# Accuracy is Not All You Need

arXiv: https://arxiv.org/abs/2407.09141 · HTML: https://mlsanity.com/paper/2407.09141/

## Key facts

- Cited by 31 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 42 references that are in the map.
- First posted on arXiv 12 Jul 2024.

## Builds on (most cited first, 27 of 27)

- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [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)
- [Pointer Sentinel Mixture Models](https://mlsanity.com/paper/1609.07843/) (2016)
- [TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension](https://mlsanity.com/paper/1705.03551/) (2017)
- [PIQA: Reasoning about Physical Commonsense in Natural Language](https://mlsanity.com/paper/1911.11641/) (2019)
- [HuggingFace's Transformers: State-of-the-art Natural Language Processing](https://mlsanity.com/paper/1910.03771/) (2019)
- [GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers](https://mlsanity.com/paper/2210.17323/) (2022)
- [The LAMBADA dataset: Word prediction requiring a broad discourse context](https://mlsanity.com/paper/1606.06031/) (2016)
- [Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formatting](https://mlsanity.com/paper/2310.11324/) (2023)
- [Yi: Open Foundation Models by 01.AI](https://mlsanity.com/paper/2403.04652/) (2024)
- [Large Language Models Are Not Robust Multiple Choice Selectors](https://mlsanity.com/paper/2309.03882/) (2023)
- [Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs](https://mlsanity.com/paper/2310.01801/) (2023)
- [SliceGPT: Compress Large Language Models by Deleting Rows and Columns](https://mlsanity.com/paper/2401.15024/) (2024)
- [The Unreasonable Ineffectiveness of the Deeper Layers](https://mlsanity.com/paper/2403.17887/) (2024)
- [What Do Compressed Deep Neural Networks Forget?](https://mlsanity.com/paper/1911.05248/) (2019)
- [Characterising Bias in Compressed Models](https://mlsanity.com/paper/2010.03058/) (2020)
- [When Benchmarks are Targets: Revealing the Sensitivity of Large Language Model Leaderboards](https://mlsanity.com/paper/2402.01781/) (2024)
- [Functional Benchmarks for Robust Evaluation of Reasoning Performance, and the Reasoning Gap](https://mlsanity.com/paper/2402.19450/) (2024)
- [Compressing LLMs: The Truth is Rarely Pure and Never Simple](https://mlsanity.com/paper/2310.01382/) (2023)
- [Decoding Compressed Trust: Scrutinizing the Trustworthiness of Efficient LLMs Under Compression](https://mlsanity.com/paper/2403.15447/) (2024)
- [Changing Answer Order Can Decrease MMLU Accuracy](https://mlsanity.com/paper/2406.19470/) (2024)
- [Beyond Preserved Accuracy: Evaluating Loyalty and Robustness of BERT Compression](https://mlsanity.com/paper/2109.03228/) (2021)
- [Going Beyond Classification Accuracy Metrics in Model Compression](https://mlsanity.com/paper/2012.01604/) (2020)
- [The Cost of Compression: Investigating the Impact of Compression on Parametric Knowledge in Language Models](https://mlsanity.com/paper/2312.00960/) (2023)

