From Fixed Keys to Readable Schemas: Small Language Models for Vehicle Agent Function Calls

• URL: http://arxiv.org/abs/2609.09476v1
• Date: Tue, 08 Sep 2026 21:46:35 GMT
• Title: From Fixed Keys to Readable Schemas: Small Language Models for Vehicle Agent Function Calls
• Abstract summary: In-vehicle assistants must translate natural-language requests into accurate vehicle function calls under strict memory and latency constraints.<n>Two approaches are to represent each function with a dedicated Functional Token (FT) or provide function schemas directly in the prompt.<n>FTs enable compact inference but are restricted to functions learned during training, whereas SIP can generalize to unseen functions at the cost of longer prompts.
• Score: 9.651124018371158
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: In-vehicle assistants must translate natural-language requests into accurate vehicle function calls under strict memory and latency constraints, making small language models (SLMs) attractive for on-device deployment. For such models, a key design choice is how the available function surface is presented. Two approaches are to represent each function with a dedicated Functional Token (FT) or provide function schemas directly in the prompt. FTs enable compact inference but are restricted to functions learned during training, whereas Schema-in-Prompt (SIP) can generalize to unseen functions at the cost of longer prompts and higher inference overhead. We introduce a benchmark of 9,822 single-turn examples spanning 79 vehicle functions derived from Android Automotive, including held-out functions and requests requiring refusal. We compare both approaches under matched fine-tuning across four SLMs from 270M to 1.7B parameters. On functions seen during training, scaling provides limited benefit: the 270M model can match the 1.7B model, while the strongest overall performance occurs at 0.6B. On held-out functions, FT achieves zero accuracy by construction, whereas SIP generalizes and improves substantially with scale. On out-of-scope requests, FT can invoke an unavailable function it was trained to emit, while SIP more reliably refuses based on the functions offered. This flexibility comes with higher memory use and latency. Our theoretical analysis explains how SIP enables generalization and why longer schema contexts increase inference cost. Overall, function-surface representation, rather than model scale alone, determines the capabilities and failure modes of SLM-based vehicle function calling.

