Workshop
How Far Are We From AGI
Jiaxuan You 路 Ge Liu 路 Yunzhu Li 路 Song Han 路 Dawn Song
Halle A 7
Fri 10 May, 11:30 p.m. PDT
The dream of Artificial General Intelligence (AGI) has been the north star of AI research since its inception. While large language models like GPT-4 and LLama-2 signal promising advances, the road to true AGI remains complex, mixed with both promise and challenges. Our workshop offers a collaborative platform to discuss AGI's potential, its historical roots, and pressing challenges, blending insights from both academia and industry. Adopting a hybrid format, we invite global researchers to join, share, and shape the discourse on our journey towards AGI.We invite submissions on a range of topics including, but not limited to:- Frontiers of AGI research- Classic AGI Attempts as Inspiration- Interdisciplinary Insights for AGI- Fundamental Limitations of LLMs- Practical Limitations of LLMs- Safety, Ethics, and Regulation in AGI Development- AGI's Economic and Societal Impacts
Chat is not available.
Timezone: America/Los_Angeles
Schedule
|
Fri 11:30 p.m. - 11:45 p.m.
|
Opening remarks: Jiaxuan You
(
Talk by organizer
)
>
SlidesLive Video |
Jiaxuan You 馃敆 |
|
Fri 11:45 p.m. - 12:45 a.m.
|
Oral presentations from submissions
(
Oral presentation
)
>
SlidesLive Video |
馃敆 |
|
Sat 12:45 a.m. - 1:00 a.m.
|
Invited talk: UK Government AI Safety Institute
(
Invited talk
)
>
SlidesLive Video |
馃敆 |
|
Sat 1:00 a.m. - 1:30 a.m.
|
Keynote talk: Oriol Vinyals
(
Keynote talk
)
>
SlidesLive Video |
Oriol Vinyals 馃敆 |
|
Sat 1:30 a.m. - 1:45 a.m.
|
Coffee break and social
|
馃敆 |
|
Sat 1:45 a.m. - 2:30 a.m.
|
Keynote talk: Yejin Choi
(
Keynote talk
)
>
SlidesLive Video |
Yejin Choi 馃敆 |
|
Sat 2:30 a.m. - 3:00 a.m.
|
Keynote talk: Andrew Gordon Wilson
(
Keynote talk
)
>
SlidesLive Video |
Andrew G Wilson 馃敆 |
|
Sat 3:00 a.m. - 3:30 a.m.
|
Keynote talk: Song Han
(
Keynote talk
)
>
SlidesLive Video |
Song Han 馃敆 |
|
Sat 3:30 a.m. - 4:30 a.m.
|
Lunch break and social
|
馃敆 |
|
Sat 4:30 a.m. - 5:30 a.m.
|
Keynote talk: Yoshua Bengio
(
Keynote talk
)
>
SlidesLive Video |
Yoshua Bengio 馃敆 |
|
Sat 5:35 a.m. - 7:00 a.m.
|
Panel discussions: How far are we from AGI?
(
Panel discussions
)
>
SlidesLive Video |
Yoshua Bengio 路 Yejin Choi 路 Andrew G Wilson 路 Song Han 路 Dawn Song 路 Jiaxuan You 路 Ge Liu 馃敆 |
|
Sat 7:00 a.m. - 8:00 a.m.
