Reasoning on the Manifold: Bidirectional Consistency for Self-Verification in Diffusion Language Models
cs.LG, cs.AI
Submitted: 2026-04-17
Updated: 2026-11-05
Terminology
Sources
- Large Language Monkeys: Scaling Inference Compute with Repeated Sampling
- Universal Self-Consistency for Large Language Model Generation
- Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
- Training Verifiers to Solve Math Word Problems
- SafeRBench: Dissecting the Reasoning Safety of Large Language Models
- Measuring Mathematical Problem Solving With the MATH Dataset
- Large Language Models Cannot Self-Correct Reasoning Yet
- Loong: Synthesize Long Chain-of-Thoughts at Scale through Verifiers
- Don't Settle Too Early: Self-Reflective Remasking for Diffusion Language Models
- Language Models (Mostly) Know What They Know
- Training Language Models to Self-Correct via Reinforcement Learning
- A Survey on Diffusion Language Models
- Large Language Diffusion Models
- Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Score-Based Generative Modeling through Stochastic Differential Equations
- Self-Correction Bench: Uncovering and Addressing the Self-Correction Blind Spot in Large Language Models
- SPG: Sandwiched Policy Gradient for Masked Diffusion Language Models
- Self-Consistency Improves Chain of Thought Reasoning in Language Models
- Revolutionizing Reinforcement Learning Framework for Diffusion Large Language Models
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- 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