# LMUnit: Fine-grained Evaluation with Natural Language Unit Tests

arXiv: https://arxiv.org/abs/2412.13091 · HTML: https://mlsanity.com/paper/2412.13091/

## Key facts

- Cited by 33 papers in the map (data through 14 Oct 2026).
- Builds on 56 of its 86 references that are in the map.
- First posted on arXiv 17 Dec 2024.

## Builds on (most cited first, 30 of 56)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment](https://mlsanity.com/paper/2303.16634/) (2023)
- [A General Language Assistant as a Laboratory for Alignment](https://mlsanity.com/paper/2112.00861/) (2021)
- [Tulu 3: Pushing Frontiers in Open Language Model Post-Training](https://mlsanity.com/paper/2411.15124/) (2024)
- [Beyond Accuracy: Behavioral Testing of NLP models with CheckList](https://mlsanity.com/paper/2005.04118/) (2020)
- [Length-Controlled AlpacaEval: A Simple Way to Debias Automatic Evaluators](https://mlsanity.com/paper/2404.04475/) (2024)
- [AlpacaFarm: A Simulation Framework for Methods that Learn from Human Feedback](https://mlsanity.com/paper/2305.14387/) (2023)
- [Can Large Language Models Be an Alternative to Human Evaluations?](https://mlsanity.com/paper/2305.01937/) (2023)
- [BARTScore: Evaluating Generated Text as Text Generation](https://mlsanity.com/paper/2106.11520/) (2021)
- [Prometheus: Inducing Fine-grained Evaluation Capability in Language Models](https://mlsanity.com/paper/2310.08491/) (2023)
- [From Generation to Judgment: Opportunities and Challenges of LLM-as-a-judge](https://mlsanity.com/paper/2411.16594/) (2024)
- [Generative Verifiers: Reward Modeling as Next-Token Prediction](https://mlsanity.com/paper/2408.15240/) (2024)
- [Is ChatGPT a Good NLG Evaluator? A Preliminary Study](https://mlsanity.com/paper/2303.04048/) (2023)
- [Large Language Models Are State-of-the-Art Evaluators of Translation Quality](https://mlsanity.com/paper/2302.14520/) (2023)
- [Fine-Grained Human Feedback Gives Better Rewards for Language Model Training](https://mlsanity.com/paper/2306.01693/) (2023)
- [CodeT: Code Generation with Generated Tests](https://mlsanity.com/paper/2207.10397/) (2022)
- [How NOT To Evaluate Your Dialogue System: An Empirical Study of Unsupervised Evaluation Metrics for Dialogue Response Generation](https://mlsanity.com/paper/1603.08023/) (2016)
- [Humans or LLMs as the Judge? A Study on Judgement Biases](https://mlsanity.com/paper/2402.10669/) (2024)
- [A Long Way to Go: Investigating Length Correlations in RLHF](https://mlsanity.com/paper/2310.03716/) (2023)
- [All That's 'Human' Is Not Gold: Evaluating Human Evaluation of Generated Text](https://mlsanity.com/paper/2107.00061/) (2021)
- [ARES: An Automated Evaluation Framework for Retrieval-Augmented Generation Systems](https://mlsanity.com/paper/2311.09476/) (2023)
- [Towards a Unified Multi-Dimensional Evaluator for Text Generation](https://mlsanity.com/paper/2210.07197/) (2022)
- [MiniCheck: Efficient Fact-Checking of LLMs on Grounding Documents](https://mlsanity.com/paper/2404.10774/) (2024)
- [WildBench: Benchmarking LLMs with Challenging Tasks from Real Users in the Wild](https://mlsanity.com/paper/2406.04770/) (2024)
- [FLASK: Fine-grained Language Model Evaluation based on Alignment Skill Sets](https://mlsanity.com/paper/2307.10928/) (2023)