|
Poster session and social
(
Poster session
)
>
|
馃敆 |
|
-
|
The Pitfalls of Next-Token Prediction ( Oral ) > link | Gregor Bachmann 路 Vaishnavh Nagarajan 馃敆 |
|
-
|
Neural Retrievers are Biased Towards LLM-Generated Content ( Poster ) > link | Sunhao Dai 路 Yuqi Zhou 路 Liang Pang 路 Weihao Liu 路 Xiaolin Hu 路 Yong Liu 路 Xiao Zhang 路 Gang Wang 路 Jun Xu 馃敆 |
|
-
|
Thought-Retriever: Don鈥檛 Just Retrieve Raw Data, Retrieve Thoughts ( Poster ) > link | Tao Feng 路 Pengrui Han 路 Guanyu Lin 路 Ge Liu 路 Jiaxuan You 馃敆 |
|
-
|
Understanding the Reasoning Ability of Language Models From the Perspective of Reasoning Paths Aggregation ( Oral ) > link | Xinyi Wang 路 Alfonso Amayuelas 路 Kexun Zhang 路 Liangming Pan 路 Wenhu Chen 路 William Wang 馃敆 |
|
-
|
Rethinking harmless refusals when fine-tuning foundation models ( Poster ) > link | Florin Pop 路 Judd Rosenblatt 路 Diogo de Lucena 路 Michael Vaiana 馃敆 |
|
-
|
Towards Human-like Machine Vision: Representing Part-Whole Relationship with hierarchically correlated neuronal activation in neural networks ( Poster ) > link | Hao Zheng 路 Hui Lin 路 Rong Zhao 路 Huanyu Qu 馃敆 |
|
-
|
CatCode: A Comprehensive Evaluation Framework for LLMs On the Mixture of Code and Text ( Poster ) > link | Zhenru Lin 路 Yiqun Yao 路 Yang Yuan 馃敆 |
|
-
|
An Embodied Generalist Agent in 3D World ( Poster ) > link | Jiangyong Huang 路 Silong Yong 路 Xiaojian Ma 路 Xiongkun Linghu 路 Puhao Li 路 Yan Wang 路 Qing Li 路 Song-Chun Zhu 路 Baoxiong Jia 路 Siyuan Huang 馃敆 |
|
-
|
Do LLM Agents Have Regret? A Case Study in Online Learning and Games ( Oral ) > link | Chanwoo Park 路 Xiangyu Liu 路 Asuman Ozdaglar 路 Kaiqing Zhang 馃敆 |
|
-
|
GPT Is Becoming a Turing Machine: Here Are Some Ways to Program It ( Poster ) > link | Ana Jojic 路 Zhen Wang 路 Nebojsa Jojic 馃敆 |
|
-
|
Can Large Language Model Agents Simulate Human Trust Behaviors? ( Poster ) > link | Chengxing Xie 路 Canyu Chen 路 Feiran Jia 路 Yeol Ye 路 Kai Shu 路 Adel Bibi 路 Ziniu Hu 路 Philip Torr 路 Bernard Ghanem 路 Guohao Li 馃敆 |
|
-
|
OpenFedLLM: Training Large Language Models on Decentralized Private Data via Federated Learning ( Poster ) > link | Rui Ye 路 WenHao Wang 路 Jingyi Chai 路 Dihan Li 路 Zexi Li 路 Yinda Xu 路 Yaxin Du 路 Yanfeng Wang 路 Siheng Chen 馃敆 |
|
-
|
Probing the Multi-turn Planning Capabilities of LLMs via 20 Question Games ( Poster ) > link | Yizhe Zhang 路 Jiarui Lu 路 Navdeep Jaitly 馃敆 |
|
-
|
LEAGUE++: EMPOWERING CONTINUAL ROBOT LEARNING THROUGH GUIDED SKILL ACQUISITION WITH LARGE LANGUAGE MODELS ( Poster ) > link | Zhaoyi Li 路 Kelin Yu 路 Shuo Cheng 路 Danfei Xu 馃敆 |
|
-
|
CodeIt: Self-Improving Language Models with Prioritized Hindsight Replay ( Poster ) > link | Natasha Butt 路 Blazej Manczak 路 Auke Wiggers 路 Corrado Rainone 路 David Zhang 路 Micha毛l Defferrard 路 Taco Cohen 馃敆 |
|
-
|
AI Alignment with Changing and Influenceable Reward Functions ( Oral ) > link | Micah Carroll 路 Davis Foote 路 Anand Siththaranjan 路 Stuart Russell 路 Anca Dragan 馃敆 |
|
-
|
