Events & Talks

Deep Models and Optimization Talk Upcoming 28-11-2025 How does data shape learning in LLMs? A case study of factual recall and the surprising role of data diversity (by Nicolas Zucchet, ETH Zurich) Data drives LLM training, yet we have limited scientific understanding of how it shapes learning dynamics and thus the final model. This talk, based on two recent works [1 <https://arxiv.org/abs/2503.21676> ][2 <https://arxiv.org/abs/2505.17863> ] will examine these questions with a focus on factual recall. We will begin by analyzing how LLMs learn a synthetic factual recall task, that serves as a test bed for knowledge acquisition and where we can precisely control data distribution properties. Our experiments reveal that learning proceeds in distinct stages, and, surprisingly, that skewe... Antonio Orvieto

Talk Jeannette Bohg Upcoming 17-11-2025 Special Talk: From Scaling Data to Scaling Capability: Compositional, Multimodal, and Embodied Robot Learning Scaling robot learning has led to impressive progress, but also to an unsustainable dependence on large datasets collected mainly through robot teleoperation. Moving forward, we need approaches that scale capability rather than data. In this talk, I will present three complementary efforts in this direction. Masquerade exploits the abundance of human egocentric videos to scale robot policy learning from rich but unstructured demonstrations. DexForce couples visual and tactile modalities to enable dexterous, contact-rich manipulation. MobiPi generalizes manipulation policies to mobile robots... Moritz Hardt Eva Lämmerhirt Nisha Tyagi

AI Safety and Alignment Talk Upcoming 14-11-2025 Lessons from the Impossibility of Safety What kind of results is impossible for safety research, and what pathways forward can we hope to achieve? First, we will discuss theoretical results on rule-following that demonstrate token-level jailbreaks as an architectural inevitability of attention (LogicBreaks). While initially pessimistic, these theoretical insights can also be leveraged to steer models to state of the art performance in five lines of code (InstABoost). Lastly, we will argue for a shift in safety strategy away from aligning model weights to stateful monitoring, as the only level at which one can hope to stop misuse (... Maksym Andriushchenko

Symposium 10-11-2025 Special Symposium on Intelligent Systems November 2025 On November 10, the Max Planck Institute for Intelligent Systems in Tübingen will host a special symposium on the future directions in intelligent systems. Keynote lectures by leading international scientists will offer insights into groundbreaking research and emerging trends, giving attendees a forward-looking perspective on the global development of intelligent systems. Moritz Hardt Michael Black Bernhard Schölkopf Eva Lämmerhirt Nisha Tyagi

Social Foundations of Computation Talk Elad Hazan 03-11-2025 Learning in Dynamical Systems Learning in dynamical systems is a fundamental challenge underlying modern sequence modeling. Despite extensive study, efficient algorithms with formal guarantees for general nonlinear systems have remained elusive. This talk presents a provably efficient framework for online learning in any bounded and Lipschitz nonlinear dynamical system, establishing the first sublinear regret guarantees in a dimension-free setting. Our approach combines Koopman lifting, Luenberger observers, and, crucially, spectral filtering to show that nonlinear dynamics are learnable. These insights motivate a new n... Moritz Hardt

Haptic Intelligence Talk Abdullah Muhammad 31-10-2025 What is so rapid about prototyping anyway? The declared objective of fabrication technology is to integrate into creative processes, i.e., sequences of design, prototyping, and testing activities, each of which commonly takes place on the minute to hour scale, often involving multiple people. Current fabrication technology, however, is too slow to be part of this process. People thus perform fabrication offline, the involved people spilt up, and the flow is broken. In this talk, I will challenge this traditional notion of rapid prototyping. Picking laser cutting as my weapon of choice, I will analyze the traditional fabrication p... Katherine J. Kuchenbecker

