Publications

V-STaR: Training Verifiers for Self-Taught Reasoners
Arian Hosseini
Xingdi Yuan
Nikolay Malkin
Aaron Courville
Alessandro Sordoni
Rishabh Agarwal
Common self-improvement approaches for large language models (LLMs), such as STaR (Zelikman et al., 2022), iteratively fine-tune LLMs on sel… (voir plus)f-generated solutions to improve their problem-solving ability. However, these approaches discard the large amounts of incorrect solutions generated during this process, potentially neglecting valuable information in such solutions. To address this shortcoming, we propose V-STaR that utilizes both the correct and incorrect solutions generated during the self-improvement process to train a verifier using DPO that judges correctness of model-generated solutions. This verifier is used at inference time to select one solution among many candidate solutions. Running V-STaR for multiple iterations results in progressively better reasoners and verifiers, delivering a 4% to 17% test accuracy improvement over existing self-improvement and verification approaches on common code generation and math reasoning benchmarks with LLaMA2 models.
2024-07-10
colmweb.org/COLM/2024/Conference (accepté)
doi.org
openreview.net
Web Retrieval Agents for Evidence-Based Misinformation Detection
Jacob-Junqi Tian
Hao Yu
Yury Orlovskiy
Tyler Vergho
Mauricio Rivera
Mayank Goel
Zachary Yang
Jean-François Godbout
Reihaneh Rabbany
Kellin Pelrine
2024-07-10
colmweb.org/COLM/2024/Conference (accepté)
openreview.net
openreview.net
What makes a good metric? Evaluating automatic metrics for text-to-image consistency
Candace Ross
Melissa Hall
Adriana Romero Soriano
Adina Williams
2024-07-10
colmweb.org/COLM/2024/Conference (accepté)
openreview.net
openreview.net
Canada’s approach to SARS-CoV-2 sero-surveillance: Lessons learned for routine surveillance and future pandemics
Sheila F. O’Brien
Michael Asamoah-Boaheng
Brian Grunau
Mel Krajden
David Buckeridge
David M. Goldfarb
Maureen Anderson
Marc Germain
Patrick Brown
Derek R. Stein
Kami Kandola
Graham Tipples
Philip Awadalla
Amanda Lang
Lesley Behl
Tiffany Fitzpatrick
Steven J. Drews
2024-07-09
Canadian journal of public health (publié)
doi.org
Adaptive Accompaniment with ReaLchords
Yusong Wu
Tim Cooijmans
Kyle Kastner
Adam Roberts
Ian Simon
Alexander Scarlatos
Chris Donahue
Cassie Tarakajian
Shayegan Omidshafiei
Aaron Courville
Pablo Samuel Castro
Natasha Jaques
Anna (Cheng-Zhi) Huang
Jamming requires coordination, anticipation, and collaborative creativity between musicians. Current generative models of music produce expr… (voir plus)essive output but are not able to generate in an online manner, meaning simultaneously with other musicians (human or otherwise). We propose ReaLchords, an online generative model for improvising chord accompaniment to user melody. We start with an online model pretrained by maximum likelihood, and use reinforcement learning to finetune the model for online use. The finetuning objective leverages both a novel reward model that provides feedback on both harmonic and temporal coherency between melody and chord, and a divergence term that implements a novel type of distillation from a teacher model that can see the future melody. Through quantitative experiments and listening tests, we demonstrate that the resulting model adapts well to unfamiliar input and produce fitting accompaniment. ReaLchords opens the door to live jamming, as well as simultaneous co-creation in other modalities.
2024-07-08
Proceedings of the 41st International Conference on Machine Learning (publié)
proceedings.mlr.press
openreview.net
All-in-one simulation-based inference
Manuel Gloeckler
Michael Deistler
Christian Dietrich Weilbach
Frank Wood
Jakob H. Macke
2024-07-08
Proceedings of the 41st International Conference on Machine Learning (publié)
doi.org
openreview.net
Autoformalizing Euclidean Geometry
Logan Murphy
Kaiyu Yang
Jialiang Sun
Zhaoyu Li
Animashree Anandkumar
Xujie Si
Autoformalization involves automatically translating informal math into formal theorems and proofs that are machine-verifiable. Euclidean ge… (voir plus)ometry provides an interesting and controllable domain for studying autoformalization. In this paper, we introduce a neuro-symbolic framework for autoformalizing Euclidean geometry, which combines domain knowledge, SMT solvers, and large language models (LLMs). One challenge in Euclidean geometry is that informal proofs rely on diagrams, leaving gaps in texts that are hard to formalize. To address this issue, we use theorem provers to fill in such diagrammatic information automatically, so that the LLM only needs to autoformalize the explicit textual steps, making it easier for the model. We also provide automatic semantic evaluation for autoformalized theorem statements. We construct LeanEuclid, an autoformalization benchmark consisting of problems from Euclid’s Elements and the UniGeo dataset formalized in the Lean proof assistant. Experiments with GPT-4 and GPT-4V show the capability and limitations of state-of-the-art LLMs on autoformalizing geometry problems. The data and code are available at https://github.com/loganrjmurphy/LeanEuclid.
