Fine-Tuning Diffusion Language Models with Context Selection and Target Weighting
cs.AI, cs.CL
Submitted: 2026-09-29
Updated: 2026-09-29
Comments: 30 pages, 4 figures, 14 tables. Main text 10 pages, references and appendix follow. Project page with interactive visualizations: https://loaym.github.io/GoldiMask/
Project page: https://loaym.github.io/GoldiMask
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Where-to-Unmask: Ground-Truth-Guided Unmasking Order Learning for Masked Diffusion Language Models
- Program Synthesis with Large Language Models
- Supportive Token Revealing for Fast Diffusion Language Model Decoding
- Self-Distilled Trajectory-Aware Boltzmann Modeling: Bridging the Training-Inference Discrepancy in Diffusion Language Models
- Evaluating Large Language Models Trained on Code
- ELF: Embedded Language Flows
- Accelerating Diffusion LLM Inference via Local Determinism Propagation
- DAWN: Dependency-Aware Fast Inference for Diffusion LLMs
- ORBSLAM3-Enhanced Autonomous Toy Drones: Pioneering Indoor Exploration
- Dream 7B: Diffusion Large Language Models
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems