nithin

nithin12342

AI & ML interests

Aspiring Software Developer | 🌐 Full-Stack Explorer | 🤖 AI Engineer-in-Progress Passionate about building scalable apps, automating workflows, and craft

Recent Activity

updated a collection about 4 hours ago

Efficient ai

reacted to mike-ravkine's post with 🔥 2 days ago

I went into this expecting to find a ~Q2 garbage dumpster but https://huggingface.co/prism-ml/Ternary-Bonsai-27B-gguf is a slick feat of QAT engineering. It is weaker then FP16 on a handful of tasks where quants usually degrade, as per their own paper the loss is "concentrated on sustained chains of reasoning / agentic" and in ReasonScape this bites on Sort, Shuffle and Dates, but counter-acting this are some noticeable improvements to thinking length without accuracy loss on several other tasks (Shapes, Cars). I haven't had a chance to run the Binary yet, but PQ2 + Bonsai QAT are confirmed to be pretty darn impressive.

upvoted a paper 3 days ago

CoLT: Teaching Multi-Modal Models to Think with Chain of Latent Thoughts

View all activity

Organizations

upvoted a paper 3 days ago

CoLT: Teaching Multi-Modal Models to Think with Chain of Latent Thoughts

Paper • 2606.31986 • Published 27 days ago • 1

upvoted a paper 6 days ago

RESOURCE2SKILL: Distilling Executable Agent Skills from Human-Created Multimodal Resources

Paper • 2606.29538 • Published 11 days ago • 139

upvoted a paper 13 days ago

MANCE: Manifold Aware Concept Erasure

Paper • 2607.03973 • Published 23 days ago • 49

upvoted a paper 19 days ago

Program-as-Weights: A Programming Paradigm for Fuzzy Functions

Paper • 2607.02512 • Published 25 days ago • 146

upvoted a paper 27 days ago

Training Open Models for Agentic Phone Use

Paper • 2606.23049 • Published Jun 22 • 17

upvoted 3 papers 28 days ago

SproutRAG: Attention-Guided Tree Search with Progressive Embeddings for Long-Document RAG

Paper • 2606.18381 • Published Jun 16 • 19

Holistic Data Scheduler for LLM Pre-training via Multi-Objective Reinforcement Learning

Paper • 2606.24133 • Published Jun 23 • 11

Moebius: 0.2B Lightweight Image Inpainting Framework with 10B-Level Performance

Paper • 2606.19195 • Published Jun 17 • 141

upvoted a paper about 1 month ago

SpatialClaw: Rethinking Action Interface for Agentic Spatial Reasoning

Paper • 2606.13673 • Published Jun 11 • 111

upvoted an article about 2 months ago

Article

LoRA Fine-Tuning BitNet b1.58 LLMs on Heterogeneous Edge GPUs via QVAC Fabric

qvac

•

Mar 17

• 19

upvoted 2 papers about 2 months ago

LaRA: Layer-wise Representation Analysis for Detecting Data Contamination in RL Post-Training

Paper • 2605.29888 • Published May 28 • 34

Unified Neural Scaling Laws

Paper • 2605.26248 • Published May 25 • 7

upvoted a paper 2 months ago

Anti-Self-Distillation for Reasoning RL via Pointwise Mutual Information

Paper • 2605.11609 • Published May 12 • 196

upvoted a collection 2 months ago

🧬 Carbon

Collection

Carbon 500M, 3B, 8B genomic models and GGUF variants for llama.cpp • 7 items • Updated Jun 2 • 44

upvoted 2 papers 2 months ago

Code as Agent Harness

Paper • 2605.18747 • Published May 18 • 225

Recursive Multi-Agent Systems

Paper • 2604.25917 • Published Apr 28 • 287

upvoted a paper 3 months ago

Stabilizing Efficient Reasoning with Step-Level Advantage Selection

Paper • 2604.24003 • Published Apr 27 • 9

upvoted a collection 3 months ago

MiMo-V2.5

Collection

4 items • Updated Apr 27 • 90

upvoted an article 3 months ago

Article

ML Intern Takes Our Post-Training Internship Test

cmpatino

•

Apr 23

• 31

upvoted a collection 3 months ago

LatentUM

Collection

6 items • Updated Apr 3 • 1

nithin

AI & ML interests

Recent Activity

Organizations

nithin12342's activity

**LoRA Fine-Tuning BitNet b1.58 LLMs on Heterogeneous Edge GPUs via QVAC Fabric**

ML Intern Takes Our Post-Training Internship Test

LoRA Fine-Tuning BitNet b1.58 LLMs on Heterogeneous Edge GPUs via QVAC Fabric