Cover Image

Matching Features, Not Tokens: Energy-Based Fine-Tuning of Language Models

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Jelassi, Samy, Kwun, Mujin, Zhao, Rosie, Li, Yuanzhi, Fusi, Nicolo, Du, Yilun, Kakade, Sham M., Domingo-Enrich, Carles
Format:	Preprint
Published:	2026
Subjects:	Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

The Recurrent Transformer: Greater Effective Depth and Efficient Decoding
by: Oncescu, Costin-Andrei, et al.
Published: (2026)

Echo Chamber: RL Post-training Amplifies Behaviors Learned in Pretraining
by: Zhao, Rosie, et al.
Published: (2025)

LoRA Soups: Merging LoRAs for Practical Skill Composition Tasks
by: Prabhakar, Akshara, et al.
Published: (2024)

Repeat After Me: Transformers are Better than State Space Models at Copying
by: Jelassi, Samy, et al.
Published: (2024)

Characterization and Mitigation of Training Instabilities in Microscaling Formats
by: Su, Huangyuan, et al.
Published: (2025)

SOAP: Improving and Stabilizing Shampoo using Adam
by: Vyas, Nikhil, et al.
Published: (2024)

Q-Probe: A Lightweight Approach to Reward Maximization for Language Models
by: Li, Kenneth, et al.
Published: (2024)

How Does Overparameterization Affect Features?
by: Duzgun, Ahmet Cagri, et al.
Published: (2024)

Universal Length Generalization with Turing Programs
by: Hou, Kaiying, et al.
Published: (2024)

Any-Order Flexible Length Masked Diffusion
by: Kim, Jaeyeon, et al.
Published: (2025)

The Role of Sparsity for Length Generalization in Transformers
by: Golowich, Noah, et al.
Published: (2025)

Adjoint Matching through the Lens of the Stochastic Maximum Principle in Optimal Control
by: Domingo-Enrich, Carles, et al.
Published: (2026)

LOTION: Smoothing the Optimization Landscape for Quantized Training
by: Kwun, Mujin, et al.
Published: (2025)

A Taxonomy of Loss Functions for Stochastic Optimal Control
by: Domingo-Enrich, Carles
Published: (2024)

Adjoint Matching: Fine-tuning Flow and Diffusion Generative Models with Memoryless Stochastic Optimal Control
by: Domingo-Enrich, Carles, et al.
Published: (2024)

Deconstructing What Makes a Good Optimizer for Language Models
by: Zhao, Rosie, et al.
Published: (2024)

A unified perspective on fine-tuning and sampling with diffusion and flow models
by: Domingo-Enrich, Carles, et al.
Published: (2026)

Reinforce Adjoint Matching: Scaling RL Post-Training of Diffusion and Flow-Matching Models
by: Bergmeister, Andreas, et al.
Published: (2026)

Feature emergence via margin maximization: case studies in algebraic tasks
by: Morwani, Depen, et al.
Published: (2023)

Mixture of Parrots: Experts improve memorization more than reasoning
by: Jelassi, Samy, et al.
Published: (2024)

Adapting Language Models via Token Translation
by: Feng, Zhili, et al.
Published: (2024)

Stochastic Optimal Control Matching
by: Domingo-Enrich, Carles, et al.
Published: (2023)

Equilibrium Matching: Generative Modeling with Implicit Energy-Based Models
by: Wang, Runqian, et al.
Published: (2025)

Selective Underfitting in Diffusion Models
by: Song, Kiwhan, et al.
Published: (2025)

Cheap Permutation Testing
by: Domingo-Enrich, Carles, et al.
Published: (2025)

Compress Then Test: Powerful Kernel Testing in Near-linear Time
by: Domingo-Enrich, Carles, et al.
Published: (2023)

Value Gradient Guidance for Flow Matching Alignment
by: Liu, Zhen, et al.
Published: (2025)

Random Scaling of Emergent Capabilities
by: Zhao, Rosie, et al.
Published: (2025)

Self-Improving Language Models with Bidirectional Evolutionary Search
by: Xu, Guowei, et al.
Published: (2026)

Collective Model Intelligence Requires Compatible Specialization
by: Pari, Jyothish, et al.
Published: (2024)

Beyond Implicit Bias: The Insignificance of SGD Noise in Online Learning
by: Vyas, Nikhil, et al.
Published: (2023)

Prescriptive Scaling Reveals the Evolution of Language Model Capabilities
by: Zhang, Hanlin, et al.
Published: (2026)

IGDA: Interactive Graph Discovery through Large Language Model Agents
by: Havrilla, Alex, et al.
Published: (2025)

EvoLM: In Search of Lost Language Model Training Dynamics
by: Qi, Zhenting, et al.
Published: (2025)

Discovering Hierarchical Latent Capabilities of Language Models via Causal Representation Learning
by: Jin, Jikai, et al.
Published: (2025)

Rare Event Analysis via Stochastic Optimal Control
by: Du, Yuanqi, et al.
Published: (2026)

Train for the Worst, Plan for the Best: Understanding Token Ordering in Masked Diffusions
by: Kim, Jaeyeon, et al.
Published: (2025)

Matching the Statistical Query Lower Bound for $k$-Sparse Parity Problems with Sign Stochastic Gradient Descent
by: Kou, Yiwen, et al.
Published: (2024)

Cognitive models can reveal interpretable value trade-offs in language models
by: Murthy, Sonia K., et al.
Published: (2025)

Peer-Predictive Self-Training for Language Model Reasoning
by: Feng, Shi, et al.
Published: (2026)