# Task-Specific Efficiency Analysis: When Small Language Models Outperform Large Language Models

arXiv: https://arxiv.org/abs/2603.21389 · HTML: https://mlsanity.com/paper/2603.21389/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- Builds on 41 of its 65 references that are in the map.
- First posted on arXiv 22 Mar 2026.

## Builds on (most cited first, 30 of 41)

- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [Know What You Don't Know: Unanswerable Questions for SQuAD](https://mlsanity.com/paper/1806.03822/) (2018)
- [MiniCPM: Unveiling the Potential of Small Language Models with Scalable Training Strategies](https://mlsanity.com/paper/2404.06395/) (2024)
- [It's Not Just Size That Matters: Small Language Models Are Also Few-Shot Learners](https://mlsanity.com/paper/2009.07118/) (2020)
- [A Comprehensive Survey of Small Language Models in the Era of Large Language Models: Techniques, Enhancements, Applications, Collaboration with LLMs, and Trustworthiness](https://mlsanity.com/paper/2411.03350/) (2024)
- [OstQuant: Refining Large Language Model Quantization with Orthogonal and Scaling Transformations for Better Distribution Fitting](https://mlsanity.com/paper/2501.13987/) (2025)
- [VL-JEPA: Joint Embedding Predictive Architecture for Vision-language](https://mlsanity.com/paper/2512.10942/) (2025)
- [FineCIR: Explicit Parsing of Fine-Grained Modification Semantics for Composed Image Retrieval](https://mlsanity.com/paper/2503.21309/) (2025)
- [VisRL: Intention-Driven Visual Perception via Reinforced Reasoning](https://mlsanity.com/paper/2503.07523/) (2025)
- [MQuant: Unleashing the Inference Potential of Multimodal Large Language Models via Full Static Quantization](https://mlsanity.com/paper/2502.00425/) (2025)
- [BALF: Simple and Efficient Blur Aware Local Feature Detector](https://mlsanity.com/paper/2211.14731/) (2022)
- [Human Motion Instruction Tuning](https://mlsanity.com/paper/2411.16805/) (2024)
- [What Is The Best 3D Scene Representation for Robotics? From Geometric to Foundation Models](https://mlsanity.com/paper/2512.03422/) (2025)
- [SGLP: A Similarity Guided Fast Layer Partition Pruning for Compressing Large Deep Models](https://mlsanity.com/paper/2410.14720/) (2024)
- [SIFThinker: Spatially-Aware Image Focus for Visual Reasoning](https://mlsanity.com/paper/2508.06259/) (2025)
- [SyreaNet: A Physically Guided Underwater Image Enhancement Framework Integrating Synthetic and Real Images](https://mlsanity.com/paper/2302.08269/) (2023)
- [Does Tone Change the Answer? Evaluating Prompt Politeness Effects on Modern LLMs: GPT, Gemini, and LLaMA](https://mlsanity.com/paper/2512.12812/) (2025)
- [GaussianDWM: 3D Gaussian Driving World Model for Unified Scene Understanding and Multi-Modal Generation](https://mlsanity.com/paper/2512.23180/) (2025)
- [FedLPA: One-shot Federated Learning with Layer-Wise Posterior Aggregation](https://mlsanity.com/paper/2310.00339/) (2023)
- [Mitigating the Risk of Health Inequity Exacerbated by Large Language Models](https://mlsanity.com/paper/2410.05180/) (2024)
- [GSQ-Tuning: Group-Shared Exponents Integer in Fully Quantized Training for LLMs On-Device Fine-tuning](https://mlsanity.com/paper/2502.12913/) (2025)
- [SILMM: Self-Improving Large Multimodal Models for Compositional Text-to-Image Generation](https://mlsanity.com/paper/2412.05818/) (2024)
- [Learn from Global Correlations: Enhancing Evolutionary Algorithm via Spectral GNN](https://mlsanity.com/paper/2412.17629/) (2024)
- [ChatMotion: A Multimodal Multi-Agent for Human Motion Analysis](https://mlsanity.com/paper/2502.18180/) (2025)
- [DynaWeb: Model-Based Reinforcement Learning of Web Agents](https://mlsanity.com/paper/2601.22149/) (2026)
- [CausalSR: Structural Causal Model-Driven Super-Resolution with Counterfactual Inference](https://mlsanity.com/paper/2501.15852/) (2025)
- [Are Small Language Models Ready to Compete with Large Language Models for Practical Applications?](https://mlsanity.com/paper/2406.11402/) (2024)
- [Contextual Bandits for Unbounded Context Distributions](https://mlsanity.com/paper/2408.09655/) (2024)

## Built on by (most cited first, 15 of 15)

- [When Safe Skills Collide: Measuring Compositional Risk in Agent Skill Ecosystems](https://mlsanity.com/paper/2606.00448/) (2026)
- [Sustainable Hybrid Document-Routed Retrieval for Financial RAG: Resolving the Robustness-Precision Trade-off](https://mlsanity.com/paper/2603.26815/) (2026)
- [Bridging Reasoning Trajectories in On-Policy Distillation via Near-Future Guidance](https://mlsanity.com/paper/2606.00305/) (2026)
- [Cornerstones or Stumbling Blocks? Deciphering the Rock Tokens in On-Policy Distillation](https://mlsanity.com/paper/2605.09253/) (2026)
- [The Efficiency Frontier: A Unified Framework for Cost-Performance Optimization in LLM Context Management](https://mlsanity.com/paper/2605.23071/) (2026)
- [MUSE: Resolving Manifold Misalignment in Visual Tokenization via Topological Orthogonality](https://mlsanity.com/paper/2605.05646/) (2026)
- [Interpretable vs Learned Encoders for High-Cardinality Fraud Detection](https://mlsanity.com/paper/2607.00477/) (2026)
- [Playing DOOM with 1.3M Parameters: Specialized Small Models vs Large Language Models for Real-Time Game Control](https://mlsanity.com/paper/2604.07385/) (2026)
- [Parametric Memory Decoding for Zero-Shot Routing in LoRA-Based External Parametric Memory](https://mlsanity.com/paper/2607.04118/) (2026)
- [SynGR: Unleashing the Potential of Cross-Modal Synergy for Generative Recommendation](https://mlsanity.com/paper/2605.18920/) (2026)
- [Agora: Toward Autonomous Bug Detection in Production-Level Consensus Protocols with LLM Agents](https://mlsanity.com/paper/2605.29910/) (2026)
- [Energy-Efficient On-Device RAG on a Mobile NPU: System Design and Benchmark on Snapdragon X Elite](https://mlsanity.com/paper/2606.11257/) (2026)
- [OCC-RAG: Optimal Cognitive Core for Faithful Question Answering](https://mlsanity.com/paper/2606.00683/) (2026)
- [Treatment Response Optimized Clinical Decision Support AI System via Digital Twin Simulation](https://mlsanity.com/paper/2606.17405/) (2026)
- [MARS: Multi-Agent Re-ranking for Repeat-Order Food Delivery Recommendation](https://mlsanity.com/paper/2607.25420/) (2026)

## Abstract (arXiv)

Large Language Models achieve remarkable performance but incur substantial computational costs unsuitable for resource-constrained deployments. This paper presents the first comprehensive task-specific efficiency analysis comparing 16 language models across five diverse NLP tasks. We introduce the Performance-Efficiency Ratio (PER), a novel metric integrating accuracy, throughput, memory, and latency through geometric mean normalization. Our systematic evaluation reveals that small models (0.5--3B parameters) achieve superior PER scores across all given tasks. These findings establish quantitative foundations for deploying small models in production environments prioritizing inference efficiency over marginal accuracy gains.