Portrait de Tianyu Zhang

Tianyu Zhang

Doctorat - UdeM
Superviseur⋅e principal⋅e
Sujets de recherche
Apprentissage multimodal
Raisonnement
Théorie des jeux
Traitement du langage naturel

Publications

MAP: Low-Compute Model Merging with Amortized Pareto Fronts via Quadratic Approximation
Zhiqi Bu
Huan He
Yonghui Wu
Jiang Bian
Yong Chen
Model merging has emerged as an effective approach to combine multiple single-task models into a multitask model. This process typically inv… (voir plus)olves computing a weighted average of the model parameters without any additional training. Existing model-merging methods focus on enhancing average task accuracy. However, interference and conflicts between the objectives of different tasks can lead to trade-offs during the merging process. In real-world applications, a set of solutions with various trade-offs can be more informative, helping practitioners make decisions based on diverse preferences. In this paper, we introduce a novel and low-compute algorithm, Model Merging with Amortized Pareto Front (MAP). MAP efficiently identifies a Pareto set of scaling coefficients for merging multiple models, reflecting the trade-offs involved. It amortizes the substantial computational cost of evaluations needed to estimate the Pareto front by using quadratic approximation surrogate models derived from a pre-selected set of scaling coefficients. Experimental results on vision and natural language processing tasks demonstrate that MAP can accurately identify the Pareto front, providing practitioners with flexible solutions to balance competing task objectives. We also introduce Bayesian MAP for scenarios with a relatively low number of tasks and Nested MAP for situations with a high number of tasks, further reducing the computational cost of evaluation.
MuPT: A Generative Symbolic Music Pretrained Transformer
Xingwei Qu
Yuelin Bai
Yinghao Ma
Ziya Zhou
Ka Man Lo
Jiaheng Liu
Ruibin Yuan
Lejun Min
Xueling Liu
Xinrun Du
Shuyue Guo
Yiming Liang
Yizhi Li
Shangda Wu
Junting Zhou
Tianyu Zheng
Ziyang Ma
Fengze Han
Wei Xue … (voir 8 de plus)
Gus Xia
Emmanouil Benetos
Xiang Yue
Chenghua Lin
Xu Tan
Ge Zhang
In this paper, we explore the application of Large Language Models (LLMs) to the pre-training of music. While the prevalent use of MIDI in m… (voir plus)usic modeling is well-established, our findings suggest that LLMs are inherently more compatible with ABC Notation, which aligns more closely with their design and strengths, thereby enhancing the model's performance in musical composition.To address the challenges associated with misaligned measures from different tracks during generation, we propose the development of a
VCR: A Task for Pixel-Level Complex Reasoning in Vision Language Models via Restoring Occluded Text
We introduce Visual Caption Restoration (VCR), a novel vision-language task that challenges models to accurately restore partially obscured … (voir plus)texts using pixel-level hints within images. This task stems from the observation that text embedded in images is intrinsically different from common visual elements and natural language due to the need to align the modalities of vision, text, and text embedded in images. While numerous works have integrated text embedded in images into visual question-answering tasks, approaches to these tasks generally rely on optical character recognition or masked language modeling, thus reducing the task to mainly text-based processing. However, text-based processing becomes ineffective in VCR as accurate text restoration depends on the combined information from provided images, context, and subtle cues from the tiny exposed areas of masked texts. We develop a pipeline to generate synthetic images for the VCR task using image-caption pairs, with adjustable caption visibility to control the task difficulty. With this pipeline, we construct a dataset for VCR called VCR-Wiki using images with captions from Wikipedia, comprising 2.11M English and 346K Chinese entities in both easy and hard split variants. Our results reveal that current vision language models significantly lag behind human performance in the VCR task, and merely fine-tuning the models on our dataset does not lead to notable improvements. We release VCR-Wiki and the data construction code to facilitate future research.