Related papers

• Small Reasoning Models are Instruction Followers in Function Calling [4.549831511476248]
  Function calling represents the core capability of agentic large language models (LLMs)<n>We introduce Instruction-Followed Function Calling (IFFC), a novel framework that decouples function-calling logic from the primary LLM.<n>IFFC consistently outperforms both native function calling (NFC) and prompt-based function calling (PFC) baselines.
  arXiv  Detail & Related papers  (2026-08-23T15:51:26Z)
• LFM2 Technical Report [87.58431408281973]
  We present LFM2, a family of Liquid Foundation Models designed for efficient on-device deployment and strong task capabilities.<n>The LFM2 family covers 350M-8.3B parameters, including dense models (350M, 700M, 1.2B, 2.6B) and a mixture-of-experts variant (8.3B total, 1.5B active)<n>We build multimodal and retrieval variants: LFM2-VL for vision-latency tasks, LFM2-Audio for speech, and LFM2-ColBERT for retrieval.
  arXiv  Detail & Related papers  (2025-11-28T17:56:35Z)
• SlsReuse: LLM-Powered Serverless Function Reuse [0.0]
  This paper presents SlsReuse, the first LLM-powered framework for serverless function reuse.<n>It learns unified semantic-enhanced representations of heterogeneous functions through effective prompt engineering.<n>Built on ChatGPT-4o, one of the most representative LLMs, SlsReuse achieves Recall@10 of 91.20%, exceeding the state-of-the-art baseline by 24.53 percentage points.
  arXiv  Detail & Related papers  (2025-11-21T14:08:12Z)
• Small Models, Big Tasks: An Exploratory Empirical Study on Small Language Models for Function Calling [6.102559098873098]
  Function calling is a complex task with widespread applications in domains such as information retrieval, software engineering and automation.<n>Large Language Models (LLMs) can automate this process but are computationally expensive and impractical in resource-constrained settings.<n>Small Language Models (SLMs) can operate efficiently, offering faster response times, and lower computational demands.
  arXiv  Detail & Related papers  (2025-04-27T15:26:51Z)
• Magnet: Multi-turn Tool-use Data Synthesis and Distillation via Graph Translation [85.68881632498909]
  We propose a principled framework for synthesizing high-quality training trajectories for large language model agents.<n>The framework is based on automatic and iterative translations from a function signature path to a sequence of queries and executable function calls.<n> Experiments show that training with the positive trajectories with supervised fine-tuning and preference optimization against negative trajectories, our 14B model, Magnet-14B-mDPO, obtains 68.01 on BFCL-v3 and 73.30 on ToolQuery.
  arXiv  Detail & Related papers  (2025-03-10T20:13:07Z)
• Reasoning with Reinforced Functional Token Tuning [70.96651128307985]
  We propose Reinforced Functional Token Tuning (RFTT) to empower Large Language Models (LLMs) with self-play learn-to-reason capabilities.<n>RFTT embeds a rich set of learnable functional tokens directly into the model vocabulary, enabling chain-of-thought construction with diverse human-like reasoning behaviors.
  arXiv  Detail & Related papers  (2025-02-19T02:59:42Z)
• Scaling Embedding Layers in Language Models [61.939921364422936]
  $SCONE$ is a new method for extending input embedding layers to enhance language model performance.<n>$SCONE$ retains the original vocabulary while introducing embeddings for a set of frequent n-grams.<n>These embeddings provide contextualized representation for each input token and are learned with a separate model during training.<n>$SCONE$ enables two new scaling strategies: increasing the number of n-gram embeddings and scaling the model used to learn them, both while maintaining fixed accelerator usage during inference.
  arXiv  Detail & Related papers  (2025-02-03T18:59:32Z)
• Unlock the Correlation between Supervised Fine-Tuning and Reinforcement Learning in Training Code Large Language Models [12.656574142412484]
  We make an attempt to understand the correlation between supervised fine-tuning and reinforcement learning.<n>We find that both atomic and synthetic functions are indispensable for SFT's generalization.
  arXiv  Detail & Related papers  (2024-06-14T03:39:01Z)
• SpaFL: Communication-Efficient Federated Learning with Sparse Models and Low computational Overhead [75.87007729801304]
  SpaFL: a communication-efficient FL framework is proposed to optimize sparse model structures with low computational overhead.<n>To optimize the pruning process itself, only thresholds are communicated between a server and clients instead of parameters.<n>Global thresholds are used to update model parameters by extracting aggregated parameter importance.
  arXiv  Detail & Related papers  (2024-06-01T13:10:35Z)
• First-Order Stable Model Semantics with Intensional Functions [2.1955512452222696]
  We extend the first-order stable model semantics by Ferraris, Lee, and Lifschitz to allow intensional functions. We use this extension as a basis for defining Answer Set Programming Modulo Theories (ASPMT) ASPMT can be applied to domains containing real numbers and alleviates the grounding problem.
  arXiv  Detail & Related papers  (2023-07-15T06:03:35Z)
• FedYolo: Augmenting Federated Learning with Pretrained Transformers [61.56476056444933]
  In this work, we investigate pretrained transformers (PTF) to achieve on-device learning goals. We show that larger scale shrinks the accuracy gaps between alternative approaches and improves robustness. Finally, it enables clients to solve multiple unrelated tasks simultaneously using a single PTF.
  arXiv  Detail & Related papers  (2023-07-10T21:08:52Z)
• AutoFIS: Automatic Feature Interaction Selection in Factorization Models for Click-Through Rate Prediction [75.16836697734995]
  We propose a two-stage algorithm called Automatic Feature Interaction Selection (AutoFIS) AutoFIS can automatically identify important feature interactions for factorization models with computational cost just equivalent to training the target model to convergence. AutoFIS has been deployed onto the training platform of Huawei App Store recommendation service.
  arXiv  Detail & Related papers  (2020-03-25T06:53:54Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.