Generative AI has been at the forefront of AI research in the most recent times. A large number of research works across different modalities (e.g., text, image and audio) have shown remarkable generation capabilities. Audio generation brings its own unique challenges and this workshop is aimed at highlighting these challenges and their solutions. It will bring together researchers working on different audio generation problems and enable a concentrated discussions on the topic. The workshop will include invited talks, high-quality papers presented through oral and poster sessions, and a panel discussion including experts in the area to further enhance the quality of discussion on audio generation research. A crucial part of audio generation research is its perceptual experience by humans. To enable this, \emph{we also propose to have an onsite demo session during the workshop where presenters can showcase their audio generation methods and technologies}, leading to a unique experience for all workshop participants.

Examining the fusion of neuroscience and AI, this workshop aims to unlock brain-inspired algorithms while advancing both biological and artificial intelligence.

The fourth version of the Efficient Natural Language and Speech Processing (ENLSP-IV) workshop will focus on how to make large language and foundation models more efficient in terms of Architecture, Training, and Inference in their real-world applications. This year, following the trend of industry and academia, we put more emphasis on investigating new architectures to make future language and foundation models more efficient. Moreover, we highlight the importance of comprehensive evaluation and benchmarking new efficient models from different practical aspects. The workshop program offers an interactive platform for gathering experts and talents from academia and industry through invited talks, panel discussion, paper submission, reviews, interactive poster sessions, oral presentations and a couple of mentorship sessions for new researchers. This will be a unique opportunity to discuss and share challenging problems, build connections, exchange ideas and brainstorm, and foster future collaborations. The topics of this workshop can be of interest for people working on general machine learning, deep learning, hardware, optimization, theory and applications.

The AI for Accelerated Materials Discovery (AI4Mat) Workshop NeurIPS 2024 provides an inclusive and collaborative platform where AI researchers and material scientists converge to tackle the cutting-edge challenges in AI-driven materials discovery and development. By taking a comprehensive look at automated materials discovery spanning AI-guided design, synthesis and automated material characterization, we hope to create an opportunity for deep, thoughtful discussion among researchers working on these interdisciplinary topics, and highlight ongoing challenges in the field. This year, AI4Mat will focus on two major themes:

Why Isn't it Real Yet? This discussion centers on why AI in materials science has not yet experienced the type of exponential growth seen in adjacent fields at the intersection of science and AI, such as large language models (LLM), multi-modal AI, drug discovery and computational biology.

AI4Mat Unique Challenges: Managing Multimodal, Incomplete Materials Data: A unique challenge in materials science is managing multimodal, incomplete data that is collected from diverse types of real-world equipment, including synthesis and characterization tools. Additionally, datasets and scientific understanding are often incomplete given the fact that fundamental physics and chemistry phenomena are sometimes unknown.

Recently-developed algorithmic innovations (e.g., transformers, diffusion models , state-space models) and large-scale datasets (e.g., Common Crawl, LAION) have given rise to machine learning models with impressive capabilities. As the cost of training such large models grows, and as systems based on them are used widely, it is increasingly important to understand how different design choices combine to induce observed behaviors. For example, we still do not fully understand how the composition of training datasets influences model behavior (e.g., how does training on code data affect reasoning capabilities in other domains?), how to attribute capabilities to subcomponents (e.g., can we identify which subnetwork of an LLM implements addition), and which algorithmic choices really drive performance (e.g., how can we best align models to human preferences?). Behavioral attribution is also important in light of recent concerns about harmful model behavior and several works suggest that these behaviors can be attributed to training data or model architecture and size.The core challenge in all of these questions is that of model behavior attribution.That is, the question of relating model behavior back to factors in the machine learning pipeline---such as the choice of training dataset or particular training algorithm---that produced this model. This workshop aims to bring together researchers and practitioners that advance our understanding of model behavior attribution in the contexts that span data, model understanding, and algorithmic interventions.

Neural models tend to learn similar representations when subject to similar stimuli; this behavior has been observed both in biological and artificial settings. The emergence of these similar representations is igniting a growing interest in the fields of neuroscience and artificial intelligence. To gain a theoretical understanding of this phenomenon, promising directions include: analyzing the learning dynamics and studying the problem of identifiability in the functional and parameter space. This has strong consequences in unlocking a plethora of applications in ML from model fusion, model stitching, to model reuse and in improving the understanding of biological and artificial neural models, including large retrained foundation models. The objective of the workshop is to discuss theoretical findings, empirical evidence and practical applications of this phenomenon, benefiting from the cross-pollination of different fields (ML, Neuroscience, Cognitive Science) to foster the exchange of ideas and encourage collaborations. Overall the questions we aim to investigate are when, why and how internal representations of distinct neural models can be unified into a common representation.

