Reasoning: From Reflection to Solution
Paper
• 2511.11712 • Published
• 2
Zixi "Oz" Li PRO
OzTianlu
AI & ML interests
My research focuses on deep reasoning with small language models, Transformer architecture innovation, and knowledge distillation for efficient alignment and transfer.
Recent Activity
upvoted a paper 1 day ago
DualPath: Breaking the Storage Bandwidth Bottleneck in Agentic LLM Inference reacted
to
their post with 🤗 1 day ago
🔥 UPGRADE in Kai: 30B Scaling! 🔥
https://huggingface.co/NoesisLab/Kai-30B-Instruct
https://huggingface.co/spaces/NoesisLab/Kai-30B-Instruct
We are incredibly excited to announce that the Kai-30B-Instruct model and its official Space are now LIVE! 🚀
If you've been following the journey from Kai-0.35B to Kai-3B, you know we're rethinking how models reason. Tired of verbose, slow Chain-of-Thought (CoT) outputs that flood your screen with self-talk? So are we.
Kai-30B-Instruct scales up our Adaptive Dual-Search Distillation (ADS) framework. By bridging classical A* heuristic search with continuous gradient descent , we use an information-theoretic log-barrier to physically prune high-entropy reasoning paths during training.
The result? Pure implicit reasoning. The model executes structured logic, arithmetic carries, and branch selections as a reflex in a single forward pass—no external scaffolding required.
At 3B, we observed a phase transition where the model achieved "logical crystallization". Now, at 30B, we are giving the ADS regularizer the massive representational capacity it needs to tackle higher-order symbolic abstractions and complex reasoning tasks.
🧪 Test Kai yourself in our new Space:
https://huggingface.co/spaces/NoesisLab/Kai-30B-Instruct
📦 Model Weights:
https://huggingface.co/NoesisLab/Kai-30B-Instruct
Bring your hardest math, logic, and coding benchmarks. We invite the community to stress-test the limits of the penalty wall! 🧱💥 posted an
update
1 day ago
🔥 UPGRADE in Kai: 30B Scaling! 🔥
https://huggingface.co/NoesisLab/Kai-30B-Instruct
https://huggingface.co/spaces/NoesisLab/Kai-30B-Instruct
We are incredibly excited to announce that the Kai-30B-Instruct model and its official Space are now LIVE! 🚀
If you've been following the journey from Kai-0.35B to Kai-3B, you know we're rethinking how models reason. Tired of verbose, slow Chain-of-Thought (CoT) outputs that flood your screen with self-talk? So are we.
Kai-30B-Instruct scales up our Adaptive Dual-Search Distillation (ADS) framework. By bridging classical A* heuristic search with continuous gradient descent , we use an information-theoretic log-barrier to physically prune high-entropy reasoning paths during training.
The result? Pure implicit reasoning. The model executes structured logic, arithmetic carries, and branch selections as a reflex in a single forward pass—no external scaffolding required.
At 3B, we observed a phase transition where the model achieved "logical crystallization". Now, at 30B, we are giving the ADS regularizer the massive representational capacity it needs to tackle higher-order symbolic abstractions and complex reasoning tasks.
🧪 Test Kai yourself in our new Space:
https://huggingface.co/spaces/NoesisLab/Kai-30B-Instruct
📦 Model Weights:
https://huggingface.co/NoesisLab/Kai-30B-Instruct
Bring your hardest math, logic, and coding benchmarks. We invite the community to stress-test the limits of the penalty wall! 🧱💥 Organizations
authored a paper about 1 month ago