جست‌وجوی استاد و پژوهشگر

Arman Cohanمشاهده پروفایل

استادیار

• Natural Language Processing
• Machine Learning
• Artificial Intelligence
• +۳ مورد دیگر

Arman Cohan is an Assistant Professor in the Department of Computer Science at Yale University, where he leads the Yale NLP Lab since its founding in January 2023. His research spans natural language processing and machine learning with emphasis on language modeling, representation learning, retrieval systems, and specialized domain applications including scientific discovery and AI for science. His primary research interests include: Natural Language Processing Machine Learning Large Language Models Information Retrieval AI for Science Scientific Problem-Solving Recent publications (2025) demonstrate intense focus on evaluating and advancing LLM capabilities across multimodal reasoning, scientific claim verification, financial domain applications, and biological modeling. The lab consistently produces high-impact work accepted at top-tier conferences including ACL, EMNLP, and ICLR, with 11 papers at ACL 2025 alone. Scientific awards include: Best Paper Award at AI4Research Workshop (IJCAI 2024) Outstanding Paper Award at EACL 2023 Best Paper Award at ACL 2024 for Olmo language model research Professor Cohan actively advises PhD students including Kaili Liu, Jacob Dunefsky, Alan Li, Yilun Zhao, and has graduated researchers such as Linyong Nan (now at Zoom) and Ansong Ni (now at Meta). The lab maintains strong industry partnerships and receives substantial research funding as evidenced by its prolific output and conference presence. The Yale NLP Lab hosts the annual New England NLP Workshop and regularly features speakers from leading institutions including Meta AI, Allen Institute for AI, and DeepMind, fostering a collaborative environment for advancing NLP research.

Robin Jiaمشاهده پروفایل

استادیار

• Natural Language Processing (NLP)
• Deep Learning
• Artificial Intelligence
• +۷ مورد دیگر

Robin Jia is an Assistant Professor in the Thomas Lord Department of Computer Science at the University of Southern California (USC) , where he leads the AI, Language, Learning, Generalization, and Robustness (Allegro) Lab . His research focuses on enhancing the reliability and robustness of large language models (LLMs) through mechanistic understanding, benchmarking under distribution shifts, and neurosymbolic integration. Key affiliations include collaborations with the USC Keck School of Medicine and contributions to legal frameworks like the EU's Digital Services Act. Robin's research spans multiple domains, including: Scientific analysis of LLM capabilities in in-context learning , data memorization , and numerical reasoning Advancements in robust NLP systems , emphasizing uncertainty estimation and calibration Development of methods combining LLMs with symbolic solvers for complex reasoning tasks Interdisciplinary applications in medicine and law , such as privacy-preserving synthetic data generation and medical misconception evaluation . His recent publications (2024-2025) address Fourier-based numerical embeddings (NeurIPS), neurosymbolic planning (NAACL), and multimodal benchmarking (COLM), with a strong emphasis on privacy , fairness , and transparency . Scientific awards include the Google Research Scholar Award (2023) , SoCalNLP Symposium Best Paper Awards , and ACL/EMNLP outstanding papers . He advises PhD students like Johnny Wei and Ameya Godbole, and has secured grants from the NSF , USC-Capital One , and USC-Amazon .

Roberto Navigliمشاهده پروفایل

استاد

• Natural Language Processing
• Multilingual Natural Language Understanding
• Large Language Models
• +۵ مورد دیگر

