8 papers
Lucy: edgerunning agentic web search on mobile with machine generated task vectors
Alan Dao, Dinh Bach Vu, Alex Nguyen +1
Small language models (SLMs) are inherently limited in knowledge-intensive tasks due to their constrained capacity. While test-time computation offers a path to enhanced performanc…
Jan-nano Technical Report
Alan Dao, Dinh Bach Vu
Most language models face a fundamental tradeoff where powerful capabilities require substantial computational resources. We shatter this constraint with Jan-nano, a 4B parameter l…
Speechless: Speech Instruction Training Without Speech for Low Resource Languages
Alan Dao, Dinh Bach Vu, Huy Hoang Ha +6
The rapid growth of voice assistants powered by large language models (LLM) has highlighted a need for speech instruction data to train these systems. Despite the abundance of spee…
ReZero: Enhancing LLM search ability by trying one-more-time
Alan Dao, Thinh Le
Retrieval-Augmented Generation (RAG) improves Large Language Model (LLM) performance on knowledge-intensive tasks but depends heavily on initial search query quality. Current metho…
VoxRep: Enhancing 3D Spatial Understanding in 2D Vision-Language Models via Voxel Representation
Alan Dao, Norapat Buppodom
Comprehending 3D environments is vital for intelligent systems in domains like robotics and autonomous navigation. Voxel grids offer a structured representation of 3D space, but ex…
AlphaSpace: Enabling Robotic Actions through Semantic Tokenization and Symbolic Reasoning
Alan Dao, Dinh Bach Vu, Bui Quang Huy
This paper presents AlphaSpace, a novel methodology designed to enhance the spatial reasoning capabilities of language models for robotic manipulation in 3D Cartesian space. AlphaS…