Perceiving Systems Talk Angela Yao 30-10-2025 Text, Touch and Trajectories: Reconstructing 3D Humans with 3T’s 3D human reconstruction requires more than just modeling geometry - it demands a nuanced understanding of semantics, contacts, and motions. This talk explores how we can rethink reconstruction through the lens of 3Ts: Text, Touch, and Trajectories, as three complementary signals to infer human and object structures in the wild. Text, or natural language, provides an intuitive prior for object shape and category and we use it as support for hand-object reconstruction. Touch, or contact, provide cues that link humans to the outside world. I will introduce our state-of-the-art approach on ... Nikos Athanasiou Melanie Feldhofer Nicole Overbaugh Michael J. Black

Robotic Materials Talk Dr. Peter Morshuis 29-10-2025 HVDC in Fast-Forward The focus of the first part of my presentation “HVDC in fast-forward” is on high voltage DC as an enabling technology for harvesting renewable energy. Ample attention will be given to the challenges and risks related to the fast roll-out of HVDC cable systems in Europe and in China. The need for developing on-line monitoring systems and non-intrusive sensors is discussed and research opportunities are highlighted. In the second part of my presentation I will provide a short overview of the expertise of the Institute of Electrical Insulation and Materials of Xi’an Jiaotong University, one ... Christoph Keplinger

AI Safety and Alignment Talk 13-10-2025 "Stress Testing Deliberative Alignment for Anti-Scheming Training" by Alexander Meinke (Tübingen) A core challenge in aligning powerful, goal-directed AI is the convergent incentive for an agent to preserve its own objectives against modification. A sufficiently capable model may therefore learn to 'scheme' by strategically appearing aligned when under oversight in order to avoid goal modification. In our latest work, we collaborated with OpenAI to study whether we can train models not to scheme by teaching o3 and o4-mini to avoid covert actions through deliberative alignment. The training reduces but doesn't eliminate covert behavior, and we show that some of the improvement comes from... Maksym Andriushchenko

Workshop 08-10-2025 Responsible Conduct in Research 2025 Remote via Zoom Principles of good scientific practice can be violated in many ways, from improper application of scientific methods or documenting data, to serious scientific misconduct through deliberate falsification or deceit. This workshop will provide an introduction to these important topics, with subject-specific input from senior IMPRS-IS and CLS researchers.

Perceiving Systems Talk Elin Hernlund 01-10-2025 Decoding animal health: Computer vision approaches to motion and emotional states in animals Animal body conformation plays an important role in health and athletic performance across species. The selective breeding process, introduced by humans in modern domestication, has shaped animal bodies to achieve visually appealing traits or functional specialisation. But these modifications can create predispositions to health issues that are challenging to detect. Assessment of body conformation, pose, and motion provides valuable insights into health parameters and forms a cornerstone of the veterinary clinical examination. However, traditional assessment methods face significant lim... Shashank Tripathi

Social Foundations of Computation Talk 25-09-2025 How benchmarking broke in the LLM era and what to salvage IMPRS-IS Keynote Lecture by Moritz Hardt Benchmarking is a process of continual improvement through competitive testing, central to engineering communities. Although benchmarking has long fueled progress in machine learning, there’s a growing crisis about recent generative models. In this talk, I'll discuss the causes of this crisis and how to achieve valid model comparisons—and, by extension, valid model rankings. Currently, different benchmarks yield contradictory comparisons, even when targeting the same task. Multi-task benchmarks exacerbate ranking disagreements, as do attempts to scale up evaluation. Toward diagnosing the pr...

Symposium 19-09-2025 2025 CaCTüS Scientific Symposium The symposium marks the conclusion of this year’s CaCTüS internship On September 19, 2025, the CaCTüS interns will present the results of their research projects to the campus community, showcasing the research and scientific soft skills they have developed throughout the internship. Following the presentations, we invite you to join us for food and drinks. Lea Nagel

Haptic Intelligence Talk Dr. Elizabeth Gregorio 18-09-2025 Snake Swimming Hydrodynamics, Biomechanics, and Maneuverability Snakes are anguilliform swimmers, they move through water by propagating a traveling wave of increasing amplitude along their body length. Their relatively simple body shape, swimming style, and high maneuverability has made them a popular model in robotics. Yet it is important to understand how snakes achieve swimming eﬃciency and maneuverability to successfully extend these qualities into robotics. As a snake swims its body pushes against the surrounding water, therefore, its undulation is the source of both its propulsion and drag force. We use digital defocusing particle tracking veloc... Katherine J. Kuchenbecker Andrew Schulz

