Discrete Diffusion Models Exploit Asymmetry to Solve Lookahead Planning Tasks
cs.LG
Submitted: 2026-02-23
Updated: 2026-10-30
Code: https://github.com/facebookresearch/flow
Terminology
Sources
- Accelerated Sampling from Masked Diffusion Models via Entropy Bounded Unmasking
- Language Models are Few-Shot Learners
- Generative Flows on Discrete State-Spaces: Enabling Multimodal Flows with Applications to Protein Co-Design
- Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality
- BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
- Set Block Decoding is a Language Model Inference Accelerator
- Train for the Worst, Plan for the Best: Understanding Token Ordering in Masked Diffusions
- Flow Matching Guide and Code
- Large Language Diffusion Models
- Deep sequence models tend to memorize geometrically; it is unclear why
- Beyond Autoregression: Discrete Diffusion for Complex Reasoning and Planning
- Dream 7B: Diffusion Large Language Models
Related papers
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis
- (Mis)generalization of Helpful-Only Fine-tuning