The rapid evolution of multimodal foundation models, capable of processing and generating language, images, video, and audio, has transformed numerous fields, including robotics, healthcare, and AI-driven media. However, these advancements bring forth significant challenges related to reliability, security, and societal impact. Instances of model hallucinations and the inadvertent generation of harmful content by Text-to-Image (T2I) models underscore the need for responsible and sustainable development practices.Our workshop aims to address these critical issues by establishing design principles that prioritize precautionary measures over reactive solutions. We will explore methodologies to enhance the reliability and robustness of multimodal models, focusing on fairness, security, and the mitigation of misinformation. By emphasizing preemptive strategies during dataset curation and model pre-training, we aim to reduce the extensive resource demands traditionally associated with iterative refinement processes.Key topics of discussion will include the identification of reliability concerns stemming from data quality, model architecture, and training strategies. Additionally, we will explore novel design principles that ensure the responsible and sustainable advancement of multimodal generative models. Our goal is to foster a collaborative environment where leading researchers and practitioners can develop actionable frameworks that align with ethical standards and maximize societal benefits.Through keynote talks, panel discussions, and interactive sessions, this workshop will provide a comprehensive platform for the AI community to converge on best practices for building the next generation of multimodal foundation models. We seek to ensure these models are not only technologically advanced but also secure, equitable, and environmentally sustainable.

Generative AI (GenAI) emerged as a strong tool that can revolutionize healthcare and medicine. Yet the public trust in using GenAI for health is not well established due to its potential vulnerabilities and insufficient compliance with health policies. The workshop aims to gather machine learning researchers and healthcare/medicine experts from both academia and industry to explore the transformative potential of GenAI for health. We will delve into the trustworthiness risks and mitigation of cutting-edge GenAI technologies applicable in health applications, such as Large Language Models, and multi-modality large models. By fostering multidisciplinary communication with experts in government policies, this workshop seeks to advance the integration of GenAI in healthcare, ensuring safe, effective, ethical, and policy-compliant deployment to enhance patient outcomes and clinical research.

Towards next-generation medical analysis: Unlock the potential of medical foundation models for more explainable, robust, secure diagnosis solutions. Workshop Homepage: https://wymancv.github.io/AIM-FM-NeurIPS-Workshop/

Ludwig Wittgenstein, in his seminal work "Philosophical Investigations", introduced the concept of "language games." This framework views language as an adaptive system where words acquire meaning through use, emphasizing its social and interactive nature. Research in cognitive science reinforces this notion, highlighting that genuine language acquisition thrives on dynamic and context-driven interactions. Language emergence simulations further demonstrate the critical role of language transmission within a population of agents in shaping modern languages. Game theory experiments showcase the superiority of interactive self-play loops compared to traditional imitation-based models. But... meanwhile... the core training paradigm in language processing remains purely based on supervised and preference losses, and it has barely changed over the past years. Besides, some limitations in LLMs, e.g., restricted planning abilities and insufficient personalization, suggest a potential deficiency in their training: the lack of interaction. Inspired by these observations, our workshop explores the concept of Language Gamification to enable interactive LLM finetuning at scale.This training paradigm encompasses interactive training or evaluation loops that enable LLMs to bootstrap and ground their language through multi-agent interactions. Following this definition, the workshop invites an exploration of Language Gamification through a diverse set of methodological perspectives and research backgrounds, offering a series of presentations and unique panel discussions

Mathematical reasoning is a fundamental aspect of human cognition that has been studied by scholars ranging from philosophers to cognitive scientists and neuroscientists. Mathematical reasoning involves analyzing complex information, identifying patterns and relationships, and drawing logical conclusions from evidence. It is central to many applications in science, engineering, finance, and everyday contexts. Recent advancements in large language models (LLMs) have unlocked new opportunities at the intersection of artificial intelligence and mathematical reasoning, ranging from new methods that solve complex problems or prove theorems, to new forms of human-machine collaboration in mathematics and beyond. Our proposed workshop is centered on the intersection of deep learning and mathematical reasoning, with an emphasis on, but not limited to, large language models. Our guiding theme is: ``To what extent can machine learning models comprehend mathematics, and what applications could arise from this capability?''

