Dec. 10, 2024

NeurIPS, the top AI and ML conference, is being held in Vancouver from December 10 to December 15. Princeton Language and Intelligence is proud to feature the work of our students, post-docs, and faculty that will be showcased at the conference. 

Presented at NeurIPS 2024


The Road Less Scheduled (Oral)
 

poster

Authors: Aaron Defazio, Xingyu Yang, Ahmed Khaled, Konstantin Mishchenko, Harsh Mehta, Ashok Cutkosky

Links: Paper 


FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision (Spotlight)
 

poster

Authors: Jay Shah, Ganesh Bikshandi, Ying Zhang, Vijay Thakkar, Pradeep Ramani, Tri Dao

Links: Paper 


CryoBench: Datasets and Benchmarks for Heterogeneous Cryo-EM Reconstruction (Spotlight)

poster

Authors: Minkyu Jeon, Rishwanth Raghu, Miro Astore, Geoffrey Woollard, Ryan Feathers, Alkin Kaz, Sonya Hanson, Pilar Cossio, Ellen Zhong

Links: Paper 


RedPajama: an Open Dataset for Training Large Language Models (Spotlight)
 

poster

Authors: Maurice Weber, Dan Fu, Quentin Anthony, Yonatan Oren, Shane Adams, Anton Alexandrov, Xiaozhong Lyu, Huu Nguyen, Xiaozhe Yao, Virginia Adams, Ben Athiwaratkun, Rahul Chalamala, Kezhen Chen, Max Ryabinin, Tri Dao, Percy Liang, Christopher Ré, Irina Rish, Ce Zhang

Links: Paper 


GREATS: Online Selection of High-Quality Data for LLM Training in Every Iteration (Spotlight)
 

poster

Authors: Jiachen (Tianhao) Wang, Tong Wu, Dawn Song, Prateek Mittal, Ruoxi Jia

Links: Paper 


Finding Transformer Circuits With Edge Pruning (Spotlight)
 

poster

Authors: Adithya Bhaskar, Alexander Wettig, Dan Friedman, Danqi Chen

Links: Paper 


A Metalearned Neural Circuit for Nonparametric Bayesian Inference
 

poster

Authors: Jake Snell, Gianluca Bencomo, Tom Griffiths

Links: Paper 


Keeping LLMs Aligned After Fine-tuning: The Crucial Role of Prompt Templates
 

poster

Authors: Kaifeng Lyu, Haoyu Zhao, Xinran Gu, Dingli Yu, Anirudh Goyal, Sanjeev Arora

Links: Paper 


SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering
 

poster

Authors: John Yang, Carlos Jimenez, Alexander Wettig, Kilian Lieret, Shunyu Yao, Karthik Narasimhan, Ofir Press

Links: Paper 


Offline Multitask Representation Learning for Reinforcement Learning
 

poster

Authors: Haque Ishfaq, Thanh Nguyen-Tang, Songtao Feng, Raman Arora, Mengdi Wang, Ming Yin, Doina Precup

Links: Paper 


Fast Best-of-N Decoding via Speculative Rejection
 

poster

Authors: Hanshi Sun, Momin Haider, Ruiqi Zhang, Huitao Yang, Jiahao Qiu, Ming Yin, Mengdi Wang, Peter Bartlett, Andrea Zanette

Links: Paper 


Nonparametric Classification on Low Dimensional Manifolds using Overparameterized Convolutional Residual Networks
 

poster

Authors: Kaiqi Zhang, Zixuan Zhang, Minshuo Chen, Yuma Takeda, Mengdi Wang, Tuo Zhao, Yu-Xiang Wang

Links: Paper 


BitDelta: Your Fine-Tune May Only Be Worth One Bit
 

poster

Authors: James Liu, Guangxuan Xiao, Kai Li, Jason Lee, Song Han, Tri Dao, Tianle Cai

Links: Paper 


A Theoretical Perspective for Speculative Decoding Algorithm
 

poster

Authors: Ming Yin, Minshuo Chen, Kaixuan Huang, Mengdi Wang

Links: Paper 


Understanding the Limits of Vision Language Models Through the Lens of the Binding Problem
 

poster

Authors: Declan Campbell, Sunayana Rane, Tyler Giallanza, Camillo Nicolò De Sabbata, Kia Ghods, Amogh Joshi, Alexander Ku, Steven Frankland, Tom Griffiths, Jonathan D Cohen, Taylor Webb

