Text to FigureText → Imageposter

Hammer Function-Calling Models with Masking — Poster

This poster introduces Hammer, a family of function-calling models using function masking to improve robustness. It details the method, shows performance gains on benchmarks like BFCL, and analyzes masking ratios.

Paper context

Paper title: Robust Function-Calling for On-Device Language Models via Function Masking Abstract: This poster introduces Hammer, a family of function-calling models using function masking to improve robustness. It details the method, shows performance gains on benchmarks like BFCL, and analyzes masking ratios. Paper body (method & results): <!DOCTYPE html> <html lang="en"> <head> <meta content="text/html; charset=utf-8" http-equiv="content-type"/> <title>Hammer: Robust Function-Calling for On-Device Language Models via Function Masking</title>  <meta content="width=device-width, initial-scale=1, shrink-to-fit=no" name="viewport"/> <link href="https://cdn.jsdelivr.net/npm/bootstrap@5.3.0/dist/css/bootstrap.min.css" rel="stylesheet" type="text/css"/> <link href="/static/browse/0.3.4/css/ar5iv.0.7.9.min.css" rel="stylesheet" type="text/css"/> <link href="/static/browse/0.3.4/css/ar5iv-fonts.0.7.9.min.css" rel="stylesheet" type="text/css"/> <link href="/static/browse/0.3.4/css/latexml_styles.css" rel="stylesheet" type="text/css"/> <script src="https://cdn.jsdelivr.net/npm/bootstrap@5.3.0/dist/js/bootstrap.bundle.min.js"></script> <script src="https://cdnjs.cloudflare.com/ajax/libs/html2canvas/1.3.3/html2canvas.min.js"></script> <script src="/static/browse/0.3.4/js/addons_new.js"></script> <script src="/static/browse/0.3.4/js/feedbackOverlay.js"></script> <base href="/html/2410.04587v1/"/></head> <body> <nav class="ltx_page_navbar"> <nav class="ltx_TOC"> <ol class="ltx_toclist"> <li class="ltx_tocentry ltx_tocentry_section"><a class="ltx_ref" href="https://arxiv.org/html/2410.04587v1#S1" title="In Hammer: Robust Function-Calling for On-Device Language Models via Function Masking">1 Introduction</a></li> <li class="ltx_tocentry ltx_tocentry_section"><a class="ltx_ref" href="https://arxiv.org/html/2410.04587v1#S2" title="In Hammer: Robust Function-Calling for On-Device Language Models via Function Masking">2 Related Works</a></li> <li class="ltx_tocentry ltx_tocentry_section"> <a class="ltx_ref" href="https://arxiv.org/html/2410.04587v1#S3" title="In Hammer: Robust Function-Calling for On-Device Language Models via Function Masking">3 Problem Statement and Analysis</a> <ol class="ltx_toclist ltx_toclist_section"> <li class="ltx_tocentry ltx_tocentry_subsection"><a class="ltx_ref" href="https://arxiv.org/html/2410.04587v1#S3.SS1" title="In 3 Problem Statement and Analysis ‣ Hammer: Robust Function-Calling for On-Device Language Models via Function Masking">3.1 Misleadingness by Function Name and Parameter Name</a></li> <li class="ltx_tocentry ltx_tocentry_subsection"><a class="ltx_ref" href="https://arxiv.org/html/2410.04587v1#S3.SS2" title="In 3 Problem Statement and Analysis ‣ Hammer: Robust Function-Calling for On-Device Language Models via Function Masking">3.2 The Impact of Excessive Focus on the Naming</a></li> </ol> </li> <li class="ltx_tocentry ltx_tocentry_section"> <a class="ltx_ref" href="https://arxiv.org/html/2410.04587v1#S4" title="In Hammer: Robust Function-Calling for On-Device Language Models via Function Masking">4 Methodology</a> <ol class="ltx_toclist ltx_toclist_section"> <li class="ltx_tocentry ltx_tocentry_subsection"><a class="ltx_ref" href="https://arxiv.org/html/2410.04587v1#S4.SS1" title="In 4 Methodology ‣ Hammer: Robust Function-Calling for On-Device Language Models via Function Masking">4.1 Function Masking</a></li> <li class="ltx_tocentry ltx_tocentry_subsection"><a class="ltx_ref" href="https://arxiv.org/html/2410.04587v1#S4.SS2" title="In 4 Methodology ‣ Hammer: Robust Function-Calling for On-Device Language Models via Function Masking">4.2 Irrelevance-Augmented Dataset</a></li> </ol> </li> <li class="ltx_tocentry ltx_tocentry_section"> <a class="ltx_ref" href="https://arxiv.org/html/2410.04587v1#S5" title="In Hammer: Robust Function-Calling for On-Device Language Models via Function Masking">5 Evaluation</a> <ol class="ltx_toclist ltx_toclist_section"> <li class="ltx_tocentry ltx_tocentry_subsection"><a class="ltx_ref" href="https://arxiv.org/html/2410.04587v1#S5.SS1" title="In 5 Evaluation ‣ Hammer: Robust Function-Calling for On-Device Language Models via Function Masking">5.1 Experimental Setup</a></li> <li class="ltx_tocentry ltx_tocentry_subsection"><a class="ltx_ref" href="https://arxiv.org/html/2410.04587v1#S5.SS2" title="In 5 Evaluation ‣ Hammer: Robust Function-Calling for On-Device Language Models via Function Masking">5.2 Overall Performance on Various Benchmarks</a></li> <li class="ltx_tocentry ltx_tocentry_subsection"><a class="ltx_ref" href="https://arxiv.org/html/2410.04587v1#S5.SS3" title="In 5 Evaluation ‣ Hammer: Robust Function-Calling for On-Device Language Models via Function Masking">5.3 Detailed Performance on Different Types of Function Calling</a></li> <li class="ltx_tocentry ltx_tocentry_subsection"><a class="ltx_ref" href="https://arxiv.org/html/2410.04587v1#S5.SS4" title="In 5 Evaluation ‣ Hammer: Robust Function-Calling for On-Device Language Models via Function Masking">5.4 Ablation on Different Base Models</a></li> <li class="ltx_tocentry ltx_tocentry_subsection"><a class="ltx_ref" href="https://arxiv.org/html/2410.04587v1#S5.SS5" title="In 5 Evaluation ‣ Hammer: Robust Function-Calling for On-Device Language Models via Function Masking">5.5 Ablation on Different Masking Ratio</a></li> <li class="ltx_tocentry ltx_tocentry_subsection"><a class="ltx_ref" href="https://arxiv.org/html/2410.04587v1#S5.SS6" title="In 5 Evaluation ‣ Hammer: Robust Function-Calling for On-Device Language Models via Function Masking">5.6 Ablation on Different Proportions of Irrelevance-Augmented Data</a></li> </ol> </li> <li class="ltx_tocentry ltx_tocentry_section"><a class="ltx_ref" href="https://arxiv.org/html/2410.04587v1#S6" title="In Hammer: Robust Function-Calling for On-Device Language Models via Function Masking">6 Conclusion</a></li> <li class="ltx_tocentry ltx_tocentry_appendix"><a class="ltx_ref" href="https://arxiv.org/html/2410.04587v1#A1" title="In Hammer: Robust Function-Calling for On-Device Language Models via Function Masking">A Extra Experimental Results</a></li> <li class="ltx_tocentry ltx_tocentry_appendix"><a class="ltx_ref" href="https://arxiv.org/html/2410.04587v1#A2" title="In Hammer: Robust Function-Calling for On-Device Language Models via Function Masking">B Different Types of Function-calling Tasks</a></li> <li class="ltx_tocentry ltx_tocentry_appendix"><a class="ltx_ref" href="https://arxiv.org/html/2410.04587v1#A3" title="In Hammer: Robust Function-Calling for On-Device Language Models via Function Masking">C Example Input to Models with Function Masking</a></li> </ol></nav> </nav> <div class="ltx_page_main"> <div class="ltx_page_content"> <article class="ltx_document ltx_authors_1line"> <h1 class="ltx_title ltx_title_document"> Hammer: Robust Function-Calling for On-Device Language Models via Function Masking </h1> <div class="ltx_authors"> Qiqiang Lin∗ linqiqiang1@oppo.com Muning Wen∗ muningwen@sjtu.edu.cn Qiuying Peng∗,+ pengqiuying@oppo.com Guanyu Nie nieg@iastate.edu Junwei Liao jwliao.ai@gmail.com Jun Wang wangjun15@oppo.com Xiaoyun Mo moxiaoyun@oppo.com Jiamu Zhou zhoujiamu@oppo.com Cheng Cheng chengcheng2@oppo.com Yin Zhao zhaoyin@oppo.com Jun Wang+ wangjun7@oppo.com Weinan Zhang+ wnzhang@sjtu.edu.cn </div> <div class="ltx_abstract"> <h6 class="ltx_title ltx_title_abstract">Abstract</h6> Large language models have demonstrated impressive value in performing as autonomous agents when equipped with external tools and API calls. Nonetheless, effectively harnessing their potential for executing complex tasks crucially relies on enhancements in their function-calling capabilities. This paper identifies a critical gap in existing function-calling models, where performance varies significantly across benchmarks, often due to being misled by specific naming conventions. To address such an issue, we introduce Hammer, a novel family of foundation models specifically engineered for on-device function calling. Hammer employs an augmented dataset that enhances models’ sensitivity to irrelevant functions and

