| 1. | | Cross-Lingual Alignment Without Joint Training (arxiv.org) |
| 1 point by codelion 3 days ago | past | discuss |
|
| 2. | | Voice AI Benchmarks: a directory of voice-agent evaluations (voicebenchmarks.com) |
| 2 points by codelion 4 days ago | past | discuss |
|
| 3. | | Neural Canvas: a fruit fly connectome demo in the browser (huggingface.co) |
| 2 points by codelion 6 days ago | past | discuss |
|
| 4. | | Iris: Climbing to the Search Frontier (arxiv.org) |
| 1 point by codelion 6 days ago | past | discuss |
|
| 5. | | The Recursive Economy: AI Self-Improvement and Scarcity (recursive-economy.pages.dev) |
| 2 points by codelion 26 days ago | past |
|
| 6. | | Neural Drive, a SuperTuxKart World Model that runs in the browser (huggingface.co) |
| 2 points by codelion 56 days ago | past |
|
| 7. | | Coding agent for local models on a Mac (mlx-optiq.com) |
| 1 point by codelion 57 days ago | past |
|
| 8. | | Frame: Grounding LLM Vulnerability Detection with a Sound Separation-Logic Core (lambdasec.github.io) |
| 2 points by codelion 73 days ago | past |
|
| 9. | | Mlx-optiq: per-layer mixed-precision LLM quantization for Apple Silicon (mlx-optiq.com) |
| 2 points by codelion 3 months ago | past |
|
| 10. | | A 1B humanizer that matches human writing on an AI detector (mlx-optiq.com) |
| 3 points by codelion 3 months ago | past |
|
| 11. | | Zero-Dependency Programming (conjure.pages.dev) |
| 2 points by codelion 5 months ago | past |
|
| 12. | | Scaling Pedagogical Pre-Training: From Optimal Mixing to 10B Tokens (huggingface.co) |
| 2 points by codelion 6 months ago | past |
|
| 13. | | From HashHop to Memory-Augmented Language Models (huggingface.co) |
| 2 points by codelion 7 months ago | past |
|
| 14. | | The Optimal Architecture for Small Language Models (huggingface.co) |
| 1 point by codelion 8 months ago | past |
|
| 15. | | Enhancing LLMs with LoRA – Standardized Recipes for Capability Enhancement (huggingface.co) |
| 3 points by codelion 9 months ago | past |
|
| 16. | | OpenEvolve: Teaching LLMs to Discover Algorithms Through Evolution (algorithmicsuperintelligence.ai) |
| 53 points by codelion 9 months ago | past | 9 comments |
|
| 17. | | Ellora: Enhancing LLMs with LoRA Standardized Recipes for Capability Enhancement (huggingface.co) |
| 1 point by codelion 9 months ago | past |
|
| 18. | | The 1B Token Challenge: Finding the Perfect Pre-Training Mix (huggingface.co) |
| 6 points by codelion 10 months ago | past |
|
| 19. | | The 1B Token Challenge: Finding the Perfect Pre-Training Mix (huggingface.co) |
| 3 points by codelion 10 months ago | past |
|
| 20. | | Pivotal Token Search (PTS): Targeting Critical Decision Points in LLM Training (huggingface.co) |
| 2 points by codelion on Aug 18, 2025 | past | 1 comment |
|
| 21. | | Internal Coherence Maximization(ICM): Label-Free Unsupervised Training Framework (github.com/codelion) |
| 1 point by codelion on Aug 9, 2025 | past |
|
| 22. | | Unsupervised Model Improvement via Internal Coherence Maximization (huggingface.co) |
| 1 point by codelion on Aug 3, 2025 | past |
|
| 23. | | Show HN: PTS Library – Analyze LLM reasoning through "thought anchors" |
| 2 points by codelion on July 23, 2025 | past |
|
| 24. | | Automated Discovery of High-Performance GPU Kernels with OpenEvolve (huggingface.co) |
| 4 points by codelion on June 28, 2025 | past |
|
| 25. | | Adaptive Classifier: Dynamic Text Classification with Continuous Learning (huggingface.co) |
| 1 point by codelion on June 20, 2025 | past |
|
| 26. | | Show HN: DeepThink Plugin – Bring Gemini 2.5's parallel reasoning to open models |
| 4 points by codelion on June 19, 2025 | past |
|
| 27. | | Eliciting Fine-Tuned Transformer Capabilities via Inference-Time Techniques (arxiv.org) |
| 1 point by codelion on June 14, 2025 | past |
|
| 28. | | Show HN: System Prompt Learning – LLMs Learn Problem-Solving from Experience |
| 48 points by codelion on June 2, 2025 | past | 13 comments |
|
| 29. | | Show HN: AutoThink – Boosts local LLM performance with adaptive reasoning |
| 397 points by codelion on May 28, 2025 | past | 68 comments |
|
| 30. | | Show HN: OpenEvolve – open-source implementation of DeepMind's AlphaEvolve |
| 8 points by codelion on May 20, 2025 | past | 3 comments |
|
|
| More |