Links: Paper 


Evaluating Copyright Takedown Methods for Language Models
 

poster

Authors: Boyi Wei, Weijia Shi, Yangsibo Huang, Noah Smith, Chiyuan Zhang, Luke Zettlemoyer, Kai Li, Peter Henderson

Links: Paper 


LoFiT: Localized Fine-tuning on LLM Representations
 

poster

Authors: Fangcong Yin, Xi Ye, Greg Durrett

Links: Paper 


Gradient Guidance for Diffusion Models: An Optimization Perspective
 

poster

Authors: Yingqing Guo, Hui Yuan, Yukang Yang, Minshuo Chen, Mengdi Wang

Links: Paper 


Inference via Interpolation: Contrastive Representations Provably Enable Planning and Inference
 

POSTER

Authors: Benjamin Eysenbach, Vivek Myers, Ruslan Salakhutdinov, Sergey Levine

Links: Paper 


SciCode: A Research Coding Benchmark Curated by Scientists
 

poster

Authors: Minyang Tian, Luyu Gao, Shizhuo Zhang, Xinan Chen, Cunwei Fan, Xuefei Guo, Roland Haas, Pan Ji, Kittithat Krongchon, Yao Li, Shengyan Liu, Di Luo, Yutao Ma, HAO TONG, Kha Trinh, Chenyu Tian, Zihan Wang, Bohao Wu, Shengzhu Yin, Minhui Zhu, Kilian Lieret, Yanxin Lu, Genglin Liu, Yufeng Du, Tianhua Tao, Ofir Press, Jamie Callan, Eliu Huerta, Hao Peng

Links: Paper 


Learning to Assist Humans without Inferring Rewards
 

poster

Authors: Vivek Myers, Evan Ellis, Sergey Levine, Benjamin Eysenbach, Anca Dragan

Links: Paper 


Are More LLM Calls All You Need? Towards the Scaling Properties of Compound AI Systems
 

poster

Authors: Lingjiao Chen, Jared Quincy Davis, Boris Hanin, Peter Bailis, Ion Stoica, Matei A Zaharia, James Zou

Links: Paper 


Metacognitive Capabilities of LLMs: An Exploration in Mathematical Problem Solving
 

poster

Authors: Aniket Didolkar, Anirudh Goyal, Nan Rosemary Ke, Siyuan Guo, Michal Valko, Timothy Lillicrap, Danilo Rezende, Yoshua Bengio, Michael Mozer, Sanjeev Arora

Links: Paper 


Learning and Transferring Sparse Contextual Bigrams with Linear Transformers
 

poster

Authors: Yunwei Ren, Zixuan Wang, Jason Lee

Links: Paper 


Regularizing Hidden States Enables Learning Generalizable Reward Model for LLMs
 

poster

Authors: Rui Yang, Ruomeng Ding, Yong Lin, Huan Zhang, Tong Zhang

Links: Paper 


Transfer Q-star : Principled Decoding for LLM Alignment
 

poster

Authors: Souradip Chakraborty, Soumya Suvra Ghosal, Ming Yin, Dinesh Manocha, Mengdi Wang, Amrit Singh Bedi, Furong Huang

Links: Paper 


Can Models Learn Skill Composition from Examples?
 

poster

Authors: Haoyu Zhao, Simran Kaur, Dingli Yu, Anirudh Goyal, Sanjeev Arora

Links: Paper 


Preference Learning Algorithms Do Not Learn Preference Rankings
 

poster

Authors: Angelica Chen, Sadhika Malladi, Lily Zhang, Xinyi Chen, Qiuyi (Richard) Zhang, Rajesh Ranganath, Kyunghyun Cho

Links: Paper 


Learning Human-like Representations to Enable Learning Human Values
 

poster

Authors: Andrea Wynn, Ilia Sucholutsky, Tom Griffiths

Links: Paper 


Hydra: Bidirectional State Space Models Through Generalized Matrix Mixers
 

poster

Authors: Sukjun Hwang, Aakash Sunil Lahoti, Ratish Puduppully, Tri Dao, Albert Gu

Links: Paper 


Stochastic Zeroth-Order Optimization under Strongly Convexity and Lipschitz Hessian: Minimax Sample Complexity
 

poster

Authors: Qian Yu, Yining Wang, Baihe Huang, Qi Lei, Jason Lee

Links: Paper 


MUVERA: Multi-Vector Retrieval via Fixed Dimensional Encoding
 

poster

Authors: Rajesh Jayaram, Laxman Dhulipala, Majid Hadian, Jason Lee, Vahab Mirrokni

