pith. sign in

arxiv: 2606.24551 · v1 · pith:OMKEXG2Rnew · submitted 2026-06-22 · 💻 cs.AI

GUI vs. CLI: Execution Bottlenecks in Screen-Only and Skill-Mediated Computer-Use Agents

Xiao Zhou , Siyue Zhang , Yilun Zhao , Jinbiao Wei , Tingyu Song , Arman Cohan , Chen Zhao This is my paper

Pith reviewed 2026-06-26 08:22 UTC · model grok-4.3

classification 💻 cs.AI
keywords computer-use agentsGUI agentsCLI agentsexecution bottlenecksmatched benchmarkskill augmentationdesktop tasksagent evaluation
0
0 comments X

The pith

A matched benchmark of 440 tasks shows GUI agents at 59.1% success outperforming original-skill CLI at 48.2%, with verifier-guided CLI reaching 69.3%.

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

The paper introduces a controlled benchmark of 440 desktop tasks across 18 applications where screen-only GUI agents and skill-mediated CLI agents receive identical goals, initial states, and final-state verifiers but are limited to actions native to their interface. In this setting the strongest GUI agent achieves a 59.1% full pass rate while the strongest original-skill CLI agent reaches only 48.2%. Augmenting CLI skills with verifier guidance lifts CLI performance to 69.3%. The results indicate that GUI agents are constrained by reliable grounded interaction across long workflows and CLI agents are constrained by the coverage and scalability of their skill sets. A reader would care because the comparison isolates modality effects and points to distinct improvement targets for each approach.

Core claim

In a matched execution-layer benchmark of 440 desktop tasks across 18 applications and 12 workflow categories, where screen-only GUI agents and skill-mediated CLI agents receive identical goals, states, and final-state verifiers while being restricted to modality-native actions, the strongest GUI agent reaches a 59.1% full pass rate, outperforming the strongest original-skill CLI agent at 48.2%; however, verifier-guided skill augmentation raises CLI success to 69.3%, showing that much of the CLI deficit comes from incomplete skill coverage rather than model capability alone. These results suggest that GUI and CLI expose different execution bottlenecks: GUI agents are limited by reliable grou

What carries the argument

The matched execution-layer benchmark of 440 desktop tasks across 18 applications and 12 workflow categories that equalizes goals, states, verifiers, and restricts agents to modality-native actions.

If this is right

  • GUI agents require advances specifically in reliable grounded interaction for long-horizon workflows.
  • CLI agents improve substantially when their skill interfaces are expanded via verifier guidance.
  • Performance differences between the two modalities can be isolated when tasks and verifiers are matched.
  • Optimization efforts for computer-use agents should target modality-specific constraints rather than model scale alone.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Hybrid agents that can switch between GUI and CLI actions within the same workflow may reach higher success rates than either modality alone.
  • The benchmark design could be reused to compare additional modalities such as voice or API-only interfaces on the same task set.
  • Dynamic skill generation during execution might address CLI coverage limits more scalably than static skill libraries.
  • Long-horizon grounding failures in GUI may be measurable by counting the step at which agents lose accurate screen localization.

Load-bearing premise

The benchmark of 440 tasks successfully eliminates confounding differences in tasks, initial states, verifiers, and permitted actions so that performance gaps can be attributed solely to interaction modality.

What would settle it

Re-running the 440-task evaluation with the same agents but allowing cross-modality actions or altered verifiers and observing that the performance ordering disappears or reverses would falsify the claim that the bottlenecks are caused by modality.

Figures

Figures reproduced from arXiv: 2606.24551 by Arman Cohan, Chen Zhao, Jinbiao Wei, Siyue Zhang, Tingyu Song, Xiao Zhou, Yilun Zhao.

Figure 1
Figure 1. Figure 1: Overview. A matched execution-layer benchmark of 440 desktop tasks compares GUI and skill-mediated CLI agents under identical goals, states, and verifiers. Results show that the observed CLI gap is strongly affected by skill coverage, while GUI embeds procedural execution within the interface. are controlled while the native action interface is systematically varied across settings. Across this benchmark, … view at source ↗
Figure 2
Figure 2. Figure 2: Benchmark composition across workflows and applications in the 440-task benchmark, spanning 18 applications across 12 workflow categories. transformation required: some tasks rely on vi￾sual layout and interface navigation, while others expose structured artifacts or application-level op￾erations. This composition supports both aggregate comparison and per-workflow analysis of where GUI and CLI execution s… view at source ↗
Figure 3
Figure 3. Figure 3: Benchmark construction pipeline. We select applications with CLI-Anything skill support, rewrite GUI [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Distribution of primary error types for failed [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: CLI agent system prompt. <GUI_SCREEN_ONLY_POLICY> Hard requirements: * All task-result changes must be made through the target application’s visible GUI. * Do not directly rewrite task files on disk or mutate their underlying data using Python, shell commands, scripts, automation APIs, databases, archives, config files, or external utilities. * Editing content through the target application’s own GUI is al… view at source ↗
Figure 6
Figure 6. Figure 6: GUI agent screen-only policy [PITH_FULL_IMAGE:figures/full_fig_p014_6.png] view at source ↗
read the original abstract