Roberto Navigli is a Full Professor in the Department of Computer, Control and Management Engineering at Sapienza University of Rome, where he leads the Sapienza NLP Group. He has held academic positions at Sapienza since 2007, progressing from Researcher to Associate Professor, and now Full Professor. Navigli is internationally recognized for his groundbreaking work in multilingual natural language processing and computational semantics. His research spans multiple areas within Natural Language Processing, with a particular focus on multilinguality , lexical semantics (Word Sense Disambiguation and induction), sentence-level semantics (Semantic Role Labeling and Semantic Parsing), and Large Language Models . As the leader of the Minerva LLM project, he pioneered the development of the first pretrained LLM in Italian. His creation of BabelNet, the largest multilingual encyclopedic computational dictionary covering 271 languages and about 14 million concepts, represents a landmark achievement in computational linguistics. Professor Navigli's recent publications demonstrate a strong focus on advancing multilingual capabilities of language models, addressing challenges in semantic understanding across languages, and developing robust evaluation frameworks. His work on Italian LLM optimization, lexical ambiguity in machine translation, and cross-lingual safety gaps reflects his commitment to building more accurate and reliable language technologies that work effectively across diverse linguistic contexts. AAAI Fellow (2025) EurAI Fellow (2024) ELLIS Fellow (2024) ACL Fellow (2023) Outstanding Paper Awards at ACL 2024, ACL 2023, and NAACL 2021 Best Resource Paper Award at ACL 2022 Prominent Paper Awards from the Artificial Intelligence Journal (2023, 2017) META Prize 2015 for BabelNet Professor Navigli has successfully secured substantial research funding, including two prestigious ERC Grants (MultiJEDI ERC Starting Grant 2010-2016 and MOUSSE ERC Consolidator Grant 2017-2023, totaling around 3 million euros), PI of PRIN 2017 and PRIN 2020 projects (overall 1.4 million euros), and a Google Focused Research Award (around 300k$). He has mentored over 30 PhD students, many of whom have gone on to successful careers in academia and industry. His leadership extends to major international conferences, serving as General Chair of ACL 2025 and Program Co-Chair of ACL-IJCNLP 2021. As the director of the Sapienza NLP Group and co-leader of the Transversal Project on Large Language/Multimodal Models in the FAIR (Future Artificial Intelligence Research) project, Professor Navigli oversees a vibrant research ecosystem focused on advancing the state-of-the-art in natural language understanding across languages. His spinoff company, Babelscape, translates academic research into practical applications for industry.

Sebastian Riedelمشاهده پروفایل

استاد

• Natural Language Processing (NLP)
• Machine Learning
• Artificial Intelligence
• +۳ مورد دیگر

Sebastian Riedel is a Professor at University College London (UCL) and a Researcher at DeepMind, leading the UCL NLP Lab. His work focuses on teaching machines to read, reason, and write, integrating Natural Language Processing (NLP) with Machine Learning. He holds an Allen Distinguished Investigator award and has held roles at FAIR, UMass Amherst, Tokyo University, and the University of Edinburgh. Education: PhD in Computer Science from the University of Edinburgh (advisor: Ewan Klein), postdoctoral research at UMass Amherst (advisor: Andrew McCallum), and research at Tokyo University (advisor: Tsujii Junichi). Research Interests: NLP, machine learning, information extraction, and multimodal models like Gemini. He develops tools such as UCLEED (BioNLP event extractor), frontlets (Scala map wrappers), and thebibbrag (BibTeX to HTML converter). Awards: Allen Distinguished Investigator. Software contributions include GitHub repositories for NLP, machine learning, and data tools. Contact: s.riedel@ucl.ac.uk | Office: 1st Floor, 90 High Holborn, London WC1V 6LJ | Office Hours: Mondays 11 AM–12 PM.

Raquel Fernándezمشاهده پروفایل

استاد

• Computational Linguistics
• Dialogue Systems
• Cognitive Science
• +۳ مورد دیگر

Raquel Fernández is Full Professor of Computational Linguistics and Dialogue Systems at the University of Amsterdam, where she leads the Dialogue Modelling Group at the Institute for Logic, Language & Computation (ILLC). As Vice-Director for Research at ILLC and a Fellow of the ELLIS Society, she bridges computational linguistics, cognitive science, and artificial intelligence through her research on language use in multimodal and conversational contexts. PhD in Computational Linguistics from King's College London Prior research positions at University of Potsdam and Stanford University's CSLI Her work explores how cognitive constraints, social interaction, and perception shape language use, with a focus on: Visually-grounded language processing Multimodal dialogue modeling Model uncertainty and calibration Language grounding in multimodal data Language learning and semantic change Dialogue reference resolution Recent publications analyze multimodal reasoning limitations, cross-lingual knowledge consistency, and uncertainty modeling in dialogue systems. She has received multiple accolades including an ERC Consolidator Grant , NWO VENI/VIDI/Aspasia fellowships , and EMNLP/GenBench awards . Outstanding Paper Award (EMNLP 2023) Best Data Award (GenBench Workshop 2023) ELLIS Society Fellow ERC Consolidator Grant #819455 recipient NWO VENI/VIDI/Aspasia awardee As a leader in academic service, she serves on the SIGDAT Executive Committee and chairs multiple conference committees. Her lab develops models for multimodal dialogue, visual storytelling, and grounded language understanding.