Links: Paper 


Directional Smoothness and Gradient Methods: Convergence and Adaptivity
 

poster

Authors: Aaron Mishkin, Ahmed Khaled, Yuanhao Wang, Aaron Defazio, Robert Gower

Links: Paper 


ConceptMix: A Compositional Image Generation Benchmark with Controllable Difficulty
 

poster

Authors: Xindi Wu, Dingli Yu, Yangsibo Huang, Olga Russakovsky, Sanjeev Arora

Links: Paper 


SimPO: Simple Preference Optimization with a Reference-Free Reward
 

poster

Authors: Yu Meng, Mengzhou Xia, Danqi Chen

Links: Paper 


CiteME: Can Language Models Accurately Cite Scientific Claims?
 

poster

Authors: Ori Press, Andreas Hochlehnert, Ameya Prabhu, Vishaal Udandarao, Ofir Press, Matthias Bethge

Links: Paper 


Neural network learns low-dimensional polynomials with SGD near the information-theoretic limit
 

poster

Authors: Jason Lee, Kazusato Oko, Taiji Suzuki, Denny Wu

Links: Paper 


CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs
 

poster

Authors: Zirui Wang, Mengzhou Xia, Luxi He, Howard Chen, Yitao Liu, Richard Zhu, Kaiqu Liang, Xindi Wu, Haotian Liu, Sadhika Malladi, Alexis Chevalier, Sanjeev Arora, Danqi Chen

Links: Paper 


Beyond Aesthetics: Cultural Competence in Text-to-Image Models
 

poster

Authors: Nithish Kannen Senthilkumar, Arif Ahmad, Marco Andreetto, Vinodkumar Prabhakaran, Utsav Prabhu, Adji Bousso Dieng, Pushpak Bhattacharyya, Shachi Dave

Links: Paper 


Optimal Aggregation of Prediction Intervals under Unsupervised Domain Shift
 

poster

Authors: Jiawei Ge, Debarghya Mukherjee, Jianqing Fan

Links: Paper 


Don't Compress Gradients in Random Reshuffling: Compress Gradient Differences
 

poster

Authors: Abdurakhmon Sadiev, Grigory Malinovsky, Eduard Gorbunov, Igor Sokolov, Ahmed Khaled, Konstantin Burlachenko, Peter Richtarik

Links: Paper 


REBEL: Reinforcement Learning via Regressing Relative Rewards
 

poster

Authors: Zhaolin Gao, Jonathan Chang, Wenhao Zhan, Owen Oertell, Gokul Swamy, Kianté Brantley, Thorsten Joachims, Drew Bagnell, Jason Lee, Wen Sun

Links: Paper 


Online Control in Population Dynamics
 

poster

Authors: Noah Golowich, Elad Hazan, Zhou Lu, Dhruv Rohatgi, Y. Jennifer Sun

Links: Paper 


FlexSBDD: Structure-Based Drug Design with Flexible Protein Modeling
 

poster

Authors: Zaixi Zhang, Mengdi Wang, Qi Liu

Links: Paper 


Scaling Laws in Linear Regression: Compute, Parameters, and Data
 

poster

Authors: Licong Lin, Jingfeng Wu, Sham Kakade, Peter Bartlett, Jason Lee

Links: Paper 


Global Convergence in Training Large-Scale Transformers
 

poster

Authors: Cheng Gao, Yuan Cao, Zihao Li, Yihan He, Mengdi Wang, Han Liu, Jason Klusowski, Jianqing Fan

Links: Paper 


One-Layer Transformer Provably Learns One-Nearest Neighbor In Context
 

poster

Authors: Zihao Li, Yuan Cao, Cheng Gao, Yihan He, Han Liu, Jason Klusowski, Jianqing Fan, Mengdi Wang

Links: Paper 


Mixture of neural fields for heterogeneous reconstruction in cryo-EM
 

poster

Authors: Axel Levy, Rishwanth Raghu, David Shustin, Adele Peng, Huan Li, Oliver Clarke, Gordon Wetzstein, Ellen Zhong

Links: Paper 


The Mamba in the Llama: Distilling and Accelerating Hybrid Models
 

poster

Authors: Junxiong Wang, Daniele Paliotta, Avner May, Alexander Rush, Tri Dao

Links: Paper