Long Lian
@LongTonyLian
Followers
387
Following
383
Media
11
Statuses
111
EECS PhD student at @berkeley_ai. Research interests: developing LLMs/VLMs with reasoning capabilities through RL.
UC Berkeley
Joined July 2022
Excited to share that Describe Anything has been accepted at ICCV 2025! 🎉. Describe Anything Model (DAM) is a powerful Multimodal LLM that generates detailed descriptions for user-specified regions in images or videos using points, boxes, scribbles, or masks. Open-source code,.
Nvidia just dropped Describe Anything on Hugging Face. Detailed Localized Image and Video Captioning
2
23
107
RT @giffmana: Gemini 2.5 paper TL;DR. Technical part in thread. Contributors: ~1k.2.5 Pro timed out counting after 600s.2.5 Flash counts 1….
0
69
0
RT @Xinyu2ML: 🚀 Super excited to share Multiverse!. 🏃 It’s been a long journey exploring the space between model design and hardware effici….
0
18
0
RT @baifeng_shi: Finally! We just released the models and code for PS3 & VILA-HD, a vision encoder **pre-trained at 4K resolution** and the….
0
27
0
RT @_dmchan: 🚀 Call for Papers! 🚀.Excited to help organize the 4th Workshop on What is Next in Multimodal Foundation Models? at ICCV in Hon….
0
6
0
RT @YifeiZhou02: With previous research in multimodal and agents, I believe the only truly useful multimodal agent before 2027 is multimoda….
0
15
0
As we all know, collecting data for robotics is very costly. This is why I’m very impressed by this work: it generates a huge amount of data for different robots without any teleoperation.
Tired of teleoperating your robots?.We built a way to scale robot datasets without teleop, dynamic simulation, or even robot hardware. Just one smartphone scan + one human hand demo video → thousands of diverse robot trajectories. Trainable by diffusion policy and VLA models
0
2
18
RT @arthurallshire: our new system trains humanoid robots using data from cell phone videos, enabling skills such as climbing stairs and si….
0
112
0
Thanks for sharing our work!.
Adaptive Parallel Reasoning (APR) released on Hugging Face paper pages!. Language models can now orchestrate both serialized and parallel computations, with a new end-to-end reinforcement learning strategy.
0
1
4
RT @xiuyu_l: Scale smarter, not harder!. Long CoT reasoning is powerful, but its sequential nature limits how efficiently and easily it can….
0
21
0
Thank you for your appreciation in our work! 感谢分享我们的工作！. 我们相信解决高难度的问题不能只依赖单线程CoT，而是需要不同的线程分工合作，就像攻克高难度的研究问题往往需要一个团队一样。期待和大家多交流！.
「Reasoning, Agent」论文. Learning Adaptive Parallel Reasoning with Language Models. 当 prompt 成了 Launch Kernel ？APR 让 LLM 学会何时分裂多线程、何时回收串行，使推理摆脱线性束缚。. 为什么要“并行推理”？.串行 CoT：一步一步写思路 -> 长 token 序列既拖慢推理，又挤爆 context。
0
1
3
RT @dongxi_nlp: 「Reasoning, Agent」论文. Learning Adaptive Parallel Reasoning with Language Models. 当 prompt 成了 Launch Kernel ？APR 让 LLM 学会何时分….
0
21
0
RT @gm8xx8: Learning Adaptive Parallel Reasoning with Language Models. APR:.- mixes serialized & parallel CoT via spawn() / join().- traine….
0
6
0
RT @YifeiZhou02: It’s a really fun project to be involved in. It’s like giving the LLM the tool to call itself in a recursive manner, and i….
0
7
0
RT @junyi42: Introducing St4RTrack!🖖. Simultaneous 4D Reconstruction and Tracking in the world coordinate feed-forwardly, just by changing….
0
51
0
Solving complex problems often requires a team of brilliant minds working in parallel—collaborating, communicating, and delivering exceptional results. The same principle applies to reasoning in LLMs. Our approach, Adaptive Parallel Reasoning, replaces the traditional, linear.
We explore a new dimension in scaling reasoning models in Adaptive Parallel Reasoning. APR lets LMs learn to orchestrate both serial & parallel compute E2E via supervised training + RL — w/ better efficiency and scalability than long CoT on Countdown. 🧵
0
4
24
RT @jiayi_pirate: We explore a new dimension in scaling reasoning models in Adaptive Parallel Reasoning. APR lets LMs learn to orchestrate….
0
70
0
RT @iScienceLuvr: Learning Adaptive Parallel Reasoning with Language Models. "we propose Adaptive Parallel Reasoning (APR), a novel reasoni….
0
31
0
RT @arankomatsuzaki: Learning Adaptive Parallel Reasoning with Language Models. - Enables LMs to orchestrate both serialized and parallel c….
0
16
0