## Built on by (most cited first, 30 of 31)

- [70% Size, 100% Accuracy: Lossless LLM Compression for Efficient GPU Inference via Dynamic-Length Float (DFloat11)](https://mlsanity.com/paper/2504.11651/) (2025)
- [INT v.s. FP: A Comprehensive Study of Fine-Grained Low-bit Quantization Formats](https://mlsanity.com/paper/2510.25602/) (2025)
- [Exploring the Trade-Offs: Quantization Methods, Task Difficulty, and Model Size in Large Language Models From Edge to Giant](https://mlsanity.com/paper/2409.11055/) (2024)
- [When LLMs get significantly worse: A statistical approach to detect model degradations](https://mlsanity.com/paper/2602.10144/) (2026)
- [SINQ: Sinkhorn-Normalized Quantization for Calibration-Free Low-Precision LLM Weights](https://mlsanity.com/paper/2509.22944/) (2025)
- [Quantization Undoes Alignment: Bias Emergence in Compressed LLMs Across Models and Precision Levels](https://mlsanity.com/paper/2605.15208/) (2026)
- [A Proximal Operator for Inducing 2:4-Sparsity](https://mlsanity.com/paper/2501.18015/) (2025)
- [PASER: Post-Training Data Selection for Efficient Pruned Large Language Model Recovery](https://mlsanity.com/paper/2502.12594/) (2025)
- [The Illusion of Equivalency: Statistical Characterization of Quantization Effects in LLMs](https://mlsanity.com/paper/2607.08734/) (2026)
- [Weight Pruning Amplifies Bias: A Multi-Method Study of Compressed LLMs for Edge AI](https://mlsanity.com/paper/2605.08137/) (2026)
- [Total Variation Distance Estimation in Autoregressive Models](https://mlsanity.com/paper/2607.19510/) (2026)
- [Certifying Compressed Language Models: An Audit and a Statistical Toolkit](https://mlsanity.com/paper/2608.15046/) (2026)
- [Broken on Arrival: Silently Defective LLM Artifacts in Public Model Registries and How to Catch Them](https://mlsanity.com/paper/2609.05881/) (2026)
- [Eigenvectors of Experts are Training-free Non-collapsing Routers](https://mlsanity.com/paper/2605.30992/) (2026)
- [Variable Bit-width Quantization: Learning Per-Group Precision for "Bigger-but-Smaller" Language Models](https://mlsanity.com/paper/2607.02893/) (2026)
- [Which Decisions Low-Bit Quantization Breaks, and How to Predict Them](https://mlsanity.com/paper/2608.06564/) (2026)
- [QUASAR: Lowering the Loss Floor of Quantization-Aware Training with Loss-Aware Reconstruction](https://mlsanity.com/paper/2608.13966/) (2026)
- [Accuracy is Not Enough: A Divergence-Based Approach to Evaluate Fidelity Loss in Quantized LLMs](https://mlsanity.com/paper/2609.07664/) (2026)
- [Cassandra: Enabling Reasoning LLMs at Edge via Self-Speculative Decoding](https://mlsanity.com/paper/2605.26558/) (2026)
- [Reducing the GPU Memory Bottleneck with Lossless Compression for ML -- Extended](https://mlsanity.com/paper/2605.30728/) (2026)
- [Reliability Scaling Laws for Quantized Large Language Models](https://mlsanity.com/paper/2607.10855/) (2026)
- [HindsightBench: A Black-Box Behavioral Audit Protocol for Parametric Hindsight in Time-Indexed LLM Decision Tasks](https://mlsanity.com/paper/2607.18867/) (2026)
- [Tied Trit-Planes: Constraining PTQTP to a Uniform Nine-Level Quantizer, with a Persistent Folded Format for Disk-Streamed Mixture-of-Experts Serving](https://mlsanity.com/paper/2608.08910/) (2026)
- [Same Request, Different Answer: Quantization Amplifies Cache-Induced Divergence in LLM Serving](https://mlsanity.com/paper/2609.04748/) (2026)
- [Permutation-Based Stegomalware in Large Language Models: Threats and Countermeasures](https://mlsanity.com/paper/2609.16193/) (2026)
- [A Calibrated Instrument for Measuring How Inference Optimizations Affect Output Quality](https://mlsanity.com/paper/2609.18005/) (2026)
- [The Undetected Damage of Quantization on Retrieval and How to Fix It](https://mlsanity.com/paper/2609.24322/) (2026)
- [G2PTQ: Improving LLM Post-Training Quantization with Generalized Gradient Compensation](https://mlsanity.com/paper/2609.31009/) (2026)
- [Saying, Not Knowing: Aggressively GGUF-Quantized Small Language Models Still Write Rare Words They Can No Longer Define](https://mlsanity.com/paper/2610.04403/) (2026)
- [What Does a Harness Repair? A Preregistered Study of Visibility, Baseline Adequacy and Evaluation Defects](https://mlsanity.com/paper/2610.05533/) (2026)

## Abstract (arXiv)

When Large Language Models (LLMs) are compressed using techniques such as quantization, the predominant way to demonstrate the validity of such techniques is by measuring the model's accuracy on various benchmarks.If the accuracies of the baseline model and the compressed model are close, it is assumed that there was negligible degradation in quality.However, even when the accuracy of baseline and compressed model are similar, we observe the phenomenon of flips, wherein answers change from correct to incorrect and vice versa in proportion.We conduct a detailed study of metrics across multiple compression techniques, models and datasets, demonstrating that the behavior of compressed models as visible to end-users is often significantly different from the baseline model, even when accuracy is similar.We further evaluate compressed models qualitatively and quantitatively using MT-Bench and show that compressed models are significantly worse than baseline models in this free-form generative task.Thus, we argue that compression techniques should also be evaluated using distance metrics.We propose two such metrics, KL-Divergence and flips, and show that they are well correlated.