Mark Steedmanمشاهده پروفایل

استاد

• Computational Linguistics
• Natural Language Processing
• Combinatory Categorial Grammar
• +۲ مورد دیگر

Mark Steedman is Professor of Cognitive Science at the University of Edinburgh's School of Informatics, with adjunct appointment at University of Pennsylvania. His research spans computational linguistics, AI, and cognitive science, focusing on Combinatory Categorial Grammar (CCG) and its applications. His research examines: Combinatory Categorial Grammar parsing and semantics Language model capabilities and limitations Cross-linguistic semantic inference Brain modeling of language processing Recent publications analyze hallucination sources in large language models, cross-linguistic entailment graphs, and brain-computer parallels in structure-building. He develops computational models integrating symbolic and distributional approaches to semantics. Honors include ACL Lifetime Achievement Award (2018) and George E. Davis Medal (2001). He serves on editorial boards of major linguistics journals and has authored influential books including 'The Syntactic Process' and 'Taking Scope'.

Xuezhe Maمشاهده پروفایل

استادیار

• Natural Language Processing
• Machine Learning
• Deep Learning
• +۵ مورد دیگر

Xuezhe Ma is an Assistant Professor in the Department of Computer Science at the University of Southern California's Viterbi School of Engineering. Previously, he was a Ph.D. student at Carnegie Mellon University's Language Technologies Institute, where he worked under the supervision of Professor Eduard Hovy. His academic journey includes a Master's degree from Shanghai Jiao Tong University's Center for Brain-like Computing and Machine Intelligence and a Bachelor's degree in Computer Science from the same institution. Ph.D. in Computer Science, Carnegie Mellon University (completed ~2020) M.S. in Brain-like Computing, Shanghai Jiao Tong University B.S. in Computer Science, Shanghai Jiao Tong University Dr. Ma's research spans multiple areas at the intersection of Natural Language Processing and Machine Learning, with particular focus on structured prediction, syntactic and semantic parsing, machine translation, language generation, and deep generative models. His recent work has expanded into vision-language models, large language model architectures, and applications across computer vision tasks. His research combines theoretical foundations with practical implementations, as evidenced by his development of tools like NeuroNLP2 and MaxParser. His publication record shows a clear trajectory from foundational NLP work during his PhD (including papers on dependency parsing and sequence labeling) to more recent contributions in generative models and large language systems. The 15 most recent publications reveal a strong focus on addressing fundamental challenges in generative modeling, context handling, and multimodal integration, with applications spanning literary translation, medical imaging, and news diffusion analysis. AI2 Outstanding Intern Award (2018) Dr. Ma has secured research funding supporting his work in generative models and language technologies, with projects focusing on improving the efficiency and capabilities of large language models. His research group at USC is actively working on next-generation language understanding and generation systems, with particular emphasis on context-aware modeling and multimodal integration. He has established collaborations with industry partners including the Allen Institute for AI and has contributed to open-source projects like Texar. At USC, Dr. Ma leads research in the Information Sciences Institute, directing projects on efficient large language model architectures and multimodal reasoning systems. His lab focuses on developing novel approaches to context handling, model efficiency, and multimodal integration, with applications across diverse domains including healthcare, literary analysis, and news media.

Wenxuan Zhangمشاهده پروفایل

استادیار

• Natural Language Processing (NLP)
• Large Language Models (LLMs)
• Multilingual Language Models
• +۵ مورد دیگر

Dr. Wenxuan Zhang is a tenure-track Assistant Professor at the Information Systems Technology and Design (ISTD) Pillar of Singapore University of Technology and Design (SUTD), supported by the prestigious SUTD Assistant Professorship (SAP) award. He holds a PhD from The Chinese University of Hong Kong and previously worked as a research scientist at Alibaba Group Singapore. His research focuses on advancing large language models (LLMs) to be both inclusive (supporting multilingual capabilities) and trustworthy (ensuring safety and robustness). Key projects include SeaLLMs (specialized for Southeast Asian languages), Babel (serving 90% of global speakers), and M3Exam (LLM evaluation framework). Education: PhD in Computer Science, The Chinese University of Hong Kong Previous roles: Research Scientist at Alibaba Singapore (2022) Research Interests: Multilingual LLMs, AI safety, model evaluation, and cross-lingual adaptation. He leads projects addressing LLM trustworthiness through safety mechanisms and fair evaluation practices. Awards & Recognition: SUTD Assistant Professorship (2025) Alibaba Star (2022) ITU Best Innovate for Impact Award (2024) Service & Leadership: Area Chair for NeurIPS 2025, ACL 2025, and multiple other top conferences. Actively contributes to program committees for conferences like ICLR and EMNLP. Current Projects: Multilingual LLMs, model compression, safety frameworks, and evaluation methodologies. Openings for PhD/Postdoc researchers in these areas.