Perceiving Systems Talk Francesco Palandra 11-09-2025 Advancing Animal Shape and Pose Reconstruction from Video: Challenges and Optimization Strategies The reconstruction of animal shape and pose from video is a challenging problem in computer vision, with many approaches utilizing the SMAL as a prior. A promising refinement in this field involves applying Gaussian Splatting on top of those approaches to enhance the accuracy of these reconstructions. However, our observations show that directly applying Gaussian Splatting to each frame independently does not yield optimal results due to the limited precision of the backbone models. In this talk we will highlight the need for an additional optimization step, potentially using temporal const... Vanessa Sklyarova

Haptic Intelligence Talk Vincenzo Marco De Luca 28-08-2025 Tempo-Relational Neural Networks for Advancing Team Modeling in the Operating Room Poor teamwork is a leading contributor to medical errors in the operating room. While team modeling has been studied from multiple perspectives, the literature still lacks a comprehensive analysis of which neural architectures best align with principles from social science. To address this gap, we propose a tempo-relational neural network that jointly predicts multiple team constructs while ensuring explainability. Our framework provides both factual explanations-clarifying the underlying causes of errors-and counterfactual explanations-offering actionable feedback for improving teamwork in... Katherine J. Kuchenbecker

Haptic Intelligence Talk Mary Kate Gale 28-08-2025 Training for Teleoperation of Surgical Robots: A Neuroscience Perspective For many routine surgical procedures, robot-assisted minimally invasive surgery (RMIS) has become the state of the art, providing surgeons with enhanced dexterity, precision, and control. However, the most effective methods for training surgeons to operate these systems remain unknown, and ineffective training can lead to errors with serious implications for patient safety. In this talk, I will present my research into optimizing training for surgical robotics, drawing on user studies involving the design, manipulation, and evaluation of multisensory performance feedback. I will also discus... Katherine J. Kuchenbecker

Haptic Intelligence IS Colloquium Prof. Malte F. Jung 11-08-2025 Machine-Mediated Teamwork As AI driven systems increasingly become integrated into team contexts, understanding how even subtle machine behaviors influence human interactions and relationships is critical. In this talk, I explore how simple machine actions—like allocation decisions by robots or language suggestions from generative AI—can profoundly affect interpersonal dynamics within teams. Drawing on empirical studies, I illustrate how machine behaviors impact perceptions of fairness, dominance, cooperation, and affiliation among team members. Findings from experiments involving robotic allocation tasks and AI... Katherine J. Kuchenbecker

Robotic Materials Talk Viktoriia Kamska 25-07-2025 Color dynamics: Novel structural color mechanisms may mediate camouflage strategies of blue sharks Numerous animals exhibit remarkable adaptations for dynamically altering skin coloration. For instance, chameleons modify guanine crystal spacing in their cells, flounders regulate skin pigment visibility, and cephalopods utilize complex light-diffusing structures to achieve effective camouflage. While shark skin is often perceived as grey, some species demonstrate diverse coloration and camouflage strategies, including countershading and bioluminescence. Recent reports suggest that great white sharks possess dynamic camouflage, as evidenced by the great white shark's skin color changes to ... Christoph Keplinger

Perceiving Systems Talk Rao Fu 24-07-2025 Capturing Dexterity: Large-Scale Demonstration Acquisition and Dynamic Contact Modeling for Bimanual Hand–Object Interaction Achieving dexterous manipulation in AI and robotics depends on learning from high-quality, generalizable demonstrations with detailed annotations. In this talk, I first present Gigahands, a massive and richly annotated dataset capturing diverse bimanual hand object interactions. Existing datasets often fall short in scale, diversity, and detail, limiting the effective training of generative models and policy learners. Gigahands offers the possibility of overcoming these limitations by providing a robust foundation of diverse demonstrations paired with comprehensive text annotations, which a...