Advantage Alignment Algorithms
The growing presence of artificially intelligent agents in everyday decision-making, from LLM assistants to autonomous vehicles, hints at a … (voir plus)future in which conflicts may arise from each agent optimizing individual interests. In general-sum games these conflicts are apparent, where naive Reinforcement Learning agents get stuck in Pareto-suboptimal Nash equilibria. Consequently, opponent shaping has been introduced as a method with success at finding socially beneficial equilibria in social dilemmas. In this work, we introduce Advantage Alignment, a family of algorithms derived from first principles that perform opponent shaping efficiently and intuitively. This is achieved by aligning the advantages of conflicting agents in a given game by increasing the probability of mutually-benefiting actions. We prove that existing opponent shaping methods, including LOLA and LOQA, implicitly perform Advantage Alignment. Compared to these works, Advantage Alignment mathematically simplifies the formulation of opponent shaping and seamlessly works for continuous action domains. We also demonstrate the effectiveness of our algorithm in a wide range of social dilemmas, achieving state of the art results in each case, including a social dilemma version of the Negotiation Game.
BigDocs: An Open Dataset for Training Multi-modal Models on Document and Code Tasks
Xiangru Jian
Akshay Kalkunte
Amirhossein Abaskohi
Pierre-Andre Noel
Sanket Biswas … (voir 23 de plus)
Sara Shanian
Noah Bolger
Kurt MacDonald
Simon Fauvel
Sathwik Tejaswi
Srinivas Sunkara
Joao Monteiro
Krishnamurthy Dj Dvijotham
Torsten Scholak
Sepideh Kharagani
Sean Hughes
M. Özsu
Christopher Pal
Sai Rajeswar
Multimodal AI has the potential to significantly enhance document-understanding tasks, such as processing receipts, understanding workflows,… (voir plus) extracting data from documents, and summarizing reports. Code generation tasks that require long-structured outputs can also be enhanced by multimodality. Despite this, their use in commercial applications is often limited due to limited access to training data and restrictive licensing, which hinders open access. To address these limitations, we introduce BigDocs-7.5M, a high-quality, open-access dataset comprising 7.5 million multimodal documents across 30 tasks. We use an efficient data curation process to ensure our data is high-quality and license-permissive. Our process emphasizes accountability, responsibility, and transparency through filtering rules, traceable metadata, and careful content analysis. Additionally, we introduce BigDocs-Bench, a benchmark suite with 10 novel tasks where we create datasets that reflect real-world use cases involving reasoning over Graphical User Interfaces (GUI) and code generation from images. Our experiments show that training with BigDocs-Bench improves average performance up to 25.8% over closed-source GPT-4o in document reasoning and structured output tasks such as Screenshot2HTML or Image2Latex generation. Finally, human evaluations showed a preference for outputs from models trained on BigDocs over GPT-4o. This suggests that BigDocs can help both academics and the open-source community utilize and improve AI tools to enhance multimodal capabilities and document reasoning. The project is hosted at https://bigdocs.github.io .
MAP: Model Merging with Amortized Pareto Front Using Limited Computation
Li Li
Zhiqi Bu
Huan He
Yonghui Wu
Jiang Bian
Yong Chen
Expected Flow Networks in Stochastic Environments and Two-Player Zero-Sum Games
Generative flow networks (GFlowNets) are sequential sampling models trained to match a given distribution. GFlowNets have been successfully … (voir plus)applied to various structured object generation tasks, sampling a diverse set of high-reward objects quickly. We propose expected flow networks (EFlowNets), which extend GFlowNets to stochastic environments. We show that EFlowNets outperform other GFlowNet formulations in stochastic tasks such as protein design. We then extend the concept of EFlowNets to adversarial environments, proposing adversarial flow networks (AFlowNets) for two-player zero-sum games. We show that AFlowNets learn to find above 80% of optimal moves in Connect-4 via self-play and outperform AlphaZero in tournaments.
Correction to: Multi-agent reinforcement learning for fast-timescale demand response of residential loads
Vincent Mai
Philippe Maisonneuve
Multi-Agent Reinforcement Learning for Fast-Timescale Demand Response of Residential Loads
Vincent Mai
Philippe Maisonneuve
To integrate high amounts of renewable energy resources, electrical power grids must be able to cope with high amplitude, fast timescale var… (voir plus)iations in power generation. Frequency regulation through demand response has the potential to coordinate temporally flexible loads, such as air conditioners, to counteract these variations. Existing approaches for discrete control with dynamic constraints struggle to provide satisfactory performance for fast timescale action selection with hundreds of agents. We propose a decentralized agent trained with multi-agent proximal policy optimization with localized communication. We explore two communication frameworks: hand-engineered, or learned through targeted multi-agent communication. The resulting policies perform well and robustly for frequency regulation, and scale seamlessly to arbitrary numbers of houses for constant processing times.
