# NTILC: Neural Tool Invocation via Learned Compression

arXiv: https://arxiv.org/abs/2606.06566 · HTML: https://mlsanity.com/paper/2606.06566/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Tool learning and function calling, one of 16 directions in LLM agents and tool use.
- Builds on 15 of its 18 references that are in the map.
- First posted on arXiv 4 Jun 2026.

## Placed in

- LLM agents and tool use / [Tool learning and function calling](https://mlsanity.com/agents/tool-learning-and-function-calling/)

## Builds on (most cited first, 15 of 15)

- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [Billion-scale similarity search with GPUs](https://mlsanity.com/paper/1702.08734/) (2017)
- [ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs](https://mlsanity.com/paper/2307.16789/) (2023)
- [Gorilla: Large Language Model Connected with Massive APIs](https://mlsanity.com/paper/2305.15334/) (2023)
- [API-Bank: A Comprehensive Benchmark for Tool-Augmented LLMs](https://mlsanity.com/paper/2304.08244/) (2023)
- [Circle Loss: A Unified Perspective of Pair Similarity Optimization](https://mlsanity.com/paper/2002.10857/) (2020)
- [MetaTool Benchmark for Large Language Models: Deciding Whether to Use Tools and Which to Use](https://mlsanity.com/paper/2310.03128/) (2023)
- [Efficient Guided Generation for Large Language Models](https://mlsanity.com/paper/2307.09702/) (2023)
- [Agentic Large Language Models, a survey](https://mlsanity.com/paper/2503.23037/) (2025)
- [ToolEyes: Fine-Grained Evaluation for Tool Learning Capabilities of Large Language Models in Real-world Scenarios](https://mlsanity.com/paper/2401.00741/) (2024)
- [Efficient Solutions For An Intriguing Failure of LLMs: Long Context Window Does Not Mean LLMs Can Analyze Long Sequences Flawlessly](https://mlsanity.com/paper/2408.01866/) (2024)
- [Context Is What You Need: The Maximum Effective Context Window for Real World Limits of LLMs](https://mlsanity.com/paper/2509.21361/) (2025)
- [Teaching Language Models to Reason with Tools](https://mlsanity.com/paper/2510.20342/) (2025)
- [Intelligence Degradation in Long-Context LLMs: Critical Threshold Determination via Natural Length Distribution Analysis](https://mlsanity.com/paper/2601.15300/) (2026)
- [Augmenting In-Context-Learning in LLMs via Automatic Data Labeling and Refinement](https://mlsanity.com/paper/2410.10348/) (2024)

## Built on by (most cited first, 1 of 1)

- [Nexus: Depth-Adaptive KV-Cache Splicing and Retrieval-Decoupled Tool Routing for Agentic LLMs on Unified Memory](https://mlsanity.com/paper/2608.20397/) (2026)

## Abstract (arXiv)

Agentic tool-calling language models depend on large registries of callable APIs, functions, and local actions. Placing full tool specifications directly in the prompt incurs a cost that scales linearly with the size of the tool registry, rapidly consuming the context budget. As the registry grows, this leads to higher latency and degrades selection accuracy, particularly due to interference from irrelevant tools. We overcome these limitations by introducing NTILC, a neural tool selection and invocation framework that replaces in-context registry look-up with learned latent retrieval. NTILC maps both user intent and tool specifications into a shared embedding space, enabling tool selection via external retrieval rather than in-context lookup. The language model is conditioned only on the selected tool schema, allowing for precise, constrained argument generation. Central to our approach is a signature-aware composite objective, which augments semantic similarity with constraints derived from tool signatures (e.g., argument schema, type compatibility, and return types). By combining Circle Loss with a Functional Margin Loss, the model enforces separation between tools that are semantically similar but incompatible under their execution signatures. We evaluate NTILC on public tool-selection and function-calling datasets and report context token usage, retrieval accuracy, and selection latency metrics. Across these settings, NTILC reduces context window consumption by over 95% and inference latency by up to 74% compared to long-context ICT baselines.