Computer-use agents can execute software tasks through either graphical interfaces or programmatic command interfaces, but existing evaluations confound interaction modality with differences in tasks, initial states, verifiers, and permitted actions. We introduce a matched execution-layer benchmark of 440 desktop tasks across 18 applications and 12 workflow categories, where screen-only GUI agents and skill-mediated CLI agents receive identical goals, states, and final-state verifiers while being restricted to modality-native actions. In this controlled setting, the strongest GUI agent reaches a 59.1% full pass rate, outperforming the strongest original-skill CLI agent at 48.2%; however, verifier-guided skill augmentation raises CLI success to 69.3%, showing that much of the CLI deficit comes from incomplete skill coverage rather than model capability alone. These results suggest that GUI and CLI expose different execution bottlenecks: GUI agents are limited by reliable grounded interaction over long-horizon workflows, whereas CLI agents are limited by the coverage and scalability of their skill interfaces.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper introduces a matched execution-layer benchmark of 440 desktop tasks across 18 applications and 12 workflow categories. Screen-only GUI agents and skill-mediated CLI agents receive identical goals, states, and final-state verifiers but are restricted to modality-native actions. The strongest GUI agent achieves a 59.1% full pass rate, outperforming the strongest original-skill CLI agent at 48.2%; verifier-guided skill augmentation raises CLI success to 69.3%. The authors conclude that GUI agents are limited by reliable grounded interaction over long-horizon workflows, whereas CLI agents are limited by the coverage and scalability of their skill interfaces.

Significance. If the benchmark successfully isolates interaction modality, the work supplies concrete empirical evidence distinguishing execution bottlenecks between GUI and CLI computer-use agents. The scale of the matched benchmark (440 tasks) and the augmentation experiment (showing CLI improvement to 69.3%) are strengths that could guide future agent design toward better grounding mechanisms for GUI and expanded skill libraries for CLI.

major comments (2)
  1. [Abstract] Abstract: The central attribution of the performance gap (59.1% GUI vs. 48.2% original-skill CLI) to interaction modality rather than agent differences requires that the 'strongest' agents per modality are comparable in base model capability, training, and prompting. The abstract supplies no selection criteria or verification that underlying models are matched, which is load-bearing for the claim that the gap reflects modality-specific bottlenecks.
  2. [Benchmark description] Benchmark description (as summarized in abstract): The matched execution-layer benchmark is stated to control for tasks, initial states, verifiers, and permitted actions, yet the description does not address control over agent implementations themselves. Without such control, differences in model strength or prompting regimes could confound the modality comparison and undermine the conclusion that GUI and CLI expose distinct bottlenecks.
minor comments (1)
  1. The abstract references 'verifier-guided skill augmentation' and 'original-skill CLI' without defining these terms or the augmentation procedure; these should be expanded with concrete examples in the methods section.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback emphasizing the need to clearly establish that performance differences arise from interaction modality rather than uncontrolled agent variations. We address the major comments point by point below.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The central attribution of the performance gap (59.1% GUI vs. 48.2% original-skill CLI) to interaction modality rather than agent differences requires that the 'strongest' agents per modality are comparable in base model capability, training, and prompting. The abstract supplies no selection criteria or verification that underlying models are matched, which is load-bearing for the claim that the gap reflects modality-specific bottlenecks.

    Authors: We agree the abstract is concise and omits explicit selection criteria. The full manuscript (Section 4.1) specifies that both the GUI and CLI agents are instantiated from the same base model (GPT-4o) with modality-specific action interfaces and standardized prompting templates; the 'strongest' designation refers to the best publicly documented configuration for each modality under these constraints. To address the concern directly in the abstract, we will insert a short clause noting the use of matched base models and prompting regimes. revision: yes

  2. Referee: [Benchmark description] Benchmark description (as summarized in abstract): The matched execution-layer benchmark is stated to control for tasks, initial states, verifiers, and permitted actions, yet the description does not address control over agent implementations themselves. Without such control, differences in model strength or prompting regimes could confound the modality comparison and undermine the conclusion that GUI and CLI expose distinct bottlenecks.

