# Aletheia: Gradient-Guided Layer Selection for Efficient LoRA Fine-Tuning Across Architectures

arXiv: https://arxiv.org/abs/2604.15351 · HTML: https://mlsanity.com/paper/2604.15351/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 13 references that are in the map.
- First posted on arXiv 4 Apr 2026.

## Builds on (most cited first, 8 of 8)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [QLoRA: Efficient Finetuning of Quantized LLMs](https://mlsanity.com/paper/2305.14314/) (2023)
- [DoRA: Weight-Decomposed Low-Rank Adaptation](https://mlsanity.com/paper/2402.09353/) (2024)
- [Reducing Transformer Depth on Demand with Structured Dropout](https://mlsanity.com/paper/1909.11556/) (2019)
- [The Truth is in There: Improving Reasoning in Language Models with Layer-Selective Rank Reduction](https://mlsanity.com/paper/2312.13558/) (2023)

## Built on by (most cited first, 3 of 3)

- [When Gradient Importance Lies: Adaptive LoRA Rank Allocation Fails Under GRPO](https://mlsanity.com/paper/2605.07366/) (2026)
- [AQLoRA: A Zero-Search Recipe for Fast Quantized LoRA Fine-Tuning](https://mlsanity.com/paper/2608.23816/) (2026)
- [Compiling Learning Problems into Adaptation Programs for Language Models](https://mlsanity.com/paper/2609.37371/) (2026)

## Abstract (arXiv)

Low-Rank Adaptation (LoRA) has become the dominant parameter-efficient fine-tuning method for large language models, yet standard practice applies LoRA adapters uniformly to all transformer layers regardless of their relevance to the downstream task. We introduce Aletheia, a gradient-guided layer selection method that identifies the most task-relevant layers via a lightweight gradient probe and applies LoRA adapters only to those layers with asymmetric rank allocation. Across 81 experiment rows covering 14 successful models from 8 architecture families (0.5B-72B parameters, including dense and Mixture-of-Experts architectures), with one additional documented failed Pythia/GPT-NeoX attempt in Campaign 2, Aletheia achieves a 15-28% training speedup (mean 23.1%, p < 0.001) with bounded extra forgetting and broadly matched downstream behavior on the evaluated MMLU, GSM8K, and HumanEval benchmark pack. Across the tested families and scales, Campaign 1 shows a 100% per-model speed win rate and Campaign 2 shows broadly preserved downstream behavior within a bounded-degradation framing. Together these results support a practical model-economics claim: intelligent layer selection can make LoRA fine-tuning materially more efficient without introducing major downstream damage on the evaluated set.