LoRA SFT

Light fine-tuning method used in E2 to reduce mismatch dr.

Neighborhood — ranked by edge-count

concept

Fine-tuning
implements
Parameter updates that reduce mismatch dr; another anchoring variant in UCCT.
Target Pattern Cohesion (ρd)
associated_with
Measures how tightly the target pattern PT clusters in representation space; one of three components of S

method

E2: Numeral-Base Arithmetic Controlled Study
uses
Quantitative study varying representational familiarity via numeral bases B10/B8/B9 at fixed computational complexity

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.

fine-tuning (SFT)method0.717
Supervised fine-tuning to adapt model parameters.
LoRA+CoTmethod0.704
Fine-tuning with chain-of-thought rationales aiming to reduce dr via procedural alignment.
LoRA (Low-Rank Adaptation)method0.698
Parameter-efficient fine-tuning method used for both SDF and expert iteration stages.
LoRA Fine-Tuning with Axolotlmethod0.684
Specific fine-tuning implementation using LoRA rank 32, learning rate 2e-4, AdamW 8-bit optimizer
Low-Rank Adaptation (LoRA)method0.677
Parameter-efficient fine-tuning method used to implement SOO fine-tuning on LLMs
SFR-DeepResearchframework0.665
The paper's core contribution: an RL-based framework for training autonomous single-agent LLMs to perform deep research with web search, browsing, and code execution.