{"id":43157,"date":"2025-07-27T15:55:31","date_gmt":"2025-07-27T19:55:31","guid":{"rendered":"https:\/\/www.bloomberg.com\/company\/stories\/bloombergs-ai-engineers-introduce-an-improved-agent-tool-calling-methodology-acl-2025\/"},"modified":"2025-07-27T15:55:31","modified_gmt":"2025-07-27T19:55:31","slug":"bloombergs-ai-engineers-introduce-an-improved-agent-tool-calling-methodology-acl-2025","status":"publish","type":"post","link":"https:\/\/www.bloomberg.com\/company\/stories\/bloombergs-ai-engineers-introduce-an-improved-agent-tool-calling-methodology-acl-2025\/","title":{"rendered":"Bloomberg’s AI engineers introduce an improved agent tool-calling methodology at ACL 2025"},"content":{"rendered":"
\n \n\t\n\t\n\t
\n\t\t\n\t\t\t
\n\t
\n \n

During the 63rd Annual Meeting of the Association for Computational Linguistics<\/a> (ACL 2025) this week in Vienna, Austria, researchers from Bloomberg\u2019s AI Engineering group in London are showcasing their expertise in large language models (LLMs) and tool-based agentic AI with their paper \u201cA Joint Optimization Framework for Enhancing Efficiency of Tool Utilization in LLM Agents<\/a>.\u201d<\/p>\n

In the paper, which is published “Findings of the Association for Computational Linguistics: ACL 2025<\/a>,” Bin Wu, a Bloomberg Data Science Ph.D. Fellow<\/a> and Ph.D. student at University College London, Edgar Meij, Head of AI Platforms in Bloomberg\u2019s AI Engineering group, and Emine Yilmaz<\/a>, a professor and EPSRC Fellow at University College London\u2019s Department of Computer Science \u2013 where she also leads the Web Intelligence Group at the UCL Centre for Artificial Intelligence \u2013 demonstrate the crucial role of the instructions provided in agent prompts and tool descriptions \u2013 collectively referred to as context. Incomplete or suboptimal context in the instructions and tool descriptions significantly increases the required number of tool calls that LLMs need to make in order to generate an adequate response, leading to computational overhead. They propose a new methodology for automatically improving agent prompts and tool descriptions, and demonstrate that it substantially reduces the number of tool calls the LLM agent needs to make.<\/p>\n\n<\/div>\n

\n \"\"\"\"\n \n<\/figure>\n
\n \n

In addition, two members of Bloomberg\u2019s AI Strategy & Research team in the company\u2019s CTO Office \u2013 Sebastian Gehrmann, Head of Responsible AI, and Enrico Santus, Principal Technical Strategist for Human-AI Interaction and Academic Engagement \u2013 are two of the organizers of the fourth iteration of the Generation, Evaluation & Metrics Workshop (GEM2)<\/a>, which will be held as part of ACL on July 31, 2025. In light of the broad accessibility of LLMs, this workshop will serve as a forum for researchers and practitioners from both the natural language processing and machine learning communities to come together to explore potential approaches and research directions to address the broader types of natural language generation (NLG) challenges \u2013 in particular, the evaluation of model-generated outputs. While these advanced models can generate fluent text, ensuring the usefulness, quality, and fairness of their output is essential to help bridge the gap between research and real-world applications.<\/p>\n

We asked the paper\u2019s lead author and one of the workshop organizers to explain why their work is notable in advancing the state-of-the-art with regards to LLMs and agentic AI:<\/p>\n

\n

Wednesday, July 30, 2025<\/u><\/strong><\/h3>\n

Session 12: IP-Posters (Findings Posters – In-Person 4)<\/em>
\n11:00-12:30 CEST<\/em><\/p>\n

A Joint Optimization Framework for Enhancing Efficiency of Tool Utilization in LLM Agents<\/strong><\/a>
\nBin Wu (Centre for Artificial Intelligence, University College London), Edgar Meij (Bloomberg), Emine Yilmaz (Centre for Artificial Intelligence, University College London)<\/p>\n\n<\/div>\n

\n \"Click\"Click<\/a>\n \n<\/figure>\n
\n \n

Please summarize your research. Why are your results notable?<\/strong><\/p>\n

Bin Wu:<\/strong> This research proposes a joint optimization framework that aims to improve the efficiency of tool-augmented LLM agents by systematically refining both agent instructions and tool descriptions. Traditional approaches have either focused on enhancing tool use effectiveness through reasoning strategies \u2013 like chain-of-thought (CoT) or tree-of-thoughts (ToT) prompting \u2013 or optimized only a single aspect (i.e., either the instructions or the tool documentation). However, these prior methods incur high computational costs and often overlook efficiency, particularly under conditions where context is incomplete.<\/p>\n

Our proposed framework introduces a three-stage process:<\/p>\n