In the past, continual learning (CL) was often overlooked as problems solvable by CL were efficiently addressed by offline learning, where resource efficiency wasn't a significant bottleneck. However, with the advent of foundation models, the landscape has shifted. For the pursuit of increasingly general intelligence, current foundation models are fundamentally limited by their training on static data, leading to outdated encoded information, saturation in knowledge accumulation, and wasteful use of compute resources. The increasing size of ML models puts ever more emphasis on scalable CL, as even fine-tuning large models is becoming increasingly resource-intensive and time-consuming. CL now emerges as a crucial framework in this new era, essential for dealing with the evolving scale and complexity of ML models. Yet, even the most recent methods in CL fall short of effectively addressing the challenges posed by the current data and compute scales. At this workshop, we discuss recent advances in scalable CL that could potentially replace static foundation model training, enabling us to model dynamic real-world information. Our workshop aims to bring together experts and researchers from various domains, including language, vision, speech, and multimodal AI to exchange ideas and foster collaboration. We are committed to advancing the development of next-generation foundation models that can learn and adapt continuously, addressing both technical and ethical aspects. With invited and contributed talks by distinguished researchers in the area, the workshop will delve into the evolving definition of CL, and how CL can enable the efficient development of foundation models. We will conclude the workshop with a panel discussion on how foundation models and their CL will arguably transform ML research, as well as their societal and environmental implications. We will also ensure there is ample time for discussions that encourage networking between researchers from different sub-communities, which we hope will result in new long-term collaborations. The workshop's program will showcase a diverse range of perspectives, reflecting the approaches pursued by specific sub-communities.

Adversarial machine learning (AdvML), a discipline that delves into the interaction of machine learning (ML) with ‘adversarial’ elements, has embarked on a new era propelled by the ever-expanding capabilities of artificial intelligence (AI). This momentum has been fueled by recent technological breakthroughs in large multimodal models (LMMs), particularly those designed for vision and language applications. The 3rd AdvML-Frontiers workshop at NeurIPS’24 continues the success of its predecessors, AdvML-Frontiers’22-23, by delving into the dynamic intersection of AdvML and LMMs. The rapid evolution of LMMs presents both new challenges and opportunities for AdvML, which can be distilled into two primary categories: AdvML for LMMs and LMMs for AdvML. This year, in addition to continuing to advance AdvML across the full theory-algorithm-application stack, the workshop is dedicated to addressing the intricate issues that emerge from these converging fields, with a focus on adversarial threats, cross-modal vulnerabilities, defensive strategies, multimodal human/AI feedback, and the overarching implications for security, privacy, and ethics. Join us at AdvML-Frontiers'24 for a comprehensive exploration of adversarial learning at the intersection with cutting-edge multimodal technologies, setting the stage for future advancements in adversarial machine learning. The workshop also hosts the 2024 AdvML Rising Star Award.

In the rapidly evolving landscape of AI, the development of adaptive foundation models represents a ground-breaking shift towards AI systems that can continually learn, adapt, and evolve in response to new information, changing environments, and user preferences. This workshop aims to explore cutting-edge advancements in adaptive foundation models, focusing on methodologies that enable continual weight updates, memory-efficient fine-tuning, and personalized adaptation to diverse tasks and domains. We feature invited talks by experts in LLMs, diffusion models, multimodal learning, continual learning, and efficient ML to explore this interdisciplinary topic. We host workshop paper submissions and invite oral papers for contributed talks. In addition, there is a panel discussion with the invited speakers.

Tables are a promising modality for representation learning and generative models with too much application potential to ignore. However, tables have long been overlooked despite their dominant presence in the data landscape, e.g. data management, data analysis, and ML pipelines. The majority of datasets in Google Dataset Search, for example, resembles typical tabular file formats like CSVs. Similarly, the top-3 most-used database management systems are all intended for relational data. Representation learning for tables, possibly combined with other modalities such as code and text, has shown impressive performance for tasks like semantic parsing, question answering, table understanding, data preparation, and data analysis (e.g. text-to-sql). The pre-training paradigm was shown to be effective for tabular ML (classification/regression) as well. More recently, we also observe promising potential in applying and enhancing generative models (e.g. LLMs) in the domain of structured data to improve how we process and derive insights from structured data.