    Authors: The benchmark description intentionally focuses on environmental controls (identical tasks, states, and verifiers) to isolate modality effects while allowing each modality to use its strongest available agent implementation; agent details and base-model matching are provided in the experimental setup to avoid confounding. We will expand the benchmark description paragraph to cross-reference the agent-matching criteria already present in Section 4, making the controls explicit without altering the core design. revision: partial

Circularity Check

0 steps flagged

No circularity: purely empirical benchmark comparison

full rationale

The paper reports results from a matched execution-layer benchmark of 440 tasks with no equations, fitted parameters, derivations, or self-citation chains. The central claim (distinct bottlenecks for GUI vs. CLI) is an interpretation of measured pass rates (59.1% GUI, 48.2% original CLI, 69.3% augmented CLI) under controlled conditions. No step reduces a prediction or uniqueness claim to its own inputs by construction. This is a standard empirical study whose validity rests on benchmark design rather than any definitional or self-referential loop.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

The central claim rests on the unverified assertion that the benchmark construction isolates modality effects; no free parameters, axioms, or invented entities are introduced in the abstract.

pith-pipeline@v0.9.1-grok · 5724 in / 1173 out tokens · 22880 ms · 2026-06-26T08:22:05.191568+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

300 extracted references · 146 canonical work pages · 1 internal anchor

  1. [1]

    arXiv preprint arXiv:2108.07732 , year =

    Program Synthesis with Large Language Models , author =. arXiv preprint arXiv:2108.07732 , year =

    Pith/arXiv arXiv

  2. [2]

    arXiv preprint arXiv:2107.03374 , year =

    Evaluating Large Language Models Trained on Code , author =. arXiv preprint arXiv:2107.03374 , year =

    Pith/arXiv arXiv

  3. [3]

    The Twelfth International Conference on Learning Representations , year=

    WebArena: A Realistic Web Environment for Building Autonomous Agents , author=. The Twelfth International Conference on Learning Representations , year=

  4. [4]

    and Del Verme, Manuel and Marty, Tom and Boisvert, L

    Drouin, Alexandre and Gasse, Maxime and Caccia, Massimo and Laradji, Issam H. and Del Verme, Manuel and Marty, Tom and Boisvert, L. arXiv preprint arXiv:2403.07718 , year =

    Pith/arXiv arXiv

  5. [5]

    2023 , url =

    Deng, Xiang and Gu, Yu and Zheng, Boyuan and Chen, Shijie and Stevens, Sam and Wang, Boshi and Sun, Huan and Su, Yu , journal =. 2023 , url =

    2023

  6. [6]

    2024 , url =

    Zheng, Boyuan and Gou, Boyu and Kil, Jihyung and Sun, Huan and Su, Yu , journal =. 2024 , url =

    2024

  7. [7]

    2024 , url =

    Rawles, Christopher and Clinckemaillie, Sarah and Chang, Yifan and Waltz, Jonathan and Lau, Gabrielle and Fair, Marybeth and Li, Alice and Bishop, William and Li, Wei and Campbell-Ajala, Folawiyo and Toyama, Daniel and Berry, Robert and Tyamagundlu, Divya and Lillicrap, Timothy and Riva, Oriana , journal =. 2024 , url =

    2024

  8. [8]

    2024 , url=

    Tianbao Xie and Danyang Zhang and Jixuan Chen and Xiaochuan Li and Siheng Zhao and Ruisheng Cao and Toh Jing Hua and Zhoujun Cheng and Dongchan Shin and Fangyu Lei and Yitao Liu and Yiheng Xu and Shuyan Zhou and Silvio Savarese and Caiming Xiong and Victor Zhong and Tao Yu , booktitle=. 2024 , url=

    2024

  9. [9]

    Xu and Shuyan Zhou and Graham Neubig , year=

    Yueqi Song and Frank F. Xu and Shuyan Zhou and Graham Neubig , year=. Beyond Browsing:

  10. [10]

    2023 , url =

    Li, Minghao and Zhao, Yingxiu and Yu, Bowen and Song, Feifan and Li, Hangyu and Yu, Haiyang and Li, Zhoujun and Huang, Fei and Li, Yongbin , booktitle =. 2023 , url =

    2023

  11. [11]

    2024 , url =

    Trivedi, Harsh and Khot, Tushar and Hartmann, Mareike and Manku, Ruskin and Dong, Vinty and Li, Edward and Gupta, Shashank and Sabharwal, Ashish and Balasubramanian, Niranjan , journal =. 2024 , url =

    2024

  12. [12]

    arXiv preprint arXiv:2604.00073 , year =

    Terminal Agents Suffice for Enterprise Automation , author =. arXiv preprint arXiv:2604.00073 , year =

