11
1
Ellora: Enhancing LLMs with LoRA Standardized Recipes for Capability Enhancement (huggingface.co)
3
The 1B Token Challenge: Finding the Perfect Pre-Training Mix (huggingface.co)
2
The 1B Token Challenge: Finding the Perfect Pre-Training Mix (huggingface.co)
2
Pivotal Token Search (PTS): Targeting Critical Decision Points in LLM Training (huggingface.co)
1
Internal Coherence Maximization(ICM): Label-Free Unsupervised Training Framework (github.com/codelion)
1
Unsupervised Model Improvement via Internal Coherence Maximization (huggingface.co)
1
Show HN: PTS Library – Analyze LLM reasoning through "thought anchors"
1
Automated Discovery of High-Performance GPU Kernels with OpenEvolve (huggingface.co)
1
Adaptive Classifier: Dynamic Text Classification with Continuous Learning (huggingface.co)
1
Show HN: DeepThink Plugin – Bring Gemini 2.5's parallel reasoning to open models
1
Eliciting Fine-Tuned Transformer Capabilities via Inference-Time Techniques (arxiv.org)
20
Show HN: System Prompt Learning – LLMs Learn Problem-Solving from Experience
112
Show HN: AutoThink – Boosts local LLM performance with adaptive reasoning
2
Show HN: OpenEvolve – open-source implementation of DeepMind's AlphaEvolve
1
Pivotal Token Search (PTS): Identify critical decision points in LLM generations (github.com/codelion)
2
Show HN: An adaptive classifier that detects hallucinations in LLM/RAG outputs (github.com/codelion)
2
Show HN: OptiLLMBench – Test how inference optimization tricks scale up LLMs
1
Adaptive Classification for Automatic LLM Temperature Optimization
4
Show HN: A classifier that learns new categories without retraining from scratch
1
Adaptive Classifier: Dynamic Text Classification with Continuous Learning (github.com/codelion)
6
Show HN: Adaptive-classifier – text classification with continuous learning
1
Entropy Decoding in Optillm and Early Results on GSM8k
1
Unbounded Context with Memory
3
Optillm: An Optimizing Inference Proxy with Plugins (github.com/codelion)
1
Discover LLM-assisted workflow opportunities in your code (patched.codes)
1
Build your own AutoFix with Patchwork (patched.codes)
1
Cot Decoding – Eliciting Reasoning from LLMs
2
Evaluating code-to-readme generation using LLMs (patched.codes)
1
A comparative study of fine-tuning GPT-4o-mini, Gemini Flash 1.5, Llama-3.1-8B (patched.codes)
1
Beating GPT-4 on Arena Bench Hard with Patched MOA (patched.codes)
1
Patched RTC: evaluating LLMs for diverse software development tasks (patched.codes)
2
Working at Heights Safety: Video Analytics Revolution in Construction (securade.ai)
1
AI in Emergency Response: Improving Situational Awareness and Decision-Making (securade.ai)
1
Advancing Workplace Safety: AI Video Analytics for Accident Reduction (securade.ai)
3