The Table Representation Learning workshop has been the key venue driving this research vision and establishing a community around TRL. The goal of the third edition of TRL at NeurIPS 2024 is to:
1) showcase the latest impactful TRL research, with a particular focus on industry insights this year,
2) explore new applications, techniques and open challenges for representation learning and generative models for tabular data,
3) facilitate discussion and collaboration across the ML, NLP, and DB communities.

At NeurIPS from 2020 to 2024, we successfully organized the 1st, 2nd, 3rd and 4t workshops on Self-Supervised Learning – Theory and Practice. These events attracted a diverse audience from multiple domains, including vision, speech, NLP, robotics, ML theory, and industry practitioners. Building on the success of these previous workshops, we are excited to continue organizing the workshop on self-supervised learning this year. Self-supervised learning (SSL) is an approach of representation learning that does not rely on human-labeled data. Instead, it creates auxiliary tasks from unlabeled input data and learns representations by solving these tasks. SSL has shown significant success across various domains such as images (e.g., MAE, DINO, MoCo, PIRL, SimCLR), speech (e.g., wav2vec, Whisper), and text (e.g., BERT, GPT, Llama). It has also demonstrated promising results in other data modalities including graphs, time-series, and audio. Recent large language models—predominantly trained on web-scale data using self-supervised methods—have exhibited remarkable generalizability and are beginning to transform numerous research fields. SSL, without using human-provided labels, can achieve performance comparable to or even surpassing that of fully supervised methods. Furthermore, generative SSL techniques such as Imagen, Stable Diffusion, and SORA have significantly enhanced the artistic capabilities of AI models. Existing research on self-supervised learning (SSL) has primarily concentrated on enhancing empirical performance without substantial theoretical underpinnings. Although SSL approaches are empirically effective across various benchmarks, their theoretical foundations and practical applications remain less explored. Key questions such as the reasons behind the superior performance of certain auxiliary tasks, the requisite amount of unlabeled data for learning effective representations, the impact of neural architectures on SSL performance, and the practical scenarios where SSL outperforms supervised models, are still largely unanswered. Our workshop aims to address these gaps by fostering a dialogue between theory and practice, especially in the context of LLMs. We plan to gather researchers interested in SSL from diverse fields to explore the theoretical bases of empirically successful SSL methods and to discuss how these theoretical insights could further enhance SSL’s practical performance. This workshop will differentiate itself from previous SSL-related workshops by prioritizing the establishment of theoretical foundations and providing theoretical frameworks to guide the development of new SSL methods. Additionally, we will attempt to close the loop from practice to theory, by inviting practitioners to share their experiences and insights regarding the practical advantages and challenges of using SSL

The transformative potential of generative AI will not be fully attained until AI researchers develop a deeper understanding of the creative process of human artists, and build constructive partnerships based on that understanding. This workshop is intended to foster such connections.

The workshop will be located at Pinnacle Hotel Harbourfront starting from 2pm.

Location: Pinnacle Harbourfront Hotel
1133 W Hastings St, Vancouver, BC V6E 3T3

In recent years, there has been a growing appreciation for the importance of respecting the topological, algebraic, or geometric structure of data in machine learning models. In parallel, an emerging set of findings in computational neuroscience suggests that the preservation of this kind of mathematical structure may be a fundamental principle of neural coding in biology. The goal of this workshop is to bring together researchers from applied mathematics and deep learning with neuroscientists whose work reveals the elegant implementation of mathematical structure in biological neural circuitry. Group theory and differential geometry were instrumental in unifying the models of 20th-century physics. Likewise, they have the potential to unify our understanding of how neural systems form useful representations of the world.

Our workshop aims to explore the synergies between causality and large models, also known as ``foundation models,'' which have demonstrated remarkable capabilities across multiple modalities (text, images, audio, etc.). Despite their high performance, the opaque nature of these large models raises crucial questions regarding their trustworthiness, especially in safety-critical domains. A growing community of researchers is turning towards a more principled framework to address these concerns, better understand the behavior of large models, and improve their reliability: causality.Specifically, this workshop will focus on four directions: causality in large models, to assess their causal reasoning abilities, causality for improving large models, causality with large models to enhance causal inference and discovery methods, and causality of large models to understand and control their internal mechanisms. The invited speakers and panelists (almost all of which have already been confirmed to attend) represent a diverse set of perspectives and expertise, across both academia and industry.The workshop is organized by a team of 12 members from six different institutions across North America, Europe, and Asia, ensuring diversity across research interests, backgrounds, and demographics. Visit our website: https://calm-workshop-2024.github.io/