    Pith/arXiv arXiv

  13. [13]

    2025 , url=

    GUI Agents: A Survey , author=. 2025 , url=

    2025

  14. [14]

    arXiv preprint arXiv:2601.11868 , year =

    Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces , author =. arXiv preprint arXiv:2601.11868 , year =

    Pith/arXiv arXiv

  15. [15]

    2026 , eprint=

    SkillsBench: Benchmarking How Well Agent Skills Work Across Diverse Tasks , author=. 2026 , eprint=

    2026

  16. [16]

    2026 , url =

    Han, Tingxu and Zhang, Yi and Song, Wei and Fang, Chunrong and Chen, Zhenyu and Sun, Youcheng and Hu, Lijie , journal =. 2026 , url =

    2026

  17. [17]

    and Yao, Ziyu , journal =

    Aghzal, Mohamed and Stein, Gregory J. and Yao, Ziyu , journal =. Why Do. 2026 , url =

    2026

  18. [18]

    2023 , url =

    Liu, Xiao and Yu, Hao and Zhang, Hanchen and Xu, Yifan and Lei, Xuanyu and Lai, Hanyu and Gu, Yu and Ding, Hangliang and Men, Kaiwen and Yang, Kejuan and Zhang, Shudan and Deng, Xiang and Zeng, Aohan and Du, Zhengxiao and Zhang, Chenhui and Shen, Sheng and Zhang, Tianjun and Su, Yu and Sun, Huan and Huang, Minlie and Dong, Yuxiao and Tang, Jie , journal =...

    2023

  19. [19]

    2026 , howpublished =

    Introducing. 2026 , howpublished =

    2026

  20. [20]

    2026 , howpublished =

    2026

  21. [21]

    2026 , eprint=

    EvoCUA: Evolving Computer Use Agents via Learning from Scalable Synthetic Experience , author=. 2026 , eprint=

    2026

  22. [22]

    2026 , eprint=

    OpenComputer: Verifiable Software Worlds for Computer-Use Agents , author=. 2026 , eprint=

    2026

  23. [23]

    2025 , eprint=

    UI-TARS: Pioneering Automated GUI Interaction with Native Agents , author=. 2025 , eprint=

    2025

  24. [24]

    2026 , eprint=

    ANCHOR: Branch-Point Data Generation for GUI Agents , author=. 2026 , eprint=

    2026

  25. [25]

    2026 , eprint=

    Android Coach: Improve Online Agentic Training Efficiency with Single State Multiple Actions , author=. 2026 , eprint=

    2026

  26. [26]

    2025 , eprint=

    OpenCUA: Open Foundations for Computer-Use Agents , author=. 2025 , eprint=

    2025

  27. [27]

    Proceedings of the 21st Workshop of Young Researchers' Roundtable on Spoken Dialogue Systems. 2025

    2025

  28. [28]

    Research on LLM s-Empowered Conversational AI for Sustainable Behaviour Change

    Chen, Ben. Research on LLM s-Empowered Conversational AI for Sustainable Behaviour Change. Proceedings of the 21st Workshop of Young Researchers' Roundtable on Spoken Dialogue Systems. 2025

    2025

  29. [29]

    Deep Reinforcement Learning of LLM s​ using RLHF

    Levandovsky, Enoch. Deep Reinforcement Learning of LLM s​ using RLHF. Proceedings of the 21st Workshop of Young Researchers' Roundtable on Spoken Dialogue Systems. 2025

    2025

  30. [30]

    Conversational Collaborative Robots

    Kranti, Chalamalasetti. Conversational Collaborative Robots. Proceedings of the 21st Workshop of Young Researchers' Roundtable on Spoken Dialogue Systems. 2025

    2025

  31. [31]

    Dialogue System using Large Language Model-based Dynamic Slot Generation

    Hashimoto, Ekai. Dialogue System using Large Language Model-based Dynamic Slot Generation. Proceedings of the 21st Workshop of Young Researchers' Roundtable on Spoken Dialogue Systems. 2025

    2025

  32. [32]

    Towards Adaptive Human-Agent Collaboration in Real-Time Environments

    Nakae, Kaito. Towards Adaptive Human-Agent Collaboration in Real-Time Environments. Proceedings of the 21st Workshop of Young Researchers' Roundtable on Spoken Dialogue Systems. 2025

    2025

  33. [33]

    Towards Human-Like Dialogue Systems: Integrating Multimodal Emotion Recognition and Non-Verbal Cue Generation

