layer-selection

Tag

Cards List
#layer-selection

Deeper is Not Always Better: Mitigating the Alignment Tax via Confident Layer Decoding

Hugging Face Daily Papers · 2026-06-20 Cached

This paper introduces Confident Decoding, a training-free decoding strategy that dynamically selects the most reliable intermediate layer in LLMs using entropy-guided search, mitigating the alignment tax and improving reasoning performance on benchmarks like GPQA-Diamond and Omni-MATH with negligible overhead.

0 favorites 0 likes
#layer-selection

FoRA: Fisher-orthogonal Rank Adaptation for Parameter-Efficient Fine-Tuning

arXiv cs.CL · 2026-05-29 Cached

FoRA introduces a parameter-efficient fine-tuning method that selects task-informative layers via Fisher scores and trains LoRA down-projections on the Stiefel manifold, reducing parameters while preserving accuracy.

0 favorites 0 likes
#layer-selection

Automatic Layer Selection for Hallucination Detection

arXiv cs.AI · 2026-05-27 Cached

This paper proposes automatic layer selection for hallucination detection in LLMs and introduces First Effective Peak of Intrinsic Dimension (FEPoID), a training-free criterion that consistently identifies optimal intermediate layers, outperforming existing heuristics.

0 favorites 0 likes
#layer-selection

Uncovering the Latent Potential of Deep Intermediate Representations

arXiv cs.LG · 2026-05-25 Cached

This paper introduces LOES (Layer-wise Optimal Embedding Selection) and GeoReg (Geometric Regularization Loss), methods that select and fuse task-relevant intermediate layers from deep models to improve transfer learning performance, demonstrating consistent gains across architectures and modalities.

0 favorites 0 likes
#layer-selection

Multilingual Steering by Design: Multilingual Sparse Autoencoders and Principled Layer Selection

arXiv cs.CL · 2026-05-25 Cached

This paper introduces a principled approach to multilingual language steering using sparse autoencoders (SAEs) trained on multilingual data and a novel layer selection rule based on the intersection of multilingual alignment and language separability, evaluated on LLaMA-3.1-8B and Gemma-2-9B for machine translation and cross-lingual summarization.

0 favorites 0 likes
#layer-selection

Predicting Where Steering Vectors Succeed

arXiv cs.CL · 2026-04-20 Cached

This paper introduces the Linear Accessibility Profile (LAP), a diagnostic method using logit lens to predict steering vector effectiveness across model layers, achieving ρ=+0.86 to +0.91 correlation on 24 concept families across five models. The work provides a systematic framework to determine which layers and concepts are suitable for steering interventions, replacing ad-hoc trial-and-error approaches.

0 favorites 0 likes
#layer-selection

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

arXiv cs.CL · 2026-04-20 Cached

Aletheia introduces a gradient-guided layer selection method for efficient LoRA fine-tuning that identifies task-relevant transformer layers via lightweight gradient probes and applies adapters selectively, achieving 15-28% training speedup across 14 models while maintaining downstream performance on MMLU, GSM8K, and HumanEval benchmarks.

0 favorites 0 likes
← Back to home

Submit Feedback