Across many application areas, machine learning (ML) systems rely on human data. Yet these systems often leave unmodelled the psychological processes that generate human data, or abstract these rich mental processes into simple models. Fortunately, there's a field full of insights about human behavior: the behavioral sciences. However, these insights are often qualitative. Integrating them into machine learning systems requires converting them into computable models and designing machine learning systems to incorporate them. The goal of this workshop is to explore incorporating insights from the behavioral sciences into machine learning systems. Our workshop will focus on one specific question in this broad area: how can we incorporate behavioral insights into formal computable models? Translating behavioral insights into computable models would enable them to interact with ML systems: behavioral models can improve ML models, and vice-versa. We hope to bring together computer scientists across many subfields with behavioral scientists to drive progress in this interdisciplinary area.

This workshop aims to contribute to the recent radical paradigm shift for fine-tuning in modern machine learning, both theoretically, computationally, and systematically. It encourages researchers to push forward the frontiers of theoretical understanding of fine-tuning, devising expeditious and resource-efficient inference and fine-tuning methods in machine learning systems, enabling their deployment within constrained computational resources.

This workshop explores theory for understanding and advancing modern ML practices, with a focus on mathematical models for empirical deep learning phenomena.

Limb loss represents a traumatic and destabilizing event in human life, significantly impacting an individual's quality of life and independence. Advancements in bionic prosthetic limbs offer a remarkable opportunity to regain mobility and functionality. Bionic limb human users (Bionic Humans) are able to learn to use those prosthetic extensions to compensate for their lost limb, and reclaim aspects of their former motor abilities. The movement generalization and environment adaptability skills displayed by humans using bionic extensions are a testament to motor intelligence, a capability yet unmatched by current artificial intelligence agents.To this end, we propose to organize MyoChallenge 2024: Physiological Dexterity and Agility in Bionic Humans, where we will provide a highly detailed neuromechanical and robotic simulation environment and invite experts worldwide to develop any type of controller for both the biological (muscle) and mechanical (bionic), including state-of-the-art reinforcement learning to solve a series of dexterous motor tasks involving human-to-bionic-limb interaction. Building on the success of the MyoChallenge on the NeurIPS 2022 and 2023 editions, this year's challenge will push the boundaries on how symbiotic human-robotic interaction needs to be coordinated to produce agile and dexterous behaviours. This year MyoChallenge will have two tracks: manipulation and locomotion. The manipulation track will require bi-manual coordination of the BionicMyoArms model -- a combination of a virtual biological arm and a Modular Prosthetic Limb (MPL). The goal will be to coordinate the use of those two limbs to manipulate a series of objects. In the locomotion track, we will exploit a new BionicMyoLegs model made from the combination of a virtual bilateral biological leg with a trans-femoral amputation together with an Open Source prosthetic Leg The goal will be to coordinate the musculo-skeleto-bionic model to navigate challenging terrains and obstacles in an oval running loop. This running circuit is inspired by the Paralympic steeplechase and Cybathlon.

We propose a workshop on the emerging frontier at the intersection between statistics and foundation models. Rigorous evaluation of large foundation models such as LLMs is necessary for reliable deployment, but it poses a towering challenge due to a lack of datasets and the black-box nature of many such models. The proposed workshop brings together the community working on understanding and improving LLMs with new statistical methodologies, and explores topics including benchmarking, measuring and correcting bias, automatic evaluation, watermarking, models/data auditing, and uncertainty quantification.

