biased+training%2E+In+Proceedings+of+the+IEEE%2FCVF+Conference+on+Computer+Vision+and+Pat)
41. Yuhao Yang, Yue Wang, Dongxu Li, Ziyang Luo, Bei Chen, Chao Huang, and Junnan Li. 2024. Aria-ui: Visual grounding for gui instructions. arXiv preprint arXiv:2412.16256. [find](https://scholar.google.com/scholar?q=Yuhao+Yang%2C+Yue+Wang%2C+Dongxu+Li%2C+Ziyang+Luo%2C+Bei+Chen%2C+Chao+Huang%2C+and+Junnan+Li%2E+2024%2E+Aria%2Dui%3A+Visual+grounding+for+gui+instructions%2E+arXiv+preprint+arXiv%3A2412%2E16256%2E)
42. Xingjian Tao, Yiwei Wang, Yujun Cai, Zhicheng Yang, and Jing Tang. 2025. Understanding gui agent localization biases through logit sharpness. arXiv preprint arXiv:2506.15425. [find](https://scholar.google.com/scholar?q=Xingjian+Tao%2C+Yiwei+Wang%2C+Yujun+Cai%2C+Zhicheng+Yang%2C+and+Jing+Tang%2E+2025%2E+Understanding+gui+agent+localization+biases+through+logit+sharpness%2E+arXiv+preprint+arXiv%3A2506%2E15425%2E)
43. Gemini Team, Rohan Anil, Sebastian Borgeaud, Jean-Baptiste Alayrac, Jiahui Yu, Radu Soricut, Johan Schalkwyk, Andrew M Dai, Anja Hauth, Katie Millican, et al. 2023. [Gemini: a family of highly capable multimodal models](https://md.research.elamur.ai/free/en/papers/gemini-team-2023-gemini-family-highly-capable-multimodal/). arXiv preprint arXiv:2312.11805.
44. Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, et al. 2023a. [Llama: Open and efficient foundation language models](https://md.research.elamur.ai/free/en/papers/touvron-2023-llama-open-efficient-foundation-language/). arXiv preprint arXiv:2302.13971.
45. Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yasmine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava, Shruti Bhosale, et al. 2023b. [Llama 2: Open foundation and fine-tuned chat models](https://md.research.elamur.ai/free/en/papers/touvron-2023-llama-2-open-foundation-fine/). arXiv preprint arXiv:2307.09288.
46. Zhengyuan Yang, Linjie Li, Kevin Lin, Jianfeng Wang, Chung-Ching Lin, Zicheng Liu, and Lijuan Wang. 2023. [The dawn of lmms: Preliminary explorations with gpt-4v (ision)](https://md.research.elamur.ai/free/en/papers/yang-2023-dawn-lmms-preliminary-explorations-gpt/). arXiv preprint arXiv:2309.17421, 9(1):1.
47. Howard Yen, Tianyu Gao, and Danqi Chen. 2024. [Long-context language modeling with parallel context encoding](https://md.research.elamur.ai/free/en/papers/yen-2024-long-context-language-modeling-parallel/).
48. Hao Yu, Tangyu Jiang, Shuning Jia, Shannan Yan, Shunning Liu, Haolong Qian, Guanghao Li, Shuting Dong, and Chun Yuan. 2025. Comrope: Scalable and robust rotary position embedding parameterized by trainable commuting angle matrices. In Proceedings of the Computer Vision and Pattern Recognition Confer… [find](https://scholar.google.com/scholar?q=Hao+Yu%2C+Tangyu+Jiang%2C+Shuning+Jia%2C+Shannan+Yan%2C+Shunning+Liu%2C+Haolong+Qian%2C+Guanghao+Li%2C+Shuting+Dong%2C+and+Chun+Yuan%2E+2025%2E+Comrope%3A+Scalable+and+robust+rotary+position+embedding+parameterized+by+trai)
49. Yi-Fan Zhang, Qingsong Wen, Chaoyou Fu, Xue Wang, Zhang Zhang, Liang Wang, and Rong Jin. 2024. Beyond llava-hd: Diving into high-resolution large multimodal models. arXiv preprint arXiv:2406.08487. — [arXiv:2406.08487](https://arxiv.org/abs/2406.08487) [find](https://scholar.google.com/scholar?q=Yi%2DFan+Zhang%2C+Qingsong+Wen%2C+Chaoyou+Fu%2C+Xue+Wang%2C+Zhang+Zhang%2C+Liang+Wang%2C+and+Rong+Jin%2E+2024%2E+Beyond+llava%2Dhd%3A+Diving+into+high%2Dresolution+large+multimodal+models%2E+arXiv+preprint+arXiv%3A2406%2E08487%2E+%E2%80%94+)
50. 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](https://md.research.elamur.ai/free/en/papers/wang-2024-qwen2-vl-enhancing-vision/). arXiv preprint arXiv:2409.12191.
51. Yiwei Wang, Muhao Chen, Wenxuan Zhou, Yujun Cai, Yuxuan Liang, Dayiheng Liu, Baosong Yang, Juncheng Liu, and Bryan Hooi. 2022. Should we rely on entity mentions for relation extraction? debiasing relation extraction with counterfactual analysis. arXiv preprint arXiv:2205.03784. [find](https://scholar.google.com/scholar?q=Yiwei+Wang%2C+Muhao+Chen%2C+Wenxuan+Zhou%2C+Yujun+Cai%2C+Yuxuan+Liang%2C+Dayiheng+Liu%2C+Baosong+Yang%2C+Juncheng+Liu%2C+and+Bryan+Hooi%2E+2022%2E+Should+we+rely+on+entity+mentions+for+relation+extraction%3F+debiasing+rela)

*Updated 2 November 2026*