[NeurIPS 2024 Oral] Aligner: Efficient Alignment by Learning to Correct
-
Updated
Jan 16, 2025 - Python
#
weak-to-strong
Here are 8 public repositories matching this topic...
Weak-to-Strong Generalization via Direct On-Policy Distillation
-
Updated
Jul 14, 2026 - Python
[TPAMI] The official implementation of our paper "Improved and Accelerated Text-to-Image Generation with Collect, Reflect, and Refine".
-
Updated
Mar 8, 2026 - Python
Co-Supervised Learning: Improving Weak-to-Strong Generalization with Hierarchical Mixture of Experts
-
Updated
Feb 26, 2024 - Python
[VLDB'25] Official repo for Paper "Weak-to-Strong Prompts with Lightweight-to-Powerful LLMs for High-Accuracy, Low-Cost, and Explainable Data Transformation"
data-transformation code-generation data-cleaning self-reflection large-language-models weak-to-strong lazy-rag
-
Updated
Aug 21, 2025 - Python
Experiments for the Neural Interactive Proofs paper
machine-learning reinforcement-learning ai multi-agent-systems ai-safety weak-to-strong scalable-oversight
-
Updated
Jun 30, 2025 - Python
🔥🔥🔥 This repository curates research on Weak-to-Strong Generalization across LLMs, multimodal learning, and beyond, focusing on how strong models learn from weak supervision and surpass their teachers. Stay tuned for the latest updates!
-
Updated
Apr 28, 2026
Official code for "Student Guides Teacher: Weak-to-Strong Inference via Spectral Orthogonal Exploration" (SOE, ACL 2026 Oral). A training-free, test-time scaling method that fixes reasoning collapse in LLM mathematical reasoning via orthogonal probing.
nlp agent math gemma reasoning self-consistency aime llm chain-of-thought vllm qwen weak-to-strong test-time-compute test-time-scaling inference-time-scaling acl2026
-
Updated
Jul 7, 2026 - Python
Improve this page
Add a description, image, and links to the weak-to-strong topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the weak-to-strong topic, visit your repo's landing page and select "manage topics."