• Human-Centered Artificial Intelligence
• Natural Language Processing
• Computer Vision
• +۵ مورد دیگر

Carlos Toxtli is an Assistant Professor at Clemson University where he leads the Human-AI Empowerment Lab. His research applies Human-Centered AI to create fair workplace tools, focusing on NLP, computer vision, and crowdsourcing ethics. He holds a PhD in Computer Science from Northeastern University and previously worked at Google, Microsoft Research, and Snap Inc. Education: Ph.D. Computer Science, Northeastern University M.S. Innovation & Technological Entrepreneurship, Monterrey Institute of Technology MBA, IEDE Business School B.S. Computer Science, University of the Valley of Mexico B.S. Computer Engineering, National Autonomous University of Mexico Research Focus: Dr. Toxtli develops AI systems promoting workplace fairness through ethical frameworks. His work spans: 1) Human-AI collaboration for task management, 2) Bias mitigation in gig economies, 3) Multimodal interfaces for accessibility, and 4) Culturally adaptive systems. His lab explores how AI can augment human capabilities while ensuring transparency. Publication Trends: Recent works demonstrate strong focus on LLM reliability (7 papers), human-AI teaming (4 studies), and ethical AI frameworks (3 publications). Over 60% of 2024-2025 output addresses real-world deployment challenges. Awards: UNESCO IRCAI Global Top 100 project Google Apps Developer Challenge 1st Place Facebook Developers World Hack 1st Prize Intel Innovation Latin App 1st Prize NSF STIR Labs Research Grant Leadership: Founded the Human-AI Empowerment Lab securing $1.2M in NSF/industry grants. Supervises 8 PhD students in human-centered computing projects. Previously co-founded 5 tech startups including ComproPago (acquired by Coca-Cola FEMSA).

• Statistical Machine Learning
• Bayesian Inference
• Variational Methods
• +۵ مورد دیگر

Michael C. Hughes ("Mike") is an Assistant Professor in the Department of Computer Science at Tufts University's School of Engineering, where he develops statistical machine learning methods for healthcare applications. His work focuses on building predictive models that extract actionable insights from complex clinical data, including electronic health records and medical imaging. PhD, Computer Science, Brown University (2016) MS, Computer Science, Brown University (2012) BS, Computer Science, Franklin W. Olin College of Engineering (2010) Research interests center on: Bayesian hierarchical models for documents, sequences, and medical images Optimization algorithms for approximate inference Model fairness and interpretability in clinical contexts Semi-supervised learning for medical diagnostics Recent publications demonstrate these capabilities through applications in cardiovascular disease diagnosis, opioid overdose forecasting, and ICU risk prediction. His lab emphasizes reproducibility through open datasets like TMED-2 and open-source tools like BNPy. Grants include NIH R01 funding for heart valve disease detection, NSF CAREER support for model interpretability, and NSF GCR funding for educational uncertainty research. Scientific awards include: NIH R01 Award (PI) for heart valve disease detection (2025) NSF CAREER Award (2024) NSF GCR Grant (2024) Best Poster Award at Time Series Workshop (ICML 2021) Top 10% Reviewer Awards at AISTATS (2023, 2022) Teaching activities include courses on Bayesian Deep Learning, Introduction to Machine Learning, and Statistical Pattern Recognition. He previously served as postdoctoral fellow at Harvard SEAS.