    Jiang, Jingjing. Towards Human-Like Dialogue Systems: Integrating Multimodal Emotion Recognition and Non-Verbal Cue Generation. Proceedings of the 21st Workshop of Young Researchers' Roundtable on Spoken Dialogue Systems. 2025

    2025

  34. [34]

    Controlling Dialogue Systems with Graph-Based Structures

    Hilgendorf, Laetitia Mina. Controlling Dialogue Systems with Graph-Based Structures. Proceedings of the 21st Workshop of Young Researchers' Roundtable on Spoken Dialogue Systems. 2025

    2025

  35. [35]

    Multimodal Agentic Dialogue Systems for Situated Human-Robot Interaction

    Sucal, Virgile. Multimodal Agentic Dialogue Systems for Situated Human-Robot Interaction. Proceedings of the 21st Workshop of Young Researchers' Roundtable on Spoken Dialogue Systems. 2025

    2025

  36. [36]

    Knowledge Graphs and Representational Models for Dialogue Systems

    Walker, Nicholas Thomas. Knowledge Graphs and Representational Models for Dialogue Systems. Proceedings of the 21st Workshop of Young Researchers' Roundtable on Spoken Dialogue Systems. 2025

    2025

  37. [37]

    Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.0

    work page doi:10.18653/v1/2025.xllm-1.0 2025

  38. [38]

    Fine-Tuning Large Language Models for Relation Extraction within a Retrieval-Augmented Generation Framework

    Efeoglu, Sefika and Paschke, Adrian. Fine-Tuning Large Language Models for Relation Extraction within a Retrieval-Augmented Generation Framework. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.1

    work page doi:10.18653/v1/2025.xllm-1.1 2025

  39. [39]

    Benchmarking Table Extraction: Multimodal LLM s vs Traditional OCR

    Nunes, Guilherme and Rolla, Vitor and Pereira, Duarte and Alves, Vasco and Carreiro, Andre and Baptista, M \'a rcia. Benchmarking Table Extraction: Multimodal LLM s vs Traditional OCR. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.2

    work page doi:10.18653/v1/2025.xllm-1.2 2025

  40. [40]

    Injecting Structured Knowledge into LLM s via Graph Neural Networks

    Li, Zichao and Ke, Zong and Zhao, Puning. Injecting Structured Knowledge into LLM s via Graph Neural Networks. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.3

    work page doi:10.18653/v1/2025.xllm-1.3 2025

  41. [41]

    Regular-pattern-sensitive CRF s for Distant Label Interactions

    Papay, Sean and Klinger, Roman and Pad \'o , Sebastian. Regular-pattern-sensitive CRF s for Distant Label Interactions. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.4

    work page doi:10.18653/v1/2025.xllm-1.4 2025

  42. [42]

    From Syntax to Semantics: Evaluating the Impact of Linguistic Structures on LLM -Based Information Extraction

    Swarup, Anushka and Bhandarkar, Avanti and Wilson, Ronald and Pan, Tianyu and Woodard, Damon. From Syntax to Semantics: Evaluating the Impact of Linguistic Structures on LLM -Based Information Extraction. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.5

    work page doi:10.18653/v1/2025.xllm-1.5 2025

  43. [43]

    Detecting Referring Expressions in Visually Grounded Dialogue with Autoregressive Language Models

    Willemsen, Bram and Skantze, Gabriel. Detecting Referring Expressions in Visually Grounded Dialogue with Autoregressive Language Models. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.6

    work page doi:10.18653/v1/2025.xllm-1.6 2025

  44. [44]

    Exploring Multilingual Probing in Large Language Models: A Cross-Language Analysis

    Li, Daoyang and Zhao, Haiyan and Zeng, Qingcheng and Du, Mengnan. Exploring Multilingual Probing in Large Language Models: A Cross-Language Analysis. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.7

    work page doi:10.18653/v1/2025.xllm-1.7 2025

  45. [45]

    Self-Contrastive Loop of Thought Method for Text-to- SQL Based on Large Language Model

    Kang, Fengrui and Tan, Mingxi and Huang, Xianying and Yang, Shiju. Self-Contrastive Loop of Thought Method for Text-to- SQL Based on Large Language Model. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.8

    work page doi:10.18653/v1/2025.xllm-1.8 2025

  46. [46]

    Combining Automated and Manual Data for Effective Downstream Fine-Tuning of Transformers for Low-Resource Language Applications

    Isaeva, Ulyana and Astafurov, Danil and Martynov, Nikita. Combining Automated and Manual Data for Effective Downstream Fine-Tuning of Transformers for Low-Resource Language Applications. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.9

    work page doi:10.18653/v1/2025.xllm-1.9 2025

  47. [47]

