9 papers
TacitVLA: Reasoning Vision-Language-Action Models Without Reasoning Latency
Jack Lu, Yurong You, Boyi Li +8
Textual reasoning before acting has improved vision-language-action (VLA) models, but at the cost of latency: every action conditioned on the reasoning text must wait for autoregre…
HyperThink: Text-to-Parameter Hypernetworks for Efficient Reasoning
Donggyun Kim, Jack Lu, Chanwoo Kim +2
Long-form thinking traces can substantially improve the multi-step reasoning performance of large language models (LLMs), but they introduce high inference-time overhead, with late…
OpenGameEval: Benchmarking Agentic Programming and Exploration in a Stateful Game Engine
Eray Turkel, Mengsha Sun, Kartik Ayyar +6
We present OpenGameEval, a benchmark and evaluation framework for agentic game development inside Roblox Studio. It runs language models as agents in reproducible, stateful game-en…
Solaris: Building a Multiplayer Video World Model in Minecraft
Georgy Savva, Oscar Michel, Daohan Lu +6
Existing action-conditioned video generation models (video world models) are limited to single-agent perspectives, failing to capture the multi-agent interactions of real-world env…
SkillFactory: Self-Distillation For Learning Cognitive Behaviors
Zayne Sprague, Jack Lu, Manya Wadhwa +3
Reasoning models leveraging long chains of thought employ various cognitive skills, such as verification of their answers, backtracking, retrying by an alternate method, and more.…
When Does Verification Pay Off? A Closer Look at LLMs as Solution Verifiers
Jack Lu, Ryan Teehan, Jinran Jin +1
Large language models (LLMs) can act as both problem solvers and solution verifiers, where the latter select high-quality answers from a pool of solver-generated candidates. This r…
EEG2TEXT-CN: An Exploratory Study of Open-Vocabulary Chinese Text-EEG Alignment via Large Language Model and Contrastive Learning on ChineseEEG
Jacky Tai-Yu Lu, Jung Chiang, Chi-Sheng Chen +3
We propose EEG2TEXT-CN, which, to the best of our knowledge, represents one of the earliest open-vocabulary EEG-to-text generation frameworks tailored for Chinese. Built on a biolo…
Context Tuning for In-Context Optimization
Jack Lu, Ryan Teehan, Zhenbang Yang +1
We introduce Context Tuning, a simple and effective method to significantly enhance few-shot adaptation of large language models (LLMs) without weight updates. In-Context Learning…
ProCreate, Don't Reproduce! Propulsive Energy Diffusion for Creative Generation
Jack Lu, Ryan Teehan, Mengye Ren
In this paper, we propose ProCreate, a simple and easy-to-implement method to improve sample diversity and creativity of diffusion-based image generative models and to prevent trai…