The prompt

Above I've shared:
(1) the full paper text,
(2) all paper figures labeled by figure number,
(3) the caption for the central poster figure I'm building.

TASK: This is a CONFERENCE POSTER. **NOT** an academic-paper figure.
Style requirements:

  - Multi-section layout with a clear poster structure: large title banner
    at the top with the paper title + author/affiliation strip, then 3-6
    distinct content panels arranged in columns or a grid.
  - Large legible fonts (text must be readable at 2 m viewing distance) —
    headings ≥ 60 pt visual size in the final image.
  - Use colour blocks / panel backgrounds to delineate sections (this is
    what makes it a poster, not a single-figure diagram).
  - Aspect ratio: portrait or landscape rectangle, NOT square.

If your output looks like a standard academic-paper figure (single panel,
no title banner, dense small text, no colour blocks), you've failed the
task. Render the COMPLETE poster, not just the central figure.

Just give me the final poster image.

Try this prompt now

Open it inside the generator with the prompt pre-filled.

Try this prompt

Related prompts

Text to Figure

Hammer Function-Calling Models with Masking — Poster

Paper context

The prompt

Try this prompt now

Related prompts

PINN Framework Pipeline for Microscopy Data

DP-FTRL Privacy Amplification Analysis — Conference Poster

NC-TTT: Noise-Contrastive Test-Time Training Poster