repoGitHubTrust 82 · PrimaryPublished 1mo agoLive · 1mo ago
sunrainyg/RandOpt
Official Codebase for "Neural Thickets: Diverse Task Experts Are Dense Around Pretrained Weights" (ICML 2026 Spotlight)
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 46%DeepSWE: new benchmark looking at how well today's frontier models can actually write code [R] →
- PossiblePossibly related (embedding) · 46%Understanding Large Language Models →
- PossiblePossibly related (embedding) · 45%Program-as-Weights: A Programming Paradigm for Fuzzy Functions →
- PossiblePossibly related (embedding) · 45%deepseek-ai/DeepSeek-V3 →
- PossiblePossibly related (embedding) · 46%Systematic Evaluation of Learning Rate Scheduling Strategies Across Heterogeneous Architectures →
- PossiblePossibly related (embedding) · 50%Similarity-Guided Curriculum Fine-Tuning of LLMs for Neural Architecture Synthesis →
Covers
Implements
Related to
Implements (incoming)
Related across the graph
modeldeepseek-ai/DeepSeek-V3paperSimilarity-Guided Curriculum Fine-Tuning of LLMs for Neural Architecture SynthesispaperProgram-as-Weights: A Programming Paradigm for Fuzzy FunctionspaperUnderstanding Large Language ModelspaperSystematic Evaluation of Learning Rate Scheduling Strategies Across Heterogeneous ArchitecturesnewsDeepSWE: new benchmark looking at how well today's frontier models can actually write code [R]