Add new papers: AIConfigurator and MoEBlaze

[AmberLJC]

Adds two new papers from Jan 12th:

• AIConfigurator: Lightning-Fast Configuration Optimization for Multi-Framework LLM Serving (added to LLM serving section)
• MoEBlaze: Breaking the Memory Wall for Efficient MoE Training on Modern GPUs (added to Pre-training section)

Resolves #49

Generated with Claude Code

---

Note

Updates the curated list with two recent papers.

• Adds MoEBlaze to Training > Pre-training
• Adds AIConfigurator to Serving > LLM serving

^{Written by Cursor Bugbot for commit ed4c98b. This will update automatically on new commits. Configure here.}

[claude]

Code Review for PR #50: Add new papers (AIConfigurator and MoEBlaze)

Summary

This PR adds two recent papers from January 2026 to the curated list. Overall, the changes are well-formatted and follow repository conventions. ✅

---

Positive Observations

1. Correct Formatting: Both entries follow the repository's formatting guidelines exactly:

   • Proper markdown link format: [Title](URL): Description
   • ArXiv URLs without www. prefix ✅
   • Consistent structure matching existing entries

2. Appropriate Section Placement:

   • MoEBlaze → Pre-training section (line 84) - Correct, as it focuses on MoE training systems
   • AIConfigurator → LLM serving section (line 241) - Correct, as it focuses on serving configuration optimization

3. No Duplicates: Verified no duplicate entries exist in the repository

4. Descriptions: Both papers have clear, concise descriptions that accurately reflect their contributions

5. Recent Research: Both papers are from January 2026 (arXiv IDs: 2601.05296 and 2601.06288), keeping the list up-to-date

---

Minor Suggestions for Improvement

While not blockers, consider these enhancements for consistency with other entries:

1. Venue/Organization Metadata (Optional):

   • Current: - [MoEBlaze](https://arxiv.org/abs/2601.05296): Breaking the Memory Wall for Efficient MoE Training on Modern GPUs
   • Consider adding organization/affiliation if available: ... | Organization Name
   • Example: - [AIConfigurator](https://arxiv.org/abs/2601.06288): Lightning-Fast Configuration Optimization for Multi-Framework LLM Serving | [Organization]
   • Note: Per CLAUDE.md line 75, venue/organization info should be included "when available"

2. Chronological Ordering:

   • Both papers appear to be added at the end of their respective sections, which is appropriate for recent papers
   • Confirmed placement follows the chronological pattern observed in other sections

---

Quality Assessment

Criterion	Status	Notes
Code Quality	✅ Pass	Markdown formatting is clean and consistent
Conventions	✅ Pass	Follows CLAUDE.md guidelines exactly
No Bugs	✅ Pass	No broken links or formatting issues
Security	✅ N/A	Content changes only, no security concerns
Test Coverage	✅ N/A	No tests needed for content updates
Performance	✅ N/A	No performance impact

---

Recommendation

APPROVE ✅

This PR is ready to merge. The additions are well-formatted, appropriately categorized, and add value to the curated list by including recent systems research on MoE training and LLM serving optimization.

---

References to Repository Guidelines

Per CLAUDE.md:

• Line 67: Paper entry format followed correctly ✅
• Line 71: ArXiv links without www. prefix ✅
• Line 102-111: Section placement is appropriate ✅
• Line 124: Chronological order maintained ✅

Great work on maintaining consistency with the repository standards!