Exploring Iterative Enhancement for Improving Learnersourced Multiple-Choice Question Explanations with Large Language Models ( Poster ) > link | Qiming Bao 路 Juho Leinonen 路 Alex Peng 路 Wanjun Zhong 路 Ga毛l Gendron 路 Timothy Pistotti 路 Alice Huang 路 Paul Denny 路 Michael Witbrock 路 Jiamou Liu 馃敆 |
|
-
|
RET-LLM: Towards a General Read-Write Memory for Large Language Models ( Poster ) > link | Seyed MohammadAli Modarressi 路 Ayyoob Imani 路 Mohsen Fayyaz 路 Hinrich Schuetze 馃敆 |
|
-
|
Speak It Out: Solving Symbol-Related Problems with Symbol-to-Language Conversion for Language Models ( Poster ) > link | Yile Wang 路 Sijie Cheng 路 Zixin Sun 路 Peng Li 路 Yang Liu 馃敆 |
|
-
|
SceneCraft: An LLM Agent for Synthesizing 3D Scene as Blender Code ( Oral ) > link | Ziniu Hu 馃敆 |
|
-
|
Incentivizing Inclusive Data Contributions in Personalized Federated Learning ( Poster ) > link | Enpei Zhang 路 Jingyi Chai 路 Rui Ye 路 Yanfeng Wang 路 Siheng Chen 馃敆 |
|
-
|
How to benchmark AGI: the Adversarial Game ( Poster ) > link | Emmanouil Seferis 馃敆 |
|
-
|
RD2Bench: Toward Data-Centric Automatic R&D ( Poster ) > link | Haotian Chen 路 Xinjie Shen 路 Zeqi Ye 路 Xiao Yang 路 Xu Yang 路 Weiqing Liu 路 Jiang Bian 馃敆 |
|
-
|
Large Language Models Are Not Strong Abstract Reasoners Yet ( Poster ) > link | Ga毛l Gendron 路 Qiming Bao 路 Michael Witbrock 路 Gillian Dobbie 馃敆 |
|
-
|
Collaborative LLM-Agents for Editable Driving Scene Simulation ( Poster ) > link | Yuxi Wei 路 Zi Wang 路 Yifan Lu 路 Chenxin Xu 路 Changxing Liu 路 Hao Zhao 路 Siheng Chen 路 Yanfeng Wang 馃敆 |
|
-
|
DEFT: FLASH TREE-ATTENTION WITH IO-AWARENESS FOR EFFICIENT TREE-SEARCH-BASED LLM INFERENCE ( Oral ) > link | Jinwei Yao 路 Kexun Zhang 路 Kaiqi Chen 路 Jiaxuan You 路 Zeke Wang 路 Binhang Yuan 路 Tao Lin 馃敆 |
|
-
|
FightLadder: A Benchmark for Competitive Multi-Agent Reinforcement Learning ( Poster ) > link | Wenzhe Li 路 Zihan Ding 路 Seth Karten 路 Chi Jin 馃敆 |
|
-
|
On the Shape of Brainscores for Large Language Models (LLMs) ( Oral ) > link | Jingkai Li 馃敆 |
|
-
|
MARS: A Benchmark for Multi-LLM Algorithmic Routing System ( Poster ) > link | Qitian Hu 路 Jacob Bieker 路 Xiuyu Li 路 Nan Jiang 路 Benjamin Keigwin 路 Gaurav Ranganath 路 Kurt Keutzer 路 Shriyash Upadhyay 馃敆 |
|
-
|
A Language Model's Guide Through Latent Space ( Oral ) > link | Dimitri von R眉tte 路 Sotiris Anagnostidis 路 Gregor Bachmann 路 Thomas Hofmann 馃敆 |
|
-
|
How Far Are We from Intelligent Visual Deductive Reasoning? ( Poster ) > link | Yizhe Zhang 路 He Bai 路 Ruixiang Zhang 路 Jiatao Gu 路 Shuangfei Zhai 路 Joshua Susskind 路 Navdeep Jaitly 馃敆 |
|
-
|
Generative Representational Instruction Tuning ( Oral ) > link | Niklas Muennighoff 路 Hongjin SU 路 Liang Wang 路 Nan Yang 路 Furu Wei 路 Tao Yu 路 Amanpreet Singh 路 Douwe Kiela 馃敆 |
|
-
|
Efficient Causal Graph Discovery Using Large Language Models ( Poster ) > link | Thomas Jiralerspong 路 Xiaoyin Chen 路 Yash More 路 Vedant Shah 路 Yoshua Bengio 馃敆 |
|
-
|
Self-Alignment of Large Language Models via Monopolylogue-based Social Scene Simulation ( Oral ) > link | Xianghe Pang 路 Shuo Tang 路 Rui Ye 路 Yuxin Xiong 路 Bolun Zhang 路 Yanfeng Wang 路 Siheng Chen 馃敆 |