Annette Riosمشاهده پروفایل

پژوهشگر

• Machine Translation
• Text Simplification
• Multimodal Processing
• +۵ مورد دیگر

Annette Rios is a Researcher at the University of Zurich within the Language, Technology and Accessibility group. Her work focuses on machine translation , text simplification , and sign language translation , particularly for multilingual and indigenous language contexts. Research highlights include contributions to the European EASIER project on sign language translation, the ZurichNLP consortium, and the development of tools for emotion detection and gender bias reduction in NLP systems. She has published extensively in top-tier venues like ACL and WMT, with a focus on multimodal translation and German language processing . Developed emotion_detection GitHub repository for German emotion analysis Contributed to ContraPro and ContraWSD projects on translation evaluation Created datasets for German text simplification and emotion detection Her publications reveal a strong emphasis on machine translation (7/15 recent works), sign language technology (3/15), and text simplification (2/15). While no formal advising is documented in the provided text, she has mentored GitHub contributors to her repositories. Current research directions include improving multimodal translation systems and expanding emotion detection capabilities beyond English.

• Artificial Intelligence
• Computer Vision
• Natural Language Processing
• +۴ مورد دیگر

Armin Kirchknopf serves as a Junior Researcher at the Media Computing Research Group within the Institute of Creative Media/Technologies, Department of Media and Digital Technologies at the University of Applied Sciences St. Pölten. His interdisciplinary work bridges artificial intelligence, computer vision, and social media analysis, with significant contributions to misinformation detection and disaster response systems. Based at Campus-Platz 1 in St. Pölten, Austria, he actively collaborates on EU-funded projects and publishes in top-tier AI venues. His educational journey spans humanities and technology: a Bachelor of Arts in Egyptology and Master of Arts in Classical Archaeology from the University of Vienna (including fieldwork at excavation sites across Austria, Germany, and Egypt), followed by a Bachelor of Science in Media Technology from FH St. Pölten. This unique background informs his human-centered AI research approach. Kirchknopf's research centers on explainable multimodal AI systems for real-world challenges. His recent work demonstrates expertise in transformer-based architectures for cross-lingual fake news detection, sexism identification, and flood monitoring through social media imagery. He pioneers techniques like Grad-CAM for object detection explainability and develops visualization tools for complex data interpretation, emphasizing transparency and social impact in AI deployment. Analysis of his 13 publications (2017-2022) reveals a strategic shift toward applied AI in societal contexts , particularly using social media data for disaster management and combating online toxicity. His projects consistently integrate computer vision with natural language processing, showing increasing sophistication in multilingual capabilities and model interpretability frameworks. His scientific recognition includes: Creative Business Award for co-developing the Tenjin learning quiz application No documented student advisement or grant leadership appears in current records, though he actively mentors through project-based collaborations. His work with the Media Computing Research Group drives innovation in educational technology and public safety applications. Kirchknopf contributes to the Media Computing Research Group's portfolio including Fake News Detection, SAiEX (Safe AI with explainable integrity), InfraBase (building footprint segmentation), and Ressel Center music therapy projects. His cross-disciplinary collaborations span computer scientists, archaeologists, and social scientists, reflecting the group's commitment to human-centric technological solutions .

Michihiro Yasunagaمشاهده پروفایل

استادیار

• Large Language Models
• Multimodal Models
• Reasoning Systems
• +۶ مورد دیگر

Michihiro Yasunaga is an Assistant Professor in the Department of Computer Science at Stanford University's School of Engineering. He received his PhD in Computer Science from Stanford, advised by Percy Liang, Jure Leskovec, and Chris Manning. Prior to his faculty position, he worked as a researcher at Google DeepMind and Meta. His research focuses on building LLMs and agents that assist humans in diverse tasks, with particular expertise in post-training techniques (RL, reward models, and evaluation), reasoning systems (AnalogicalReasoner), retrieval and tool use for LLMs (LinkBERT, QAGNN, DRAGON, REPLUG, HippoRAG), and multimodality (RA-CM3, Med-Flamingo, Transfusion). His work spans both theoretical foundations and practical applications of large language models. Yasunaga's publication record demonstrates significant contributions to the field of AI, with 15 recent articles (2023-2025) covering diverse aspects of language model development, evaluation, and application. His research shows a clear trajectory toward building more capable, efficient, and reliable multimodal AI systems, with particular emphasis on knowledge integration and robust evaluation frameworks. Among his notable achievements is the Best Paper Award at AAAI 2023 Deep Learning on Graphs Workshop for the DRAGON paper. He has also been deeply involved in major benchmarking efforts including HELM and HEIM, which provide comprehensive evaluation frameworks for language and vision-language models. Yasunaga actively contributes to the research community through service roles including Organizing Committee for the Workshop on Knowledge-Augmented Methods for NLP (ACL 2024), Workshop on Structured and Unstructured Knowledge Integration (NAACL 2022), and the Workshop on Scientific Document Summarization (SIGIR 2017-2020). He has also served on program committees for top conferences including NeurIPS, ICML, ICLR, ACL, EMNLP, CVPR, and ICCV from 2020-2025.