Robotic Materials Talk Prof. Arjang Ruhparwar, MD 24-07-2025 Prospects of artificial muscles for cardiac support Medical soft robots represent a promising field in the treatment of cardiovascular diseases, offering a potential for millions of patients suffering from heart failure worldwide. In this context, we present our work and review the current state and future direction of artificial muscle-based soft robotic biomedical devices that support the inotropic function of the heart, with a particular focus on emerging electrothermal and pnumatic powered artificial heart muscles. These artificial muscle-powered soft robotic devices are capable of mimicking the actions of complex biological systems, suc... Christoph Keplinger

Robotic Materials Talk Aurore Claverie 11-07-2025 The S3® WEC (Wave Energy Converter) Journey SBM will present the S3® WEC (Wave Energy Converter) journey an innovative concept developed by SBM for over ten years, which has captured the imagination and interest of many people. The S3® WEC aimed to harness the power of ocean waves using electro-active polymers - advanced materials capable of generating energy through mechanical deformation. Despite notable technical achievements, testing, and the development of deep expertise across multiple domains, SBM made the strategic decision to discontinue the project. Ultimately, the S3® WEC could not demonstrate a clear path to commercial ... Christoph Keplinger

Social Foundations of Computation Talk David Blei 30-06-2025 Hierarchical Causal Models Analyzing nested data with hierarchical models is a staple of Bayesian statistics, but causal modeling remains largely focused on “flat” models. In this talk, we will explore how to think about nested data in causal models, and we will consider the advantages of nested data over aggregate data (such as means) for causal inference. We show that disaggregating your data replacing a flat causal model with a hierarchical causal model can provide new opportunities for identification and estimation. As examples, we will study how to identify and estimate causal effects under unmeasured confounder... Moritz Hardt

Haptic Intelligence PhD Thesis Defense 30-06-2025 Towards Robust and Flexible Robot State and Motion Estimation through Optimization and Learning For autonomous agents to operate reliably in the real world, they must perceive their environment, estimate their own and the environment's state, and derive appropriate actions based on their beliefs and the given task. While extensive research has been conducted on robot state estimation and geometric mapping, many existing methods remain tailored to specific hardware configurations or problem setups, limiting their scalability and applicability to future, more diverse scenarios. This thesis explores new directions for state and motion estimation, mapping, and localization, rendering i... Katherine J. Kuchenbecker Julian Nubert

Symposium 26-06-2025 - 27-06-2025 2025 Special Symposium on Intelligent Systems On June 26 and 27, the Max Planck Institute for Intelligent Systems in Tübingen will host a special symposium on the future directions in intelligent systems. Keynote lectures by leading international scientists will offer insights into groundbreaking research and emerging trends, giving attendees a forward-looking perspective on the global development of intelligent systems. Moritz Hardt Bernhard Schölkopf Michael Black Eva Lämmerhirt Nisha Tyagi

Haptic Intelligence IS Colloquium Mathieu Chollet 10-06-2025 Socially Aware Computing for Augmenting Virtual Social Interactions Human-computer interaction is increasingly concerned with the social. Social signal processing – a sub-field of artificial intelligence dealing with the recognition of social cues – enables computers to detect our social behaviour reliably; at the same time, new forms of remote social interaction, such as social virtual reality, attempt to reproduce genuine social interactions such as through tracking non-verbal behaviour and reproducing them on avatars. My research explores how these new computing paradigms dealing with social behaviour can augment, or create new forms of, virtual social i... Katherine J. Kuchenbecker

Haptic Intelligence Talk Dr. Chelsea Sidrane 30-05-2025 Neural Network Verification for Autonomous Systems Data-driven methods have proven very effective for control, planning and estimation onboard complex robotic systems. Unfortunately, the safety and reliability of the resulting system consequently becomes much more difficult to analyze. One answer to this problem is the small but growing body of work focused on formal verification of neural networks. However, many of the first works published in this area considered simple settings where the networks were analyzed in isolation. My PhD thesis, completed in 2022, and current research as a postdoctoral fellow focuses on developing methods to fo... Katherine J. Kuchenbecker