We are proposing the Algorithmic Fairness through the lens of Metrics and Evaluation (AFME)workshop, which is the fifth edition of this workshop series on algorithmic fairness. While previouseditions have explored foundational work on causal approaches to fairness and the intersection offairness with other fields of trustworthy machine learning namely interpretability, robustness, privacyand temporal aspects, this year’s workshop aims to timely reflect on fairness metrics definitions andevaluation methods.Indeed, with rapid advances in large generative models and international regulatory efforts as well aspertinent calls to understand fairness in context, it is crucial to revisit the suitability of existing fairnessmetrics and explore new bias evaluation frameworks. Our workshop aims to provide a venue to haverigorous interdisciplinary discussions around these critical topics and foster reflections on the necessityand challenges in defining adaptable fairness metrics and designing reliable evaluation techniques.## TopicThe discussion on defining and measuring algorithmic (un)fairness has predominantly been afocus in the early stages of algorithmic fairness research [Dwork et al., 2012, Zemel et al., 2013, Hardtet al., 2016, Zafar et al., 2017, Agarwal et al., 2018] resulting in four main fairness denominations:individual or group [Binns, 2020], statistical or causal [Makhlouf et al., 2023], equalizing or non-equalizing [Diana et al., 2021], and temporal or non-temporal fairness [Rateike, 2024]. Since, muchwork in the field had been dedicated to providing methodological advances within each denominationand understanding various trade-offs between fairness metrics [Binns, 2020, Heidari et al., 2019,Kleinberg et al., 2017]. However, given the changing machine learning landscape, with both increasingglobal applications and the emergence of large generative models, the question of understanding anddefining what constitutes “fairness” in these systems has become paramount again.On one hand, definitions of algorithmic fairness are being critically examined regarding the historicaland cultural values they encode [Asiedu et al., 2024, Arora et al., 2023, Bhatt et al., 2022]. Themathematical conceptualization of these definitions and their operationalization through satisfyingstatistical parities has also raised criticism of not taking into account the context within which thesesystems are deployed [Weinberg, 2022, Green and Hu, 2018].On another hand, it is still unclear how to reconcile standard fairness metrics and evaluationsdeveloped mainly for prediction and classification tasks with large generative models. While someworks proposed adapting existing fairness metrics, e.g., to large language models [Li et al., 2023,Zhang et al., 2023, Gallegos et al., 2023], questions remain on how to systematically measure fairnessfor textual outputs, or even multi-modal generative models [Schmitz et al., 2022, Chen et al., 2023,Lum et al., 2024]. Large generative models also pose new challenges to fairness evaluation withrecent work showcasing how biases towards specific tokens in large language models can influencefairness assessments during evaluation [Ding et al., 2024]. Finally, regulatory requirements introducenew challenges in defining, selecting, and assessing algorithmic fairness [Deck et al., 2024, Lauxet al., 2024, Hellman, 2020].Given these critical and timely considerations, this workshop aims to investigate how to defineand evaluate (un)fairness in today’s machine learning landscape. We are particularly interested inaddressing open questions in the field, such as:• Through a retrospective lens, what are the strengths and limitations of existing fairnessmetrics?• How to operationalize contextual definitions of fairness in diverse deployment domains?• Given the plethora of use-cases, how to systematically evaluate fairness and bias in largegenerative models?• How do recent regulatory efforts demand the utilization of fairness metrics and evaluationtechniques, and do existing ones comply with regulations?

Recent advances in ML and AI have led to impressive achievements, yet models often struggle to express uncertainty, and more importantly, make decisions that account for uncertainty. This hinders the deployment of AI models in critical applications, ranging from scientific discovery, where uncertainty quantification is essential, to real-world scenarios with unpredictable and dynamic environments, where models may encounter data vastly different from their training sets.Through the use of probability, Bayesian methods offer a powerful framework to address these limitations by quantifying uncertainty, incorporating prior knowledge, enabling adaptive decision-making and information gathering in uncertain environments. These approaches have led to significant progress and success in relevant fields, tackling critical problems such as drug discovery, hyperparameter tuning and environmental monitoring. However, challenges remain in both theory and practice, such as establishing performance guarantees and scaling up these methods to handle the complexity and dimensionality of larger data and models. On the other hand, the development of frontier models (e.g., based on large language models) presents new opportunities to enhance Bayesian methods with stronger priors and tools not previously available.This workshop aims to bring together researchers from different but closely related areas, including Bayesian optimization, active learning, uncertainty quantification, Gaussian processes, spatiotemporal modeling, and sequential experimental design. We seek to foster a vibrant exchange of ideas, showcase successful applications, and prompt fruitful discussion to collaboratively tackle the emerging challenges and shape the future directions of Bayesian decision-making and uncertainty in the new era of ML and AI.

Aligning AI with human preferences and societal values is increasingly important. Yet, today’s AI alignment methods have been shown to be insufficient for capturing the vast space of complex – and often conflicting – real-world values. Our workshop will discuss how to integrate diverse perspectives, values, and expertise into pluralistic AI alignment. We aim to explore new methods for multi-objective alignment by drawing inspiration from governance and consensus-building practices to address conflicting values in pluralistic AI alignment. Discussion will include technical approaches for dataset collection, algorithms development, and the design of human-AI interaction workflows that reflect pluralistic values among diverse populations. By gathering experts from various fields, this workshop seeks to foster interdisciplinary collaboration and push the boundaries of the understanding, development and practice of pluralistic AI alignment.

