LoRA adapter

Fine-tuning technique that can effectively add a constant vector to the residual stream, equivalent to steering along a persona direction

Neighborhood — ranked by edge-count

paper

cosine ≥ 0.65 · no typed edge

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

LoRA Adaptersmethod0.952
Parameter-efficient fine-tuning method used in both distillation and introspection stages
LoRA (Low-Rank Adaptation)method0.775
Parameter-efficient fine-tuning method used for both SDF and expert iteration stages.
Low-Rank Adaptation (LoRA)method0.754
Parameter-efficient fine-tuning method used to implement SOO fine-tuning on LLMs
LoRA Fine-Tuningmethod0.747
Adaptation method used via Tinker API for DeepSeek-V3.1 and Qwen3-235B fine-tuning with rank 32
LoRA+CoTmethod0.739
Fine-tuning with chain-of-thought rationales aiming to reduce dr via procedural alignment.
LoRA SFTmethod0.738
Light fine-tuning method used in E2 to reduce mismatch dr.
A LoRA adapter can learn to add a constant vector equivalent to steering along a pre-existing persona direction, making fine-tuning literally equivalent to activation steeringclaim0.732
Mechanistic explanation of how fine-tuning can shift persona vectors without directly updating activations
LoRA: Low-Rank Adaptation of Large Language Models (Hu et al., 2022)concept0.714
Fine-tuning method paper whose technique is used in the fine-tuning experiments