Robotic Materials Talk Prof. Dr. Friedrich Kremer 22-05-2025 Glassy dynamics, charge transport and electrode polarization in low molecular weight (ionic) liquids Broadband Dielectric Spectroscopy (BDS) [1], as one of the major tools in molecular physics, benefits from several advantages; (i) based on one of Maxwell´s equations it measures both, charge transport of mobile charge carriers and relaxations of the polar moieties in a molecular system, (ii) it covers an extraordinary wide spectral range from typically 10-5Hz – 109Hz or higher, (iii) with modern equipment the experiments can be carried out at widely varying electrical field strength and temperatures and (iv) its sensitivity increases with decreasing thickness of a sample capacitor and he... Christoph Keplinger Adrian Koh

Haptic Intelligence IS Colloquium Prof. Dr. Jürgen Steimle 12-05-2025 Human-Computer Integration The integration of computing with the human body is unlocking exciting opportunities for intelligent systems that enhance human capabilities. This talk will explore approaches to human-computer integration at the physical, sensory, and behavioral levels. First, seamless physical integration with the body requires a new generation of soft sensing and acting devices with unique form factors and material properties. I will present results from my research on multi-modal electronic skin. Second, I will discuss technologies for rich sensory integration using physiological sensors and feel-throug... Katherine J. Kuchenbecker

Haptic Intelligence IS Colloquium Prof. Dr. Maren Bennewitz 05-05-2025 Active Perception, Manipulation, and Personalized Navigation for Advanced Service Robots Service robots acting in human environments must actively perceive their environment, act foresightedly, and adapt to individual user preferences. In this talk, I will first present methods to enable robots to efficiently perceive the environment and carefully manipulate objects in cluttered scenes. In particular, I will introduce an active mapping approach for confined spaces such as household shelves or horticultural crop rows, where significant occlusions pose challenges for service robots. In these scenarios, robots need to manipulate objects to reduce occlusions and improve visibility.... Katherine J. Kuchenbecker

Outreach 03-04-2025 - 19-03-2025 Girls' Day Interessante Einblicke in Forschung und Technik Linda Behringer

Social Foundations of Computation Talk Bryan Wilder 03-04-2025 Predictive vs causal targeting of social interventions Machine learning is increasingly used to inform which people receive limited interventions in a wide range of domains, including healthcare, human services, education, development, and more. What is the right quantity for such models to predict? Moritz Hardt

Haptic Intelligence IS Colloquium Dr. Adam Spiers 26-03-2025 Dexterity and Haptics at the Manipulation and Touch Lab The Manipulation and Touch Lab (MTL) at Imperial College London conducts research on topics related to human and robotic manipulation, tactile sensing, haptic interfaces and applied machine learning. In this talk the PI of the lab, Dr Ad Spiers, (formerly of MPI-IS) will discuss several of the MTL’s novel mechatronic approaches to these topics, focusing particularly on dexterous robotic manipulation and tactile sensors. Ad will also give an overview of his startup’s work on a novel haptic interface to aid navigation for vision-impaired pedestrians. Katherine J. Kuchenbecker

Social Foundations of Computation Talk Jason Hartline 24-03-2025 Optimization of Scoring Rules Scoring rules are everywhere. Any decision problem where an agent has beliefs about an unknown state and takes an action and realizes payoffs according to the action and the realized state is a scoring rule. Behavioral subjects in experiments are evaluated and rewarded according to scoring rules. Machine learning algorithms are trained and evaluated according to scoring rules. Students' coursework is graded according to scoring rules. Moritz Hardt

Talk Nicolas Papernot 12-03-2025 Special Talk: Verifiable Approaches to Trustworthy Machine Learning: Lessons from Researching Unlearning The talk presents open problems in the study of trustworthy machine learning. We begin by broadly characterizing the attack surface of modern machine learning algorithms. We then illustrate the challenges of having end user’s trust that machine learning algorithms were deployed responsibly, i.e., verify its trustworthiness, through a deep dive on the problem of unlearning. The need for machine unlearning, i.e., obtaining a model one would get without training on a subset of data, arises from privacy legislation and as a potential solution to data poisoning or copyright claims. As we present... Moritz Hardt Eva Lämmerhirt Nisha Tyagi