    Seamlessly Integrating Tree-Based Positional Embeddings into Transformer Models for Source Code Representation

    Bartkowiak, Patryk and Grali \'n ski, Filip. Seamlessly Integrating Tree-Based Positional Embeddings into Transformer Models for Source Code Representation. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.10

    work page doi:10.18653/v1/2025.xllm-1.10 2025

  48. [48]

    Enhancing AMR Parsing with Group Relative Policy Optimization

    Barta, Botond and Hamerlik, Endre and Nyist, Mil \'a n and Ito, Masato and Acs, Judit. Enhancing AMR Parsing with Group Relative Policy Optimization. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.11

    work page doi:10.18653/v1/2025.xllm-1.11 2025

  49. [49]

    Structure Modeling Approach for UD Parsing of Historical M odern J apanese

    Ozaki, Hiroaki and Omura, Mai and Komiya, Kanako and Asahara, Masayuki and Ogiso, Toshinobu. Structure Modeling Approach for UD Parsing of Historical M odern J apanese. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.12

    work page doi:10.18653/v1/2025.xllm-1.12 2025

  50. [50]

    BARTABSA ++: Revisiting BARTABSA with Decoder LLM s

    Pfister, Jan and V. BARTABSA ++: Revisiting BARTABSA with Decoder LLM s. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.13

    work page doi:10.18653/v1/2025.xllm-1.13 2025

  51. [51]

    Typed- RAG : Type-Aware Decomposition of Non-Factoid Questions for Retrieval-Augmented Generation

    Lee, DongGeon and Park, Ahjeong and Lee, Hyeri and Nam, Hyeonseo and Maeng, Yunho. Typed- RAG : Type-Aware Decomposition of Non-Factoid Questions for Retrieval-Augmented Generation. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.14

    work page doi:10.18653/v1/2025.xllm-1.14 2025

  52. [52]

    Do we still need Human Annotators? Prompting Large Language Models for Aspect Sentiment Quad Prediction

    Hellwig, Nils Constantin and Fehle, Jakob and Kruschwitz, Udo and Wolff, Christian. Do we still need Human Annotators? Prompting Large Language Models for Aspect Sentiment Quad Prediction. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.15

    work page doi:10.18653/v1/2025.xllm-1.15 2025

  53. [53]

    Can LLM s Interpret and Leverage Structured Linguistic Representations? A Case Study with AMR s

    Raut, Ankush and Zhu, Xiaofeng and Pacheco, Maria Leonor. Can LLM s Interpret and Leverage Structured Linguistic Representations? A Case Study with AMR s. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.16

    work page internal anchor Pith review doi:10.18653/v1/2025.xllm-1.16 2025

  54. [54]

    LLM Dependency Parsing with In-Context Rules

    Ginn, Michael and Palmer, Alexis. LLM Dependency Parsing with In-Context Rules. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.17

    work page doi:10.18653/v1/2025.xllm-1.17 2025

  55. [55]

    Cognitive Mirroring for D oc RE : A Self-Supervised Iterative Reflection Framework with Triplet-Centric Explicit and Implicit Feedback

    Han, Xu and Wang, Bo and Sun, Yueheng and Zhao, Dongming and Qu, Zongfeng and He, Ruifang and Hou, Yuexian and Hu, Qinghua. Cognitive Mirroring for D oc RE : A Self-Supervised Iterative Reflection Framework with Triplet-Centric Explicit and Implicit Feedback. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025)...

    work page doi:10.18653/v1/2025.xllm-1.18 2025

  56. [56]

    Cross-Document Event-Keyed Summarization

    Walden, William and Kuchmiichuk, Pavlo and Martin, Alexander and Jin, Chihsheng and Cao, Angela and Sun, Claire and Allen, Curisia and White, Aaron. Cross-Document Event-Keyed Summarization. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.19

    work page doi:10.18653/v1/2025.xllm-1.19 2025

  57. [57]

    Transfer of Structural Knowledge from Synthetic Languages

    Budnikov, Mikhail and Yamshchikov, Ivan. Transfer of Structural Knowledge from Synthetic Languages. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.20

    work page doi:10.18653/v1/2025.xllm-1.20 2025

  58. [58]

    Language Models are Universal Embedders

    Zhang, Xin and Li, Zehan and Zhang, Yanzhao and Long, Dingkun and Xie, Pengjun and Zhang, Meishan and Zhang, Min. Language Models are Universal Embedders. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.21

    work page doi:10.18653/v1/2025.xllm-1.21 2025

  59. [59]

    D ia DP @ XLLM 25: Advancing C hinese Dialogue Parsing via Unified Pretrained Language Models and Biaffine Dependency Scoring

