FUSE: Ensembling Verifiers with Zero Labeled Data

↘ Related papers

• Too Correct to Learn: Reinforcement Learning on Saturated Reasoning Data
  Zhenwen Liang34% match
• Train Separately, Merge Together: Modular Post-Training with Mixture-of-Experts
  Jacob Morrison32% match
• Randomly Initialized Networks Can Learn from Peer-to-Peer Consensus
  Esteban Rodríguez-Betancourt31% match
• GSQ: Highly-Accurate Low-Precision Scalar Quantization for LLMs via Gumbel-Softmax Sampling
  Alireza Dadgarnia29% match
• Latent Phase-Shift Rollback: Inference-Time Error Correction via Residual Stream Monitoring and KV-Cache Steering
  Manan Gupta29% match