Perceiving Systems Talk Robin Courant 10-03-2025 How and what to film in virtual environments? Content creation for movies and video games has been transformed with the rise of virtual environments, yet filming within these digital worlds remains a complex challenge. This talk explores the question: how and what to film in virtual environments? We examine the role of camera control and human interaction across different virtual settings, including NeRF, 3D engines, and video generation. Victoria Fernandez Abrevaya

Symposium 25-02-2025 2025 Scientific Symposium All current and former employees and partners of the Max Planck Institute for Intelligent Systems are welcome to attend this event. If you have any questions, please contact Eva Lämmerhirt, Institute Management Officer, at eva.laemmerhirt@tuebingen.mpg.de Katherine J. Kuchenbecker Eva Lämmerhirt

Perceiving Systems Talk Ailing Zeng 18-02-2025 The Dawn of Video Generation: Preliminary Explorations with SORA-like Models High-quality video generation—encompassing text-to-video (T2V), image-to-video (I2V), and video-to-video (V2V) generation—plays a pivotal role in content creation and world simulation. While several DiT-based models have advanced rapidly in the past year, a thorough exploration of their capabilities, limitations, and alignment with human preferences remains incomplete. In this talk, I will present recent advancements in SORA-like T2V, I2V, and V2V models and products, bridging the gap between academic research and industry applications. Through live demonstrations and comparative analyses, ... Nikos Athanasiou Michael Black

Social Foundations of Computation Talk Stratis Tsirtsis 11-02-2025 - 11-03-2025 Counterfactual Token Generation in Large Language Models Imagine the following story, generated by a large language model: "Captain Lyra stood at the helm of her trusty ship, the Maelstrom's Fury, gazing out at the endless sea. [...] Lyra's eyes welled up with tears as she realized the bitter truth—she had sacrificed everything for fleeting riches, and lost the love of her crew, her family, and herself.” Now, let’s conduct a thought experiment: how would the story have unfolded if the model had chosen “Captain Maeve” as the protagonist instead? Moritz Hardt

Perceiving Systems Talk Yannis Siglidis 06-02-2025 Computer Vision at the Mirror Stage: Questioning and Refining Visual Categorization Computer vision advancements in predicting and visualizing labels, often motivate us to consider the relationship between labels and images as a given. Yet, the prototypical nature of coherent labels, such as the alphabet of handwritten characters, can help us question assumed families of handwritten variation. Nikos Athanasiou

Haptic Intelligence PhD Thesis Defense 15-01-2025 Capturing and Recognizing Multimodal Surface Interactions as Embedded High-Dimensional Distributions Exploring a surface with a handheld tool generates complex contact signals that uniquely encode the surface's properties—a needle hidden in a haystack of data. Humans naturally integrate visual, auditory, and haptic sensory data during these interactions to accurately assess and recognize surfaces. However, enabling artificial systems to perceive and recognize surfaces with human-like proficiency remains a significant challenge. The complexity and dimensionality of multi-modal sensor data, particularly in the intricate and dynamic modality of touch, hinders effective sensing and processing.... Behnam Khojasteh Katherine J. Kuchenbecker

Haptic Intelligence PhD Thesis Defense 17-12-2024 Precision Haptics in Gait Retraining for Knee Osteoarthritis Gait retraining, or teaching patients to walk in ways that reduce joint loading, shows promise as a conservative intervention for knee osteoarthritis. However, its use in clinical settings remains limited by challenges in prescribing optimal gait patterns and delivering precise, real-time biofeedback. This thesis presents four interconnected studies that aim to address these barriers to clinical adoption: First, a regression model was developed to predict patient-specific biomechanical responses to a gait modification using only simple clinical measures, reducing the need for instrumented g... Katherine J. Kuchenbecker Nataliya Rokhmanova

