Cover Image

Massive Supervised Fine-tuning Experiments Reveal How Data, Layer, and Training Factors Shape LLM Alignment Quality

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Harada, Yuto, Yamauchi, Yusuke, Oda, Yusuke, Oseki, Yohei, Miyao, Yusuke, Takagi, Yu
Format:	Preprint
Published:	2025
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Is Structure Dependence Shaped for Efficient Communication?: A Case Study on Coordination
by: Kajikawa, Kohei, et al.
Published: (2024)

How a Bilingual LM Becomes Bilingual: Tracing Internal Representations with Sparse Autoencoders
by: Inaba, Tatsuro, et al.
Published: (2025)

Exclusive Unlearning
by: Sasaki, Mutsumi, et al.
Published: (2026)

Instability in Downstream Task Performance During LLM Pretraining
by: Nishida, Yuto, et al.
Published: (2025)

Do Self-Supervised Speech Models Exhibit the Critical Period Effects in Language Acquisition?
by: Koga, Yurie, et al.
Published: (2025)

The Imperfective Paradox in Large Language Models
by: Ma, Bolei, et al.
Published: (2026)

Analyzing Correlations Between Intrinsic and Extrinsic Bias Metrics of Static Word Embeddings With Their Measuring Biases Aligned
by: Katô, Taisei, et al.
Published: (2024)

A Comparative Analysis of LLM Memorization at Statistical and Internal Levels: Cross-Model Commonalities and Model-Specific Signatures
by: Chen, Bowen, et al.
Published: (2026)

Does it Chug? Towards a Data-Driven Understanding of Guitar Tone Description
by: Sutar, Pratik, et al.
Published: (2024)

An Empirical Study of LLM-as-a-Judge: How Design Choices Impact Evaluation Reliability
by: Yamauchi, Yusuke, et al.
Published: (2025)

Do LLMs Need to Think in One Language? Correlation between Latent Language and Task Performance
by: Ozaki, Shintaro, et al.
Published: (2025)

A Multi-Perspective Analysis of Memorization in Large Language Models
by: Chen, Bowen, et al.
Published: (2024)

A Statistical and Multi-Perspective Revisiting of the Membership Inference Attack in Large Language Models
by: Chen, Bowen, et al.
Published: (2024)

Self-Emotion Blended Dialogue Generation in Social Simulation Agents
by: Zhang, Qiang, et al.
Published: (2024)

Why Are Parsing Actions for Understanding Message Hierarchies Not Random?
by: Kato, Daichi, et al.
Published: (2025)

Exploring the Effect of Segmentation and Vocabulary Size on Speech Tokenization for Speech Language Models
by: Kando, Shunsuke, et al.
Published: (2025)

How Much Do Large Language Models Know about Human Motion? A Case Study in 3D Avatar Control
by: Li, Kunhang, et al.
Published: (2025)

Are Emotions Arranged in a Circle? Geometric Analysis of Emotion Representations via Hyperspherical Contrastive Learning
by: Yamauchi, Yusuke, et al.
Published: (2026)

Regulators on some abelian coverings of $\mathbb{P}^1$ minus $n+2$ points
by: Nemoto, Yusuke, et al.
Published: (2025)

llm-jp-modernbert: A ModernBERT Model Trained on a Large-Scale Japanese Corpus with Long Context Length
by: Sugiura, Issa, et al.
Published: (2025)

Flexible Mesopores in Nanoscrolls: Extraordinarily Large Alteration of Pore Sizes and Their Reversibility
by: Yusuke Asakura, et al.
Published: (2024)

Dual Alignment Between Language Model Layers and Human Sentence Processing
by: Kuribayashi, Tatsuki, et al.
Published: (2026)

Developmentally-plausible Working Memory Shapes a Critical Period for Language Acquisition
by: Mita, Masato, et al.
Published: (2025)

Tree-Planted Transformers: Unidirectional Transformer Language Models with Implicit Syntactic Supervision
by: Yoshida, Ryo, et al.
Published: (2024)

Tracking World States with Language Models: State-Based Evaluation Using Chess
by: Harang, Romain, et al.
Published: (2025)

Improving Unsupervised Constituency Parsing via Maximizing Semantic Information
by: Chen, Junjie, et al.
Published: (2024)

Human-Grounded Multimodal Benchmark with 900K-Scale Aggregated Student Response Distributions from Japan's National Assessment of Academic Ability
by: Takami, Kyosuke, et al.
Published: (2026)

Unsupervised Parsing by Searching for Frequent Word Sequences among Sentences with Equivalent Predicate-Argument Structures
by: Chen, Junjie, et al.
Published: (2024)

Textless Dependency Parsing by Labeled Sequence Prediction
by: Kando, Shunsuke, et al.
Published: (2024)

Surface Insights of Mesoporous Fragile Organic Materials Under Ultra‐Low‐Voltage Directed by Gemini Column
by: Yusuke Asakura, et al.
Published: (2024)

A Universal Approach Using Water‐Soluble Templates for Meso‐ and Macro‐Porous Organic Polymers
by: Yusuke Asakura, et al.
Published: (2025)

How to Make the Most of LLMs' Grammatical Knowledge for Acceptability Judgments
by: Ide, Yusuke, et al.
Published: (2024)

Factorization envelopes and enveloping vertex algebras
by: Nishinaka, Yusuke
Published: (2025)

Zipping the Thought: When and How Compressed Reasoning Data Works in LLM Post-Training
by: Matsutani, Kohsei, et al.
Published: (2026)

Engineering faster double-array Aho-Corasick automata
by: Kanda, Shunsuke, et al.
Published: (2022)

Samarium(II) Ate Species Combined With Aminoalcohols as Reductants for Hydrogenation of Unsaturated Organic Compounds
by: Yusuke Oda, et al.
Published: (2026)

When ‘Skilled Migrants’ Becomes an Expansive Category: Multi‐Layered Inequality Among Former International Students in Japan
by: Yusuke Mazumi
Published: (2026)

FLARE-SSM: Deep State Space Models with Influence-Balanced Loss for 72-Hour Solar Flare Prediction
by: Takagi, Yusuke, et al.
Published: (2025)

Laryngeal stimulation test to identify the optimal timing for deep tracheal extubation in children
by: Tatsuo Kajino, et al.
Published: (2024)

Composition, Attention, or Both?
by: Yoshida, Ryo, et al.
Published: (2022)