Shrimai Prabhumoyeمشاهده پروفایل

استادیار مدعو

• Large Language Models (LLMs)
• Machine Learning
• Natural Language Processing (NLP)
• +۴ مورد دیگر

Shrimai Prabhumoye is a Senior Research Scientist at NVIDIA's Applied Deep Learning Research group and an Adjunct Assistant Professor at Boston University. Her research focuses on advancing large language models (LLMs), particularly in enhancing their reasoning capabilities and ensuring safety through toxicity and bias reduction. She is a core contributor to the Nemotron family of LLMs, including the state-of-the-art Nemotron4-15B. Previously, she earned her PhD from Carnegie Mellon University (2021), advised by Prof. Ruslan Salakhutdinov and Prof. Alan W. Black, with a thesis on controllable text generation and computational ethics in NLP. Education: PhD in Machine Learning (CMU, 2021), MS in Language Technologies (CMU, 2017), Undergraduate at National Institute of Technology Karnataka, India. Research Interests: LLMs, controllable generation, ethical AI, bias mitigation, data curation, and model safety. Her work emphasizes rigorous evaluation frameworks and practical applications of NLP systems. Publications span key venues like EACL, ACL, NAACL, and EMNLP, with notable contributions to bias testing, toxicity control, and model scalability. She has been recognized as a Rising Star finalist in the VentureBeat Women in AI Awards 2022 and has led teams in competitions like the Alexa Prize. Labs/Teams: Part of NVIDIA's Applied Deep Learning Research Group and contributed to CMU's Computational Ethics for NLP initiatives. Active in mentoring students on projects like politeness transfer and dataset creation.

• Natural Language Processing
• Large Language Models
• Soft Robotics
• +۴ مورد دیگر

Peiyi Wang is an Assistant Professor at Peking University's School of Electronics Engineering and Computer Science, Institute for Artificial Intelligence. With strong research output spanning both natural language processing and robotics, Wang maintains significant collaborations with Southern University of Science and Technology and National University of Singapore, particularly in soft robotics research with Professor Cecilia Laschi. Additionally, Wang is actively involved with DeepSeek-AI, contributing to several major language model initiatives including DeepSeek-R1 and DeepSeek-V2. Peking University, School of EECS, Institute for Artificial Intelligence (Primary) Southern University of Science and Technology (Collaborative) National University of Singapore (Collaborative) DeepSeek-AI Research Organization Dr. Wang's research spans two primary domains with significant intersection points. In natural language processing, Wang focuses on large language model reasoning capabilities, mathematical verification, uncertainty estimation, and preference alignment. The robotics work centers on soft robotics, particularly origami-inspired designs, strain-based modeling, and control systems for continuum manipulators. These domains converge in Wang's work on vision-language models, embodied AI, and multimodal reasoning systems. Recent work demonstrates particular innovation in mathematical reasoning verification (Math-Shepherd), soft robotic control systems, and red teaming frameworks for language model safety. Wang's publication record shows remarkable productivity, with over 40 publications between 2021-2025 across top-tier venues including ACL, EMNLP, CVPR, and IEEE Transactions on Robotics. The work demonstrates consistent progression from foundational NLP tasks to increasingly sophisticated multimodal and reasoning systems. The most recent publications (2024-2025) show particular emphasis on mathematical reasoning verification, soft robotics control, and language model safety evaluation. While specific awards aren't documented in the provided materials, Wang's work has clearly gained significant recognition through acceptance at top-tier conferences and collaborations with leading researchers in both NLP and robotics fields. Wang's research demonstrates strong interdisciplinary connections, bridging theoretical NLP work with practical robotics applications. The work with DeepSeek-AI suggests active industry collaboration while maintaining strong academic research output. Current research directions appear focused on improving language model reasoning reliability while developing novel soft robotic systems that can interact safely and effectively with complex environments.