## Built on by (most cited first, 30 of 33)

- [Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models](https://mlsanity.com/paper/2503.09567/) (2025)
- [Skywork-Reward-V2: Scaling Preference Data Curation via Human-AI Synergy](https://mlsanity.com/paper/2507.01352/) (2025)
- [Checklists Are Better Than Reward Models For Aligning Language Models](https://mlsanity.com/paper/2507.18624/) (2025)
- [A Survey of Frontiers in LLM Reasoning: Inference Scaling, Learning to Reason, and Agentic Systems](https://mlsanity.com/paper/2504.09037/) (2025)
- [Learning to Plan & Reason for Evaluation with Thinking-LLM-as-a-Judge](https://mlsanity.com/paper/2501.18099/) (2025)
- [Evaluating Judges as Evaluators: The JETTS Benchmark of LLM-as-Judges as Test-Time Scaling Evaluators](https://mlsanity.com/paper/2504.15253/) (2025)
- [Improving LLM-as-a-Judge Inference with the Judgment Distribution](https://mlsanity.com/paper/2503.03064/) (2025)
- [Does Context Matter? ContextualJudgeBench for Evaluating LLM-based Judges in Contextual Settings](https://mlsanity.com/paper/2503.15620/) (2025)
- [TD-EVAL: Revisiting Task-Oriented Dialogue Evaluation by Combining Turn-Level Precision with Dialogue-Level Comparisons](https://mlsanity.com/paper/2504.19982/) (2025)
- [IF-RewardBench: Benchmarking Judge Models for Instruction-Following Evaluation](https://mlsanity.com/paper/2603.04738/) (2026)
- [RLBFF: Binary Flexible Feedback to bridge between Human Feedback & Verifiable Rewards](https://mlsanity.com/paper/2509.21319/) (2025)
- [Bridging Human and LLM Judgments: Understanding and Narrowing the Gap](https://mlsanity.com/paper/2508.12792/) (2025)
- [Full-Stack Alignment: Co-Aligning AI and Institutions with Thick Models of Value](https://mlsanity.com/paper/2512.03399/) (2025)
- [Foundational Automatic Evaluators: Scaling Multi-Task Generative Evaluator Training for Reasoning-Centric Domains](https://mlsanity.com/paper/2510.17793/) (2025)
- [Debatable Intelligence: Benchmarking LLM Judges via Debate Speech Evaluation](https://mlsanity.com/paper/2506.05062/) (2025)
- [Evalet: Evaluating Large Language Models through Functional Fragmentation](https://mlsanity.com/paper/2509.11206/) (2025)
- [On the Shelf Life of Fine-Tuned LLM-Judges: Future-Proofing, Backward-Compatibility, and Question Generalization](https://mlsanity.com/paper/2509.23542/) (2025)
- [Entropy Aware Reward Guidance for Diffusion Language Model Alignment](https://mlsanity.com/paper/2602.05000/) (2026)
- [Skill-RM: Unifying Heterogeneous Evaluation Criteria via Agent Skill](https://mlsanity.com/paper/2606.03980/) (2026)
- [AURA: A Diagnostic Framework for Tracking User Satisfaction of Interactive Planning Agents](https://mlsanity.com/paper/2505.01592/) (2025)
- [FormalRewardBench: A Benchmark for Formal Theorem Proving Reward Models](https://mlsanity.com/paper/2605.10141/) (2026)
- [Assessing the Pedagogical Readiness of Large Language Models as AI Tutors in Low-Resource Contexts: A Case Study of Nepal's K-10 Curriculum](https://mlsanity.com/paper/2604.09619/) (2026)
- [Support Vector Rubrics: Closing the Gap Between Self-Generated and Human Rubrics](https://mlsanity.com/paper/2606.08077/) (2026)
- [Graph-Structured Rubrics: Compiling Rubrics into Typed Evaluation Graphs for LLM Judges](https://mlsanity.com/paper/2608.12097/) (2026)
- [Grounded Checklist Partial Credit for Agent Skill Trajectories](https://mlsanity.com/paper/2608.27487/) (2026)
- [MutaGReP: Execution-Free Repository-Grounded Plan Search for Code-Use](https://mlsanity.com/paper/2502.15872/) (2025)
- [The AI Consumer Index (ACE)](https://mlsanity.com/paper/2512.04921/) (2025)
- [Stable Reasoning, Unstable Responses: Mitigating LLM Deception via Stability Asymmetry](https://mlsanity.com/paper/2603.26846/) (2026)
- [From Feelings to Metrics: Understanding and Formalizing How Users Vibe-Test LLMs](https://mlsanity.com/paper/2604.14137/) (2026)
- [Decoupled Smart Contract Audits: Lightweight LLM Framework via Distillation and Aggregation](https://mlsanity.com/paper/2606.03128/) (2026)

## Abstract (arXiv)

As language models become integral to critical workflows, assessing their behavior remains a fundamental challenge -- human evaluation is costly and noisy, while automated metrics provide only coarse, difficult-to-interpret signals. We introduce natural language unit tests, a paradigm that decomposes response quality into explicit, testable criteria, along with a unified scoring model, LMUnit, which combines multi-objective training across preferences, direct ratings, and natural language rationales. Through controlled human studies, we show this paradigm significantly improves inter-annotator agreement and enables more effective LLM development workflows. LMUnit achieves state-of-the-art performance on evaluation benchmarks (FLASK, BigGenBench) and competitive results on RewardBench. These results validate both our proposed paradigm and scoring model, suggesting a promising path forward for language model evaluation and development.