arXiv Machine Learning September 23, 2026 By Junyi Ye, Mengjia Yu, Debapriya Hazra, Guiling Wang

Damage Predicts Recovery: When Calibration Data Matters in Compressing Financial LLMs

Read the original on arXiv Machine Learning →

The paper investigates whether domain‑matched calibration data is necessary when compressing large language models for financial tasks. It finds that if compression causes little task damage, the choice of calibration corpus has minimal impact, whereas significant damage—especially from pruning—can be mitigated by using a finance‑specific calibration set (FinMix). The study tests this across multiple models, compression settings, and financial tasks, consistently supporting the link between damage and recovery.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

arXiv Machine Learning

Sep 2

The Structure of Quantization Damage in LLMs: Why the Next Bit Should Be Spent Globally

The paper investigates where post‑training quantization (PTQ) harms large language models (LLMs) and how to best allocate a limited precision budget. By causally raising each layer to 8‑bit precision across nine open‑weight models, the authors find that quantization damage is diffuse rather than concentrated in specific task circuits or weight statistics, and that globally refining quantization granularity outperforms selectively protecting the most recoverable layers. They also observe that the residual accuracy loss is budget‑limited and that peak recovery locations correlate with architecture within families but not across families.

By Jundong Hu, Shekar Ramachandran

llms efficiency

arXiv Machine Learning

Aug 10

Quantization Damage Is Multiplicative, Not Additive

arXiv:2608. 06564v1 Announce Type: new Abstract: Quantization is how large language models are actually deployed, and below four bits it is known to hurt.

By Zekun Wu, Swati Dhiman, Adriano Koshiyama

llms agents efficiency benchmarks safety

arXiv AI

Jun 3

How Quantization Changes Interpretable Features: A Sparse Autoencoder Analysis of Language Models

arXiv:2606. 03002v1 Announce Type: cross Abstract: Quantization is a standard path to deploying large language models, and a quantized model is typically judged acceptable when its perplexity or downstream accuracy stays close to the full-precision original.

By Evan Duan

llms efficiency safety

Hugging Face Trending Papers

Aug 5

The Calibration Floor: Format Repair Can Masquerade as Self-Correction at Small-to-Mid Scale

Accuracy changes after language-model self-revision are usually interpreted as changes in reasoning. We show this can fail at the answer-extraction boundary, and test the failure causally rather than only observationally.

arXiv AI

Aug 24

Target-Aware Calibration Data Selection for Preserving Uncertainty in Quantized Language Models

arXiv:2608.21019v1 Announce Type: cross Abstract: Quantization is widely used to deploy large language models, but its effect on uncertainty behavior, such as confidence, margins, and abstention, is...

By Zhen Yang, Sizai Hou, Kaiwen Zheng, Yaofang Liu, Liang He, Yixuan Chen, Kangning Cui

llms nlp efficiency benchmarks

Hugging Face Trending Papers

Sep 21

The Undetected Damage of Quantization on Retrieval and How to Fix It

We show that a quantized model that keeps its classification accuracy still changes $14$ to $46\%$ of its top-1 retrieval results, and that aggregate ranking metrics reveal only part of this damage. W...

efficiency