We propose a challenge organised in conjunction with the Fair Universe project, a collaborative effort funded by the US Department of Energy and involving the Lawrence Berkeley National Laboratory, Université Paris-Saclay, University of Washington, and ChaLearn. This initiative aims to forge an open AI ecosystem for scientific discovery. The challenge will focus on measuring the physics properties of elementary particles with imperfect simulators due to differences in modelling systematic errors. Additionally, the challenge will leverage a large-compute-scale AI platform for sharing datasets, training models, and hosting machine learning competitions. Our challenge will bring together the physics and machine learning communities to advance our understanding and methodologies in handling systematic (otherwise known as epistemic) uncertainties within AI techniques.

We are launching the first AI track for the popular Meta Hacker Cup programmingcompetition, designed to assess the capabilities of Generative AI in performingautonomous code generation tasks. We aim to test the limits of AI in complexcoding challenges and measure the performance gap between AI systems andhuman programmers. We will provide access to all Hacker Cup problems since 2011alongside their respective solutions in a multimodal (image and text) format, andutilize the existing Hacker Cup infrastructure for competitor evaluation. Featuringboth "open evaluation, open model" and "open evaluation, closed model" tracks,this competition invites diverse participation from research institutions of variedinterests and resource constraints, including academic labs, AI startups, largetechnology companies, and AI enthusiasts. Our goal is to develop and democratizemeaningful advancements in code automation with the very first open evaluationprocess for competitive AI programmers.

The Ariel Data Challenge 2024 tackles one of astronomy's hardest data analysis problems - extracting faint exoplanetary signals from noisy space telescope observations like the upcoming Ariel Mission. A major obstacle are systematic noise sources, such as ``jitter noise" arising from spacecraft vibrations, which corrupts spectroscopic data used to study exoplanet atmospheres. This complex spatio-temporal noise challenges conventional parametric denoising techniques. In this challenge, the jitter time series is simulated based on Ariel's payload design and other noise effects are taken from in-flight data from JWST, in order to provide a realistic representation of the effect.To recover minute signals from the planet's atmosphere, participants must push boundaries of current approaches to denoise this multimodality data across image, time, and spectral domains. This requires novel solutions for non-Gaussian noise, data drifts, uncertainty quantification, and limited ground truth. Success will directly improve the Ariel pipeline design and enable new frontiers in characterising exoplanet atmospheres - a key science priority in the coming decades for understanding planetary formation, evolution, and habitability.

NeurIPS 2024 workshop Socially Responsible Language Modelling Research (SoLaR), proposed herein has two goals: (a) highlight novel and important research directions in responsible LM research across various sub-communities. (b) Promote interdisciplinary collaboration and dialogue on socially responsible LM research across communities. For example, between i) the AI safety and FATE (fairness, accountability, transparency, and ethics) communities and ii) technical and policy communities. To achieve this goal, we have assembled a diverse line-up of speakers who will talk about LM research in the context of governance, ethics, fairness, safety and alignment. We will also be holding a panel on whether or not it is socially responsible to continue the pursuit for AGI-like, more capable and more general-purpose LMs; an extremely timely topic considering multiple leading AI labs are explicitly focusing on achieving this goal.

The growing relevance of video-language models in both academia and industry highlights the necessity for a dedicated workshop to address the unique challenges and opportunities this field presents. This workshop is designed to accelerate the development and practical application of video foundation models, which are crucial for interpreting and utilizing the extensive amounts of video data that make up a significant portion of global data. These models are increasingly vital for a range of applications, from video search and content creation to surveillance and robotics. Confirmed speakers are leading researchers in this field from UT Austin, University of Tübingen, and University of Bristol (Tentative), as well as prominent industry figures from Meta, Google DeepMind, and Microsoft, ensuring a rich exchange of knowledge. The diverse organizing team from universities, industry, and non-profit research institutes aims to foster broad participation and collaboration. This workshop aims to push the boundaries of video-language models, ensuring their development and deployment are ethical and responsible. It will serve as a platform for sharing knowledge, fostering collaborations, and setting future research directions in this rapidly advancing field.