AI For Global Climate Cooperation 2023 Competition Proceedings
Prateek Arun Gupta
Li Li
Soham Rajesh Phade
Sunil Srinivasa
Andrew Robert Williams
Yang Zhang
Stephan Zheng
The international community must collaborate to mitigate climate change and sustain economic growth. However, collaboration is hard to achie… (voir plus)ve, partly because no global authority can ensure compliance with international climate agreements. Combining AI with climate-economic simulations offers a promising solution to design international frameworks, including negotiation protocols and climate agreements, that promote and incentivize collaboration. In addition, these frameworks should also have policy goals fulfillment, and sustained commitment, taking into account climate-economic dynamics and strategic behaviors. These challenges require an interdisciplinary approach across machine learning, economics, climate science, law, policy, ethics, and other fields. Towards this objective, we organized AI for Global Climate Cooperation, a Mila competition in which teams submitted proposals and analyses of international frameworks, based on (modifications of) RICE-N, an AI-driven integrated assessment model (IAM). In particular, RICE-N supports modeling regional decision-making using AI agents. Furthermore, the IAM then models the climate-economic impact of those decisions into the future. Whereas the first track focused only on performance metrics, the proposals submitted to the second track were evaluated both quantitatively and qualitatively. The quantitative evaluation focused on a combination of (i) the degree of mitigation of global temperature rise and (ii) the increase in economic productivity. On the other hand, an interdisciplinary panel of human experts in law, policy, sociology, economics and environmental science, evaluated the solutions qualitatively. In particular, the panel considered the effectiveness, simplicity, feasibility, ethics, and notions of climate justice of the protocols. In the third track, the participants were asked to critique and improve RICE-N.
Biological Sequence Design with GFlowNets
Alex-Hernandez Garcia
Bonaventure F. P. Dossou
Chanakya Ekbote
Michael Kilgour
Payel Das
Design of de novo biological sequences with desired properties, like protein and DNA sequences, often involves an active loop with several r… (voir plus)ounds of molecule ideation and expensive wet-lab evaluations. These experiments can consist of multiple stages, with increasing levels of precision and cost of evaluation, where candidates are filtered. This makes the diversity of proposed candidates a key consideration in the ideation phase. In this work, we propose an active learning algorithm leveraging epistemic uncertainty estimation and the recently proposed GFlowNets as a generator of diverse candidate solutions, with the objective to obtain a diverse batch of useful (as defined by some utility function, for example, the predicted anti-microbial activity of a peptide) and informative candidates after each round. We also propose a scheme to incorporate existing labeled datasets of candidates, in addition to a reward function, to speed up learning in GFlowNets. We present empirical results on several biological sequence design tasks, and we find that our method generates more diverse and novel batches with high scoring candidates compared to existing approaches.
ClimateGAN: Raising Climate Change Awareness by Generating Images of Floods
Alexandra Luccioni
Mélisande Teng
Gautier Cosne
Adrien Juraver
Climate change is a major threat to humanity, and the actions required to prevent its catastrophic consequences include changes in both poli… (voir plus)cy-making and individual behaviour. However, taking action requires understanding the effects of climate change, even though they may seem abstract and distant. Projecting the potential consequences of extreme climate events such as flooding in familiar places can help make the abstract impacts of climate change more concrete and encourage action. As part of a larger initiative to build a website that projects extreme climate events onto user-chosen photos, we present our solution to simulate photo-realistic floods on authentic images. To address this complex task in the absence of suitable training data, we propose ClimateGAN, a model that leverages both simulated and real data for unsupervised domain adaptation and conditional image generation. In this paper, we describe the details of our framework, thoroughly evaluate components of our architecture and demonstrate that our model is capable of robustly generating photo-realistic flooding.