2024-07-08
Proceedings of the 41st International Conference on Machine Learning (publié)
doi.org
openreview.net
CKGConv: General Graph Convolution with Continuous Kernels
Liheng Ma
Soumyasundar Pal
Yitian Zhang
Jiaming Zhou
Yingxue Zhang
Mark Coates
The existing definitions of graph convolution, either from spatial or spectral perspectives, are inflexible and not unified. Defining a gene… (voir plus)ral convolution operator in the graph domain is challenging due to the lack of canonical coordinates, the presence of irregular structures, and the properties of graph symmetries. In this work, we propose a novel and general graph convolution framework by parameterizing the kernels as continuous functions of pseudo-coordinates derived via graph positional encoding. We name this Continuous Kernel Graph Convolution (CKGConv). Theoretically, we demonstrate that CKGConv is flexible and expressive. CKGConv encompasses many existing graph convolutions, and exhibits a stronger expressiveness, as powerful as graph transformers in terms of distinguishing non-isomorphic graphs. Empirically, we show that CKGConv-based Networks outperform existing graph convolutional networks and perform comparably to the best graph transformers across a variety of graph datasets. The code and models are publicly available at https://github.com/networkslab/CKGConv.
2024-07-08
Proceedings of the 41st International Conference on Machine Learning (publié)
doi.org
openreview.net
On PI Controllers for Updating Lagrange Multipliers in Constrained Optimization
Motahareh Sohrabi
Juan Ramirez
Tianyue H. Zhang
Simon Lacoste-Julien
Jose Gallego-Posada
Constrained optimization offers a powerful framework to prescribe desired behaviors in neural network models. Typically, constrained problem… (voir plus)s are solved via their min-max Lagrangian formulations, which exhibit unstable oscillatory dynamics when optimized using gradient descent-ascent. The adoption of constrained optimization techniques in the machine learning community is currently limited by the lack of reliable, general-purpose update schemes for the Lagrange multipliers. This paper proposes the νPI algorithm and contributes an optimization perspective on Lagrange multiplier updates based on PI controllers, extending the work of Stooke, Achiam and Abbeel (2020). We provide theoretical and empirical insights explaining the inability of momentum methods to address the shortcomings of gradient descent-ascent, and contrast this with the empirical success of our proposed νPI controller. Moreover, we prove that νPI generalizes popular momentum methods for single-objective minimization. Our experiments demonstrate that νPI reliably stabilizes the multiplier dynamics and its hyperparameters enjoy robust and predictable behavior.
2024-07-08
Proceedings of the 41st International Conference on Machine Learning (publié)
proceedings.mlr.press
openreview.net
EiG-Search: Generating Edge-Induced Subgraphs for GNN Explanation in Linear Time
Shengyao Lu
Bang Liu
Keith G Mills
Jiao He
Di Niu
2024-07-08
Proceedings of the 41st International Conference on Machine Learning (publié)
proceedings.mlr.press
openreview.net
Estimating Unknown Population Sizes Using the Hypergeometric Distribution
Liam Hodgson
Danilo Bzdok
The multivariate hypergeometric distribution describes sampling without replacement from a discrete population of elements divided into mult… (voir plus)iple categories. Addressing a gap in the literature, we tackle the challenge of estimating discrete distributions when both the total population size and the category sizes are unknown. Here, we propose a novel solution using the hypergeometric likelihood to solve this estimation problem, even in the presence of severe under-sampling. Our approach accounts for a data generating process where the ground-truth is a mixture of distributions conditional on a continuous latent variable, as seen in collaborative filtering, using the variational autoencoder framework. Empirical data simulation demonstrates that our method outperforms other likelihood functions used to model count data, both in terms of accuracy of population size estimate and learning an informative latent space. We showcase our method’s versatility through applications in NLP, by inferring and estimating the complexity of latent vocabularies in reading passage excerpts, and in biology, by accurately recovering the true number of gene transcripts from sparse single-cell genomics data.
2024-07-08
Proceedings of the 41st International Conference on Machine Learning (publié)
proceedings.mlr.press
Experts Don't Cheat: Learning What You Don't Know By Predicting Pairs
Daniel D. Johnson
Danny Tarlow
David Duvenaud
Chris J. Maddison
Identifying how much a model …
2024-07-08
Proceedings of the 41st International Conference on Machine Learning (publié)
doi.org
openreview.net