The integration of machine learning (ML) techniques for addressing intricate physics problems is increasingly recognized as a promising avenue for expediting simulations. However, assessing ML-derived physical models poses a significant challenge for their adoption within industrial contexts. This competition is designed to promote the development of innovative ML approaches for tackling physical challenges, leveraging our recently introduced unified evaluation framework known as Learning Industrial Physical Simulations (LIPS). Building upon the preliminary edition held from November 2023 to March 20241, this iteration centers on a task fundamental to a well-established physical application: airfoil design simulation, utilizing our proposed AirfRANS dataset. The competition evaluates solutions based on various criteria encompassing ML accuracy, computational efficiency, Out-Of-Distribution performance, and adherence to physical principles. Notably, this competition represents a pioneering effort in exploring ML-driven surrogate methods aimed at optimizing the trade-off between computational efficiency and accuracy in physical simulations. Hosted on the Codabench platform, the competition offers online training and evaluation for all participating solutions.

The proposed competition revolves around testing the limits of agents (e.g rule-based or Meta RL agents) when it comes to adapting to a game with changing dynamics. We propose a unique 1v1 competition format where both teams face off in a sequence of 5 games. The game mechanics, along with partial observability are designed to ensure that optimal gameplay requires agents to efficiently explore and discover the game dynamics. They ensure that the strongest agents may play "suboptimally" in game 1 to explore, but then win easily in games 2 to 5 by leveraging information gained through game 1 and adapting. This competition provides a GPU parallelized game environment via jax to enable fast training/evaluation on a single GPU, lowering barriers of entry to typically industry-level scales of research. Participants can submit their agents to compete against other submitted agents on a online leaderboard hosted by Kaggle ranked by a Trueskill ranking system. The results of the competition will provide a dataset of top open-sourced rule-based agents as well as many game episodes that can lead to unique analysis (e.g. quantifying emergence/surprise) past competitions cannot usually provide thanks to the number of competitors the Lux AI Challenges often garner.

Decision-making in large-scale games is an essential research area in artificial intelligence with significant real-world impact. An agent confronts the critical task of making high-frequency strategic decisions in an uncertain and competitive environment, characterized by significant randomness and rapidly changing strategies from massive competitors. However, the shortage of large-scale, realistic game systems and datasets has hindered research progress in this area. To provide opportunities for in-depth research on this highly valuable problem, we present the Auto-Bidding in Large-Scale Auctions challenge derived from online advertising, a booming \$626.8 billion industry in 2023. We have developed a standardized ad auction system for the competition, which reproduces the characteristics of real-world large-scale games and incorporates essential features that deserve research attention. We also provide a training framework with a 500-million-record dataset and several industry-proven methods as baselines to help participants quickly start and deeply optimize their strategies.Furthermore, we have prepared a comprehensive promotional strategy, raised sufficient funds, and offered varied incentives to attract more participants from diverse backgrounds.We believe that the proposed competition will provide opportunities for more researchers to gain insights and conduct research in this field, driving technical innovation for both research and real-world practical applications.

Speech enhancement (SE) is the task of improving the quality of the desired speech while suppressing other interference signals.Tremendous progress has been achieved in the past decade in deep learning-based SE approaches.However, existing SE studies are often limited in one or multiple aspects of the following: coverage of SE sub-tasks, diversity and amount of data (especially real-world evaluation data), and diversity of evaluation metrics.As the first step to fill this gap, we establish a novel SE challenge, called URGENT, to promote research towards universal SE.It concentrates on the universality, robustness, and generalizability of SE approaches.In the challenge, we extend the conventionally narrow SE definition to cover different sub-tasks, thus allowing the exploration of the limits of current SE models.We start with four SE sub-tasks, including denoising, dereverberation, bandwidth extension, and declipping.Note that handling the above sub-tasks within a single SE model has been challenging and underexplored in the SE literature due to the distinct data formats in different tasks.As a result, most existing SE approaches are only designed for a specific subtask.To address this issue, we propose a technically novel framework to unify all these sub-tasks in a single model, which is compatible to most existing SE approaches.Several state-of-the-art baselines with different popular architectures have been provided for this challenge, including TF-GridNet, BSRNN, and Conv-TasNet.We also take care of the data diversity and amount by collecting abundant public speech and noise data from different domains.This allows for the construction of diverse training and evaluation data.Additional real recordings are further used for evaluating robustness and generalizability.Different from existing SE challenges, we adopt a wide range of evaluation metrics to provide comprehensive insights into the true capability of both generative and discriminative SE approaches.We expect this challenge would not only provide valuable insights into the current status of SE research, but also attract more research towards building universal SE models with strong robustness and good generalizability.