    Duan, Shuoqiu and Chen, Xiaoliang and Miao, Duoqian and Gu, Xu and Li, Xianyong and Du, Yajun. D ia DP @ XLLM 25: Advancing C hinese Dialogue Parsing via Unified Pretrained Language Models and Biaffine Dependency Scoring. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.22

    work page doi:10.18653/v1/2025.xllm-1.22 2025

  60. [60]

    LLMSR @ XLLM 25: Less is More: Enhancing Structured Multi-Agent Reasoning via Quality-Guided Distillation

    Yuan, Jiahao and Sun, Xingzhe and Yu, Xing and Wang, Jingwen and Du, Dehui and Cui, Zhiqing and Di, Zixiang. LLMSR @ XLLM 25: Less is More: Enhancing Structured Multi-Agent Reasoning via Quality-Guided Distillation. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.23

    work page doi:10.18653/v1/2025.xllm-1.23 2025

  61. [61]

    S peech EE @ XLLM 25: End-to-End Structured Event Extraction from Speech

    Chaudhuri, Soham and Biswas, Diganta and Saha, Dipanjan and Das, Dipankar and Bandyopadhyay, Sivaji. S peech EE @ XLLM 25: End-to-End Structured Event Extraction from Speech. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.24

    work page doi:10.18653/v1/2025.xllm-1.24 2025

  62. [62]

    Luu, Son and Van Nguyen, Kiet

    Pham Hoang Le, Nguyen and Dinh Thien, An and T. Luu, Son and Van Nguyen, Kiet. D oc IE @ XLLM 25: Z ero S emble - Robust and Efficient Zero-Shot Document Information Extraction with Heterogeneous Large Language Model Ensembles. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.25

    work page doi:10.18653/v1/2025.xllm-1.25 2025

  63. [63]

    D oc IE @ XLLM 25: In-Context Learning for Information Extraction using Fully Synthetic Demonstrations

    Popovic, Nicholas and Kangen, Ashish and Schopf, Tim and F. D oc IE @ XLLM 25: In-Context Learning for Information Extraction using Fully Synthetic Demonstrations. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.26

    work page doi:10.18653/v1/2025.xllm-1.26 2025

  64. [64]

    LLMSR @ XLLM 25: Integrating Reasoning Prompt Strategies with Structural Prompt Formats for Enhanced Logical Inference

    Tai, Le and Van, Thin. LLMSR @ XLLM 25: Integrating Reasoning Prompt Strategies with Structural Prompt Formats for Enhanced Logical Inference. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.27

    work page doi:10.18653/v1/2025.xllm-1.27 2025

  65. [65]

    D oc IE @ XLLM 25: UIEP rompter: A Unified Training-Free Framework for universal document-level information extraction via Structured Prompt

    Qiu, Chengfeng and Zhou, Lifeng and Wei, Kaifeng and Li, Yuke. D oc IE @ XLLM 25: UIEP rompter: A Unified Training-Free Framework for universal document-level information extraction via Structured Prompt. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.28

    work page doi:10.18653/v1/2025.xllm-1.28 2025

  66. [66]

    LLMSR @ XLLM 25: SWRV : Empowering Self-Verification of Small Language Models through Step-wise Reasoning and Verification

    Chen, Danchun. LLMSR @ XLLM 25: SWRV : Empowering Self-Verification of Small Language Models through Step-wise Reasoning and Verification. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.29

    work page doi:10.18653/v1/2025.xllm-1.29 2025

  67. [67]

    LLMSR @ XLLM 25: An Empirical Study of LLM for Structural Reasoning

    Li, Xinye and Wan, Mingqi and Sui, Dianbo. LLMSR @ XLLM 25: An Empirical Study of LLM for Structural Reasoning. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.30

    work page doi:10.18653/v1/2025.xllm-1.30 2025

  68. [68]

    LLMSR @ XLLM 25: A Language Model-Based Pipeline for Structured Reasoning Data Construction

    Xing, Hongrui and Liu, Xinzhang and Jiang, Zhuo and Yang, Zhihao and Yao, Yitong and Wang, Zihan and Deng, Wenmin and Wang, Chao and Song, Shuangyong and Yang, Wang and He, Zhongjiang and Li, Yongxiang. LLMSR @ XLLM 25: A Language Model-Based Pipeline for Structured Reasoning Data Construction. Proceedings of the 1st Joint Workshop on Large Language Model...

    work page doi:10.18653/v1/2025.xllm-1.31 2025

  69. [69]

    S peech EE @ XLLM 25: Retrieval-Enhanced Few-Shot Prompting for Speech Event Extraction

