88 days 0:52 A brain-inspired agentic architecture to improve planning with LLMs
Register for the series to learn about future episodes: aka.ms/registerresearchforumYTs2e2 Watch all talks in this episode: aka.ms/researchforumYTs2e2 Explore all previous episodes
1 820
30
89 days 6:40 Data Formulator: Vibe with your data, in control
Data Formulator is an AI-powered tool for analysts to iteratively explore and visualize data.
732
8.1
89 days 6:57 Accelerating MRI image reconstruction with Tyger
The Tyger framework enables faster, more accessible medical imaging by streaming raw data to the cloud for accelerated reconstruction—reducing patient wait times and discomfort—while empowering
391
6.9
89 days 6:31 Tool-space Interference: An emerging problem for LLM agents
Tool-space interference occurs when adding an otherwise reasonable agent or tool to a team or agent reduces end-to-end task performance.
493
16.1
89 days 5:46 Publicly-verifiable elections
Microsoft’s free, open-source ElectionGuard tools enable voters to verify their votes were accurately counted without compromising privacy or trusting election equipment or personnel.
696
9.1
89 days 6:04 A brain-inspired agentic architecture to improve planning with LLMs
Inspired by human collective cognition and neuroscience, we conducted two studies showing that a) multi-LLM architectures with mixed communication connectivity lead to better collaborative
870
5
89 days 1:42 Watch now: The last episode of Microsoft Research Forum of 2025
Register for the series to learn about future episodes: aka.ms/registerresearchforumYTs2e2 Watch all talks in this episode: aka.ms/researchforumYTs2e2 Explore all previous episodes
1 830
37.8
89 days 34:02 Microsoft Research Forum | Season 2, Episode 2
Microsoft Research Forum is a virtual series highlighting purposeful research and its real-world impact, from fundamental exploration to advancing AI responsibly, scaling innovation through products
1 160
11
96 days 45:03 Leveraging Loanword Constraints for Improving Machine Translation in Low-resource Settings
Translating from high-resource to low-resource languages like Emakhuwa remains a challenge due to limited parallel data, orthographic variation, and frequent loanwords and code-switching.
225
97 days 57:38 Ideas: Community building, machine learning, and the future of AI
As the Women in Machine Learning Workshop (WiML) marks its 20th annual gathering, cofounders, friends, and collaborators Jenn Wortman Vaughan and Hanna Wallach reflect on WiML’s evolution
909
18.6
97 days 1:18:27 Coauthor roundtable: Reflecting on healthcare economics, biomedical research, and medical education
For the series finale, Peter Lee, Carey Goldberg, and Dr.
313
8.4
103 days 1:56 Promptions
Promptions (“Prompt” plus “options”) dynamically generates UI to help users steer AI responses more effectively.
974
13.7
103 days 0:44 Coming December 9th! Microsoft Research Forum | Season 2, Episode 2
aka.ms/researchforum
724
59.8
104 days 1:04:38 Spatial Audio Rendering for Speech Live Translation
Language barriers in virtual meetings remain a persistent challenge to global collaboration.
310
109 days 1:01:01 Designing safe digital systems for the humanitarian sector
Speaker: Carmela Troncoso Host: Betül Durak In this talk we overview our collaboration with the International Committee of the Red Cross, in which we help them to digitalize their aid
332
21.8
112 days 1:01:57 The Illusion of Inclusion How LLMs Misrepresent African Languages and Cultural Contexts
Multilingual LLMs have become so powerful, yet African languages remain misrepresented.
532
11.5
122 days 5:04 Data Formulator v0.5: Vibe with data, in control
Data Formulator is an AI-powered visualization tool that helps analysts iteratively explore and visualize data from any source—clean or messy, small or large—without coding.
1 621
15.1
123 days 57:55 Knowledge-Coin Fair Exchange
Fair exchange has been studied in computer science for many decades.
690
28.4
142 days 1:28:41 Distant conversational speech recognition: Challenges and Opportunities
Host: Sunit Sivasankaran, Microsoft Research Speaker: Dr.
382
20.9
142 days 1:56:28 Ultra Ethernet for next-generation AI and HPC workloads
Host: Sujata Banerjee Speakers: Torsten Hoefler, Microsoft and ETH Zurich; Abdul Kabbani, Microsoft The Ultra Ethernet Consortium set out to redefine Ethernet-based interconnects for AI and
2 351
12.7
142 days 53:58 BRAIN SIGNALS TO ACTION: Monitoring and Explaining User Cognitive Load with Foundation Models
Host: Dimitra Emmanouilidou, Microsoft Research Speaker: Deeksha Moodasarige Shama Passive monitoring of cognitive load can enable personalized user experiences and even accelerate human learning
707
9
142 days 34:11 IronDict: Transparent Dictionaries from Polynomial Commitments
Host: Melissa Chase, Microsoft Research Speaker: Hossein Hafezi, NYU We present IronDict, a transparent dictionary construction based on polynomial commitment schemes.
178
142 days 54:46 Lattice-Based Accumulator and Application to Anonymous Credential Revocation
Host: Betül Durak, Microsoft Research Speaker: Victor Youdom Kemmoe An accumulator is a cryptographic system for compactly representing a set of elements such that every element in the set has a
230
142 days 54:08 FOA Tokenizer: Learning Discrete Representations of Spatial Audio with Multichannel VQ-GAN
Host: Hannes Gamper, Microsoft Research Speaker: Parthasaarathy Sudarsanam, Tampere University Spatial audio captures the directional and environmental characteristics of sound, enabling immersive
252
142 days 56:10 Efficient Secure Aggregation for Federated Learning
Host: Melissa Chase, Microsoft Research Speaker: Varun Madathil, Yale University Federated Learning (FL) trains a global model by having each selected device push only its model update to a
377
9.3
144 days 1:16 What does it mean to paraphrase a protein?
This explainer video unpacks the science behind paraphrasing a protein—a defining technique of Microsoft’s Paraphrase Project.
737
6.1
152 days 1:07:22 Evaluating the Cultural Relevance of AI Models and Products: Insights from the YUX Team
How do we ensure that AI systems are not only technologically advanced but also deeply resonant and beneficial for the specific communities they aim to serve?
230
152 days 28:52 Evaluating Privacy Policies under Modern Privacy Laws At Scale: An LLM-Based Automated Approach
Website privacy policies detail an online service's information practices, including how they handle user data and rights.
439
7.8
153 days 52:07 Ideas: More AI-resilient biosecurity with the Paraphrase Project
Microsoft’s Eric Horvitz and guests Tessa Alexanian, James Diggans, and Bruce Wittmann discuss the Paraphrase Project—a red-teaming effort that exposed and secured a biosecurity vulnerability in
502
8.9