Demasking Fine-Tuning for Knowledge Injection in Autoregressive and Diffusion LLMs
cs.CL, cs.AI
Submitted: 2025-10-10
Updated: 2026-11-05
Code: https://github.com/huggingface/math-verify
Terminology
Sources
- The Reversal Curse: LLMs trained on "A is B" fail to learn "B is A"
- Webscale-RL: Automated Data Pipeline for Scaling RL Data to Pretraining Levels
- Everything is Editable: Extend Knowledge Editing to Unstructured Data in Large Language Models
- BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
- Reverse Training to Nurse the Reversal Curse
- The Llama 3 Herd of Models
- AnyEdit: Edit Any Knowledge Encoded in Language Models
- Scaling Laws for Neural Language Models
- On the generalization of language models from in-context learning and finetuning: a controlled study
- Lost in the Middle: How Language Models Use Long Contexts
- Injecting New Knowledge into Large Language Models via Supervised Fine-Tuning
- Diffusion Language Models are Super Data Learners
- Large Language Diffusion Models
- Fine-Tuning or Retrieval? Comparing Knowledge Injection in LLMs
- Knowledge-Instruct: Effective Continual Pre-training from Limited Data using Instructions
- Memorization and Knowledge Injection in Gated LLMs
- Diffusion Beats Autoregressive in Data-Constrained Settings
- GPQA: A Graduate-Level Google-Proof Q&A Benchmark
- On the Theoretical Limitations of Embedding-Based Retrieval
- Any-Order GPT as Masked Diffusion Model: Decoupling Formulation and Architecture
Related papers
- ALARM: Audio-Language Alignment for Reasoning Models
- ThinkSum: Probabilistic reasoning over sets using large language models
- Coherence boosting: When your pretrained language model is not paying enough attention
- From Facts to Insights: A Persona-Driven Dual Memory Framework and Dataset for Role-Playing Agents
- Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization
- The Art of Asking: Multilingual Prompt Optimization for Synthetic Data