    Gedeon, M \'a t \'e. S peech EE @ XLLM 25: Retrieval-Enhanced Few-Shot Prompting for Speech Event Extraction. Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025). 2025. doi:10.18653/v1/2025.xllm-1.32

    work page doi:10.18653/v1/2025.xllm-1.32 2025

  70. [70]

    Computational Sanskrit and Digital Humanities - World Sanskrit Conference 2025. 2025

    2025

  71. [71]

    An introduction to computational identification and classification of Upam \= a alaṇk \= a ra

    Jadhav, Bhakti and Dutta, Himanshu and Kanitkar, Shruti and Kulkarni, Malhar and Bhattacharyya, Pushpak. An introduction to computational identification and classification of Upam \= a alaṇk \= a ra. Computational Sanskrit and Digital Humanities - World Sanskrit Conference 2025. 2025

    2025

  72. [72]

    Aesthetics of S anskrit Poetry from the Perspective of Computational Linguistics: A Case Study Analysis on \'S ikṣ \= a ṣṭaka

    Sandhan, Jivnesh and Barbadikar, Amruta and Maity, Malay and Satuluri, Pavankumar and Sandhan, Tushar and Gupta, Ravi M and Goyal, Pawan and Behera, Laxmidhar. Aesthetics of S anskrit Poetry from the Perspective of Computational Linguistics: A Case Study Analysis on \'S ikṣ \= a ṣṭaka. Computational Sanskrit and Digital Humanities - World Sanskrit Confere...

    2025

  73. [73]

    Itaretara Dvandva: A challenge for Dependency Tree semantics

    Kulkarni, Amba and Neelamana, Vasudha. Itaretara Dvandva: A challenge for Dependency Tree semantics. Computational Sanskrit and Digital Humanities - World Sanskrit Conference 2025. 2025

    2025

  74. [74]

    A Case Study of Handwritten Text Recognition from Pre-Colonial era S anskrit Manuscripts

    Chincholikar, Kartik and Dwivedi, Shagun and Gopalan, Kaushik and Awasthi, Tarinee. A Case Study of Handwritten Text Recognition from Pre-Colonial era S anskrit Manuscripts. Computational Sanskrit and Digital Humanities - World Sanskrit Conference 2025. 2025

    2025

  75. [75]

    Towards Accent-Aware V edic S anskrit Optical Character Recognition Based on Transformer Models

    Tsukagoshi, Yuzuki and Kuroiwa, Ryo and Ohmukai, Ikki. Towards Accent-Aware V edic S anskrit Optical Character Recognition Based on Transformer Models. Computational Sanskrit and Digital Humanities - World Sanskrit Conference 2025. 2025

    2025

  76. [76]

    Vedavani: A Benchmark Corpus for ASR on V edic S anskrit Poetry

    Kumar, Sujeet and Ray, Pretam and Beerukuri, Abhinay and Kamoji, Shrey and Jagadeeshan, Manoj Balaji and Goyal, Pawan. Vedavani: A Benchmark Corpus for ASR on V edic S anskrit Poetry. Computational Sanskrit and Digital Humanities - World Sanskrit Conference 2025. 2025

    2025

  77. [77]

    Compound Type Identification in S anskrit

    Krishnan, Sriram and Satuluri, Pavankumar and Barbadikar, Amruta and Prasanna Venkatesh, T S and Kulkarni, Amba. Compound Type Identification in S anskrit. Computational Sanskrit and Digital Humanities - World Sanskrit Conference 2025. 2025

    2025

  78. [78]

    IKML : A Markup Language for Collaborative Semantic Annotation of I ndic Texts

    Lakkundi, Chaitanya S and Rajaraman, Gopalakrishnan and Susarla, Sai Rama Krishna. IKML : A Markup Language for Collaborative Semantic Annotation of I ndic Texts. Computational Sanskrit and Digital Humanities - World Sanskrit Conference 2025. 2025

    2025

  79. [79]

    Challenges in Processing V edic S anskrit: Towards creating a normalized dataset for the Ṛgveda-saṃhit \= a

    Krishnan, Sriram and Gayathri, Sepuri and Kulkarni, Amba. Challenges in Processing V edic S anskrit: Towards creating a normalized dataset for the Ṛgveda-saṃhit \= a. Computational Sanskrit and Digital Humanities - World Sanskrit Conference 2025. 2025

    2025

  80. [80]

    P \= a ṇḍitya: Visualizing S anskrit Intellectual Networks

    Neill, Tyler. P \= a ṇḍitya: Visualizing S anskrit Intellectual Networks. Computational Sanskrit and Digital Humanities - World Sanskrit Conference 2025. 2025

    2025

Showing first 80 references.