Talk Dr. Isabella Fiorello 10-12-2024 Next-Generation Biohybrids: Engineering Miniature Machines Inspired by Plant Systems Among living organisms, plants are an ideal source of inspiration for robotics and engineering due to their remarkable evolutionary adaptations to almost every habitat. When miniaturized, plant-inspired machines can navigate confined and complex unstructured surfaces. We introduce a new class of plant-inspired, microfabricated hybrid machines designed for multifunctional tasks such as in situ monitoring and targeted cargo delivery. These machines combine bioinspired design with biohybrid approaches, incorporating the morphological and biomechanical features of both terrestrial and aquatic ... Katherine J. Kuchenbecker Christoph Keplinger

Perceiving Systems Talk Sergi Pujades 28-11-2024 How to predict the inside from the outside? Segment, register, model and infer! Observing and modeling the human body has attracted scientific efforts since the very early times in history. In the recent decades, though, several imaging modalities, such as Computed Tomography scanners (CT), Magnetic Resonance Imaging (MRI), or X-ray have provided the means to “see” inside the body. Most interestingly, there is growing evidence pointing that the shape of the surface of the human body is highly correlated with its internal properties, for example, the body composition, the size of the bones, and the amount of muscle and adipose tissue (fat). In this talk I will go over ... Marilyn Keller

Social Foundations of Computation Talk Dr. Krishna P. Gummadi 19-11-2024 Towards Better Foundations for Foundational Models: A Cognitivist Approach to Studying Large Language Models (LLMs) The talk will begin with a short demo of an LLM-based assistant that allows scientists to convert their papers (with a simple drag and drop) into short podcasts for communicating their research to a general audience. While we built the tool, we can’t explain its unreasonable (in)effectiveness, i.e., we don’t really understand why it works or when it might fail. So in the rest of the talk, I will present our investigations into some curiosity-driven questions about LLMs; specifically, how do LLMs receive, process, organize, store, and retrieve information. Moritz Hardt

Haptic Intelligence PhD Thesis Defense 23-10-2024 Data-Driven Needle Puncture Detection for Urgent Medical Care Delivery in Space Needle decompression (ND) is a surgical procedure that treats one of the most preventable causes of trauma-related death: dangerous accumulations of air between the chest wall and the lungs. However, needle-tip overshoot of the target space can result in the inadvertent puncture of critical structures like the heart. This type of complication is fatal without urgent surgical care, which is not available in resource-poor environments like space. Since ND is done blind, operators rely on tool sensations to identify when the needle has reached its target. Needle instrumentation could enable pu... Katherine J. Kuchenbecker Rachael Lorsa

Event 17-10-2024 The Atomic Human: Understanding ourselves in the age of AI The Max Planck Institute for Intelligent Systems is delighted to invite you to its 2024 Max Planck Lecture in Stuttgart. Michael J. Black Valeria Rojas

Max Planck Lecture Neil Lawrence 17-10-2024 The Atomic Human: Understanding ourselves in the age of AI The Max Planck Institute for Intelligent Systems is delighted to invite you to its 2024 Max Planck Lecture in Stuttgart. Michael Black Barbara Kettemann

Outreach 15-10-2024 Max Planck Institute for Intelligent Systems exhibits research in Stuttgart City Hall As part of the Stuttgart Festival of Science, robots are swimming and walking through the corridors. Barbara Kettemann

Perceiving Systems Talk Guy Tevet 14-10-2024 Diffusion Models for Human Motion Synthesis Character motion synthesis stands as a central challenge in computer animation and graphics. The successful adaptation of diffusion models to the field boosted synthesis quality and provided intuitive controls such as text and music. One of the earliest and most popular methods to do so is Motion Diffusion Model (MDM) [ICLR 2023]. In this talk, I will review how MDM incorporates domain know-how into the diffusion model and enables intuitive editing capabilities. Then, I will present two recent works, each suggesting a refreshing take on motion diffusion and extending its abilities to new... Omid Taheri

Research

Departments

Research Groups

People

Contact

Our Institute

Our History

Career

Doctoral Programs

Training

Service Units

Central Scientific Facilities

Workshops

Campus